nanomedicines and combination therapy of doxorubicin and ... · nanomedicines and combination...
TRANSCRIPT
Nanomedicines and Combination Therapy of
Doxorubicin and Olaparib for Treatment of Ovarian Cancer
by
Sina Eetezadi
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Department of Pharmaceutical Sciences University of Toronto
© Copyright by Sina Eetezadi 2016
ii
Nanomedicines and Combination Therapy
of Doxorubicin and Olaparib for Treatment of Ovarian Cancer
Sina Eetezadi
Doctor of Philosophy
Department of Pharmaceutical Sciences
University of Toronto
2016
Abstract
Ovarian cancer is the fourth leading cause of death in women of developed countries, with
dismal survival improvements achieved in the past three decades. Specifically, current
chemotherapy strategies for second-line treatment of relapsed ovarian cancer are unable to
effectively treat recurrent disease. This thesis aims to improve the therapeutic outcome
associated with recurrent ovarian cancer by (1) creating a 3D cell screening method as an in
vitro model of the disease (2) developing a nanomedicine of doxorubicin (DOX) that is more
efficacious than PEGylated liposomal doxorubicin (PLD / Doxil®) and (3) evaluating additional
strategies to enhance treatment efficacy such as mild hyperthermia (MHT) and combination
therapy with inhibitors of the poly(ADP-ribose) polymerase enzyme family (PARP). Overall,
this work demonstrates the use of 3D multicellular tumor spheroids (MCTS) as an in vitro
drug testing platform which more closely reflects the clinical presentation of recurrent ovarian
cancer relative to traditional monolayer cultures. With the use of this technology, it was found
that tissue penetration of drug is not only an issue for large tumors, but also for invisible,
microscopic lesions that result from metastasis or remain following cytoreductive surgery. A
novel block-copolymer micelle formulation for DOX was developed and fulfilled the goal of
iii
controlling drug release while enhancing intratumoral distribution and MCTS bioavailability of
DOX, which resulted in a significant improvement in growth inhibition, relative to PLD. MHT
appeared to enhance drug accumulation in MCTS in the short term, but not after 48 h of drug
treatment. Drug combination studies of DOX together with the PARP inhibitor, olaparib (OLP,
Lynparza®) were conducted in 2D monolayers and 3D MCTS. In these studies, the
effectiveness of the DOX:OLP combination therapy in monolayers and MCTS was found to be
ratio dependent such that equimolar ratios resulted in an additive effect, while a greater level
of synergy was observed with more extreme ratios. The synergistic effect observed bears
promise for future evaluation in vivo which warrants an appropriate delivery method to ensure
that the determined molar ratios of both drugs accumulate at the tumor as such, despite
differences in the pharmacokinetic profile of each drug, respectively.
iv
Dedication
Dr. Ebrahim Eetezadi
Mortui Viventes Obligant.
v
Acknowledgements
Good science is always the result of good teamwork and this thesis is no exception.
First and foremost I would like to thank Prof. Christine Allen for her continuous support
over the past 5 years, especially during the rather unconventional endeavors I embarked on.
Christine has been a role model with an unparalleled work ethic and has taught me important
lessons in perseverance and leadership. I am especially grateful for the responsibilities she
entrusted me with, which helped me to grow and develop my own leadership skills.
Second, I would like to express my special gratitude to Dr. Payam Zahedi and Dr. Raquel
De Souza. Payam gave me the most valuable advice at the beginning and during this PhD
project. Raquel was an incredible support and enabled to the most productive time during the
last 6 months of this doctorate.
Furthermore, I would like to thank the Drs. Andrew Mikhail and Changhai Lu who had the
patience to train me in the lab and made substantial contributions to the third and fourth
chapter of this thesis. I would also like to thank Sandra Ekdawi for her relentless strive for
perfection, which made the second chapter what it had become. I would have never come
even close to this result without her.
Mohammad Ali Amini, Mike Dunne, Sohyoung Her, Karen Lam and Rida Mourtada have
been the greatest friends inside and outside the lab. Thank you for all the scientific and more
importantly non-scientific talks. I would also like to thank my students Rodrigo Lessa Cataldi,
Mirugashini Vythilingam and Larissa Da Gama, who all did outstanding experimental work.
vi
I would like to acknowledge Dr. Andreas Rummelt, Prof. Cornelia Halin Winter, Prof. Jean-
Christophe Leroux, Prof. Beat Ernst, Prof. Rob Macgregor, Prof. Heiko Heerklotz, Prof. David
Hedley and Prof. Raymond Reilly for their direct or indirect support at different stages of this
dissertation.
This work would not have been possible without my parents, who fostered my curiosity
during my early years, taught me to never accept the world as it is and had to be extremely
patient, while their son was for years somewhere on the other side of the Atlantic studying
something of which neither duration nor outcome was ever really apparent.
Last but not least I would like to thank my wife Shokoufeh for being there for me from start
to finish with all the up and downs along the way. She has been the light and warmth when
science puts you in a cold, dark tunnel where time and speed is lost.
Nihil tam difficile est, quin quaerendo investigari possit.
vii
Table of Contents
Abstract .................................................................................................................................... 2
Dedication ................................................................................................................................ 4
Acknowledgements ................................................................................................................ 5
Table of Contents ................................................................................................................... 7
List of Tables ......................................................................................................................... 12
List of Figures ........................................................................................................................ 13
List of Abbreviations............................................................................................................. 19
........................................................................................................................... 1
1.1. Introduction ............................................................................................................... 2
1.1.1. Ovarian Cancer Treatment in Clinical Practice ................................................ 2
1.1.2. The Role of PEGylated Liposomal Doxorubicin in Ovarian Cancer .............. 4
1.1.3. Targeting DNA Repair in Ovarian Cancer ......................................................... 6
1.2. Rationale .................................................................................................................. 10
1.3. Hypothesis .............................................................................................................. 13
1.4. Overview of Thesis Chapters ............................................................................... 14
1.5. References .............................................................................................................. 16
......................................................................................................................... 22
2.1. Abstract ................................................................................................................... 23
2.2. Introduction ............................................................................................................. 24
2.3. Biology vs. Block Copolymer Micelles ............................................................... 28
2.3.1. First line of defense: the blood compartment ............................................... 28
2.3.2. Second line of defense: tumor extravasation and accumulation .............. 34
2.3.3. Third line of defense, and the Achilles’ heel of nanomedicines: tumor penetration and tumor drug bioavailability .................................................................... 43
2.4. Translatability: best practices and lessons learned ........................................ 52
2.4.1. Towards clinically relevant nanoformulations............................................... 53
viii
2.4.2. The management of heterogeneity at the pre-clinical and clinical levels 57
2.5. Conclusions ............................................................................................................ 67
2.6. Acknowledgements ............................................................................................... 68
2.7. References .............................................................................................................. 69
......................................................................................................................... 80
3.1. Abstract ................................................................................................................... 81
3.2. Introduction ............................................................................................................. 82
3.3. Materials and Methods ......................................................................................... 85
3.3.1. Materials ............................................................................................................... 85
3.3.2. Synthesis of CH3O-PEG-b-PCL (PEG-b-PCL) Copolymers .......................... 85
3.3.3. Preparation and Characterization of BCM+DTX ........................................... 85
3.3.4. Sizing of BCM+DTX ............................................................................................ 86
3.3.5. Transmission Electron Microscopy (TEM) ..................................................... 86
3.3.6. Drug Release ........................................................................................................ 87
3.3.7. Tissue Culture and Growth of MCTS............................................................... 87
3.3.8. Immunohistochemical Analysis of MCTS ...................................................... 87
3.3.9. Measurement of MCTS Growth ....................................................................... 89
3.3.10. Cytotoxicity in Monolayer and Spheroid Tissue Cultures ............................ 89
3.3.11. Growth Inhibition of MCTS ................................................................................ 90
3.3.12. Clonogenic Survival Assay ................................................................................ 90
3.4. Results ..................................................................................................................... 92
3.4.1. Characterization of BCM+DTX ......................................................................... 92
3.4.2. Growth of MCTS .................................................................................................. 93
3.4.3. Cytotoxicity in Monolayer and MCTS Culture ................................................ 93
3.4.4. Inhibition of MCTS Growth ................................................................................ 93
3.4.5. Immunohistochemistry...................................................................................... 95
3.4.6. Clonogenic Survival ............................................................................................. 95
3.5. Discussion ............................................................................................................... 96
3.6. Supporting Information ..................................................................................... 105
3.7. References ........................................................................................................... 108
ix
....................................................................................................................... 112
4.1. Abstract ................................................................................................................ 113
4.2. Introduction .......................................................................................................... 114
4.3. Materials and Methods ...................................................................................... 117
4.3.1. Materials ............................................................................................................. 117
4.3.2. Preparation of BCM-DOX ................................................................................. 117
4.3.3. Physico-Chemical Characterization of BCM-DOX ...................................... 118
4.3.4. Evaluation of Drug Release ............................................................................. 119
4.3.5. Cell Lines ............................................................................................................. 119
4.3.6. Monolayer Cytotoxicity .................................................................................... 119
4.3.7. MCTS Growth Studies ...................................................................................... 120
4.3.8. MCTS Cell Packing Density ............................................................................. 121
4.3.9. Drug Penetration into MCTS ........................................................................... 121
4.3.10. MCTS Growth Inhibition ................................................................................... 124
4.3.11. MHT Experiments ............................................................................................. 124
4.3.12. Statistical Analysis ............................................................................................ 125
4.4. Results .................................................................................................................. 126
4.4.1. Physicochemical Characterization of BCM-DOX ........................................ 126
4.4.2. DOX Release from BCM-DOX ......................................................................... 128
4.4.3. Monolayer Cytotoxicity Studies ...................................................................... 130
4.4.4. MCTS 3D Cell Model ......................................................................................... 132
4.4.5. MCTS Cell Organization and Density ............................................................ 134
4.4.6. MCTS DOX Penetration Studies ..................................................................... 136
4.4.7. MCTS Drug Distribution ................................................................................... 142
4.4.8. MCTS Growth Inhibition Studies .................................................................... 142
4.4.9. MHT Effects on Monolayer Cytotoxicity ....................................................... 144
4.5. Discussion ............................................................................................................ 145
4.6. Conclusion ........................................................................................................... 149
4.7. Acknowledgements ............................................................................................ 150
4.8. References ........................................................................................................... 151
....................................................................................................................... 156
x
5.1. Abstract ................................................................................................................ 157
5.2. Introduction .......................................................................................................... 158
5.3. Materials and Methods ...................................................................................... 162
5.3.1. Materials ............................................................................................................. 162
5.3.2. Cell Culture ......................................................................................................... 162
5.3.3. Cell Line Authentication ................................................................................... 163
5.3.4. BRCA1 and BRCA2 Mutation Sequencing .................................................... 164
5.3.5. Doubling Time .................................................................................................... 165
5.3.6. Monolayer IC50 Evaluation ............................................................................... 165
5.3.7. Determination of Combination Index (CI) Values ....................................... 166
5.3.8. MCTS Growth Studies ...................................................................................... 167
5.3.9. MCTS Growth Inhibition Studies .................................................................... 167
5.3.10. Statistical Analysis ............................................................................................ 168
5.4. Results .................................................................................................................. 169
5.4.1. Properties of the HGSOC Cell Panel .............................................................. 169
5.4.2. Cell Monolayer Viability Studies of DOX and OLP treatment .................... 171
5.4.3. DOX-OLP Combination Studies ...................................................................... 173
5.4.4. MCTS Model Development and Morphology Studies ................................ 176
5.4.5. MCTS Growth and DOX Sensitivity Studies ................................................. 180
5.4.6. MCTS DOX and OLP Combination Studies .................................................. 182
5.5. Discussion ............................................................................................................ 186
5.6. Conclusions ......................................................................................................... 190
5.7. Acknowledgements ............................................................................................ 191
5.8. Supplementary Information .............................................................................. 192
5.9. References ........................................................................................................... 197
....................................................................................................................... 202
6.1. Thesis Conclusion and Summary of Findings .............................................. 203
6.2. Future Directions ................................................................................................. 207
6.2.1. Expanding on the MCTS model for Ovarian Cancer ................................... 207
6.2.2. Delivering localized drug combinations ........................................................ 209
xi
6.3. References ........................................................................................................... 216
xii
List of Tables
Table 4-1: Characteristics of BCM-DOX at different drug to copolymer ratios and a constant copolymer concentration of 50 mg/mL (n = 3 independently produced batches). DLS = Dynamic Light Scattering, TEM = Transmission Electron Microscopy, LC = Loading Content, EE = Encapsulation Efficiency. ............................................................... 127
Table 5-1: Primers used in sequencing BRCA1 and BRCA2 regions of interest..................... 164
Table 5-2: Characteristics of the HGSOC cell panel selected for these studies. .................... 170
Table 5-3: Confirmation of BRCA1 or 2 status for PEO1, PEO4, UWB1.289, UWB1.289+BRCA1 and COV362 ovarian cancer cell lines. Gene sequences were confirmed as described in the Materials and Methods section. ............................................... 171
Table 5-4: MCTS growth studies for each cell line. Images show MCTS grown using the liquid-overlay technique 36 after initial seeding of 3000 cells per well. All scale bars represent 100 µm. .............................................................................................................................. 177
Table 5-5: Short Tandem Repeat Analysis for the cell lines used. All cell lines were identified according to their respective references. .................................................................... 192
xiii
List of Figures
Figure 2-1: Versatility of BCM chemistry. BCMs provide a flexible platform for the design of NDDSs given the synthetic versatility that enables customization at the molecular level. Among a wide range of factors which influence key physico-chemical properties and in vivo performance, we illustrate select parameters which have been varied as a means to control size, stability, loading and labeling of BCMs. ................................................... 26
Figure 2-2: Number of annual publications for select NDDSs. Despite the exponential growth in research activity over the last several decades, very few NDDSs have achieved translation from “bench to bedside”. Data show the total number of publications compiled as of March 2014 using Scopus® search engine with search terms "micelle", "liposome" or "polymer-drug conjugate” paired separately with "drug delivery”........................ 40
Figure 2-3: Challenges for BCM-based drug delivery. For the successful translation of BCM technology into the clinical space, three main delivery challenges are presented: I) stability in the blood compartment and prolonged circulation, II) penetration into deep layers of the tumor tissue and III) tumor bioavailability of the drug within cancer cells. I) Following systemic administration, BCMs must remain intact for effective delivery of the drug payload to the target site (a). However, a challenge to this feat is leakage of drug from the nanocarrier (b) or disruption of the integrity of the BCM and subsequent release of drug in its free form (c). II) Within the tumor, a dynamic network of blood vessels exists wherein perfusion of certain vessels may be deficient (e.g. transient or non-existent) (a). Following extravasation, tumor penetration of NDDSs is possible by means of diffusion (b), albeit hindered by the dense extra-cellular matrix and/or sequestration via stromal cells (c). III) Upon delivery of the shuttled therapy via BCMs at the tumor site, three scenarios may present themselves: the carrier system need not penetrate the tumor interstitium and the drug is released in the extracellular space (a), the carrier system penetrates into the tumor and drug is released into the extracellular space (b) and finally, the carrier system penetrates through the tissue and is internalized by a cell where the drug is released (c). ............................................................................................ 49
Figure 2-4: Beyond the EPR effect. For many years, the EPR-centric paradigm has dominated the development of NDDSs including BCMs. In this paradigm, NDDS stability, drug retention and long circulation in the bloodstream have been sought after in order to exploit the EPR effect which has been assumed to be directly predictive of therapeutic effect. However, given the tremendous heterogeneity in cancer, it is now understood that we must design NDDSs beyond the EPR effect alone and that parameters such as tumor penetration, tumor drug bioavailability and, in particular, patient EPR status (i.e. extent to which EPR is operative), have a combined impact on therapeutic efficacy. The proposed ultra-EPR paradigm incorporates this deeper understanding of biological response to NDDS mediated drug delivery and may enable patient stratification based on EPR status which could potentially not only increase therapeutic efficacy, but also reduce toxicities. .............................................................................. 62
Figure 3-1: 3-D cultures as intermediary between 2-D cultures and animal models. Intermediate in complexity, 3-D cultures permit the systematic, high-throughput assessment of formulation properties in a controlled environment that approximates
xiv
important properties of in vivo tumors in the absence of complex parameters which may confound data interpretation. ..................................................... Error! Bookmark not defined.
Figure 3-2: In vitro assays used in this study for analysis of formulation efficacy in spheroids. ............................................................................................................................................... 88
Figure 3-3: Characterization of micelle morphology and size. a) Transmission electron micrograph (Scale bar in represents 100 nm) and b) size distribution of BCM+DTX as determined by dynamic light scattering at 37 °C. ........................................................................... 91
Figure 3-4: Drug release. Release of docetaxel from dialysis bags containing BCM+DTX, Taxotere®, and DTX in DMSO, n = 3 ................................................................................................... 92
Figure 3-5: Spheroid packing density and growth. a) Cells per HeLa and HT29 spheroid of given volume, n = 12. b) Growth of HeLa and HT29 spheroids, n = 6. Data was fit using the Gompertz equation for tumor growth. The dashed lines indicate spheroid properties used in the studies. ........................................................................................................... 94
Figure 3-6: .Cytotoxicity of Taxotere® and BCM+DTX in spheroid and monolayer cultures. Viability of a) HeLa and b) HT29 cells cultured as monolayers and spheroids as measured using the APH assay. Data is expressed as the percent viability relative to untreated controls and fit to the Hill equation. c) Cytotoxicity of blank PEG-b-PCL micelles as a function of copolymer concentration. Each plot represents the mean of three independent experiments 6 SD (n = 3). .................................................................................. 98
Figure 3-7: Inhibition of spheroid growth. a) Sequential images of the same HeLa and HT29 spheroids following treatment with BCM+DTX at a concentration of 20 ng/mL. Bars represent 100 mm. Growth inhibition of HeLa (b,c) and HT29 (d,e) MCTS by BCM+DTX and Taxotere® at concentrations of 2, 20 and 200 ng/mL. Cells were re-treated after two weeks (arrow). Box represents expanded region of plots b) and d). Data is expressed as the mean volume of six spheroids (n = 6) ± SD. ‘‘*’’ represents a significant difference between BCM 20 and TAX 20, p < 0.05. ................................................. 100
Figure 3-8: Histological assessment of spheroid microenvironment. HeLa (a–c) and HT29 (d–f) MCTS cross-sections stained with H&E (a, d), Ki67 proliferation marker (b, e) and EF5 (c, f), a marker of hypoxia. Scale bars represent 100 mm. g) Properties of the spheroid microenvironment and their spatial distribution. ‘‘++’’, ‘‘+’’, and ‘‘–’’, indicate high, intermediate and low levels of the corresponding feature, respectively. ....................... 102
Figure 3-9: Spatial distribution of proliferating cells in spheroids. Ki67 positive signal distribution relative to radial position in a) HeLa and b) HT29 MCTS as a percent of total positive stain, n = 6 ............................................................................................................................ 103
Figure 3-10: Clonogenic potential of cells following treatment. Clonogenic survival of HeLa and HT29 cells following 24 h treatment with 20 ng/mL of BCM+DTX or Taxotere® as a) monolayers, b) disaggregated spheroids and c) intact spheroids. ............. 104
Figure 3-11: Measurement of spheroid volume. a) Schematic representation of the analysis process using a macro developed for ImageJ (version 1.44 m). b) Correlation between manual and automated volume measurements of HeLa MCTS. MCTS were
xv
imaged at selected intervals of growth. Manual measurement of MCTS volume was performed by determining the average of the largest and smallest diameters using the captured images and assuming a spherical MCTS morphology. Automated volume measurement was achieved using an image recognition technique in ImageJ. Firstly, MCTS images were converted into 8-bit greyscale and the perimeter of the MCTS was recognized by an automated threshold function. The area of the 2-D MCTS mask was recorded and converted to mm2 by calibration using an image of known scale and subsequently used to calculate the volume. ................................................................................ 105
Figure 3-12: Validation of the acid phosphatase (APH) assay. Results from the APH assay using HeLa (left column) and HT29 cells (right column) grown as spheroids (top row) and monolayers (bottom row) demonstrate a linear relationship between cell number and UV absorption at 405 nm. Each data point represents the mean of three independent experiments 6 SD (n=3). ............................................................................................ 106
Figure 3-13: Failure of WST-8 assay. Results from the WST-8 assay demonstrate a non-linear correlation between the number of cells and OD450 in spheroid culture. .................. 107
Figure 3-14: Fluorescence images of HT29 (a) and HeLa (b) tumor xenografts displaying markers of hypoxia (EF5 - blue) and blood vessels (CD31 - red). Scale bars represent 100 mm. ............................................................................................................................................... 107
Figure 4-1: Workflow for evaluating drug penetration in MCTS. Three MCTS per treatment group were imaged using fluorescent live cell imaging on a confocal microscope. Images were taken at 30, 60 and 90 µm (slices 1, 2 and 3) from the MCTS surface. For each of these optical slices, a custom MATLAB algorithm was used to determine the fluorescence intensity per unit area for three equally spaced concentric regions (periphery/P/red, intermediate/I/blue and core/C/green) that mirror the MCTS perimeter. For each of the three defined regions (red, blue and green), the average signal intensity per unit area within the three optical slices per MCTS was calculated. This value was finally reported as mean ± SD for n = 3 MCTS. ......................................................... 123
Figure 4-2: Representative TEM images of final BCM-DOX formulations at 0.1, 0.2 and 0.3 drug to copolymer feed ratios [wt/wt]. BCM-DOX formulations were imaged using a Hitachi 7000 microscope at 40000x magnification. ................................................................... 127
Figure 4-3: In vitro release of DOX from BCMs prepared from drug-to-copolymer ratios of 0.1 to 0.3 (wt:wt). Release studies were conducted in PBS (pH 7.4) and/or PBS containing BSA (50 mg/mL). ........................................................................................................... 129
Figure 4-4: Assessment of cytotoxicity of DOX, BCM-DOX and PLD in cell monolayers using the APH assay in HEYA8 (A), OV-90 (B) and SKOV3 (C) ovarian cancer cell lines after 48 h treatment (n = 3 per treatment). Under these conditions, BCM-DOX cytotoxicity was superior in all cell lines when compared to PLD (p < 0.05). A (*) denotes statistically significant differences between IC50 values resulting from the three treatments. .......................................................................................................................................... 131
Figure 4-5: MCTS growth studies for HEYA8 and OV-90 ovarian cancer cell lines. Different numbers of cells were seeded, and growth was monitored over a 15-day
xvi
period. The horizontal line indicates a diameter of 500 µm. Subsequent studies were conducted with initial cell numbers of 1000 cells/well and 2000 cells/well for HEYA8 and OV-90, respectively. This yielded a size of about 500 µm after 7 days of growth. Growth curves were fit to the Gompertz growth equation of tumor growth on a log plot of MCTS volume as a function of time.35 ...................................................................................... 133
Figure 4-6: Confocal images of MCTS with CellMaskTM Green Plasma Membrane Stain (A) and MCTS cellular density, expressed as cells per volume for both cell lines (B). Scale bars represent 100 µm. ..................................................................................................................... 135
Figure 4-7: Qualitative assessment of DOX penetration into MCTS formed from HEYA8 or OV-90 ovarian carcinoma cells. Representative images show drug penetration at a 50-µm depth from the MCTS surface, imaged using a Zeiss LSM700 confocal microscope with a 20x objective after 2 h incubation with each treatment. PLD visualization is limited due to low levels of free drug released, yielding low fluorescence intensity. Individual images represent 635 µm fields of view. Scale bars: 100 µm. .............. 139
Figure 4-8 (next page): Assessment of free DOX penetration into MCTS of HEYA8 and OV-90 ovarian carcinoma cell lines following incubation with DOX alone, BCM-DOX or PLD at 37 °C (NT) or 42 °C (MHT) with free drug penetration quantified using fluorescence imaging. The total fluorescence intensity per unit area was separately calculated for periphery (P), intermediate (I) and core (C) regions of each MCTS as outlined in the methods section. Data points represent the mean fluorescence intensity of each area within 3 MCTS ± SD. Statistically significant differences between each region within the same treatment are denoted by (*); between treatments for each specific region are denoted by (#); and between MHT and NT for each treatment and region are denoted by (). ................................................................................................................. 139
Figure 4-9: Representative images of free DOX when delivered as DOX alone or as BCM-DOX after 2 h incubation at a depth of 30 µm under NT and MHT. In general, DOX alone quickly accumulated within cell nuclei, whereas drug delivered via BCM-DOX accumulated substantially in the cell membrane (white arrows). Mild hyperthermia appeared to create a more diffuse drug distribution pattern for both delivery strategies and cell lines. PLD could not be visualized due to undetectably low free drug levels that yielded low fluorescence intensity. Scale bars: 10 µm. .............................................................. 141
Figure 4-10: HEYA8 and OV-90 MCTS growth inhibition of free DOX, BCM-DOX and PLD after 48 h drug exposure under NT or 1 h MHT followed by 47 h NT. Data points represent the mean volume of 3 - 6 MCTS ± SD. A () denotes instances when BCM-DOX performance is superior to PLD and comparable to that of free DOX (p < 0.05). ......... 143
Figure 4-11: Cytotoxicity of DOX under NT and MHT in HEYA8 and OV-90 cell monolayers, assessed after 48 h treatment using the APH assay. Data for MHT is reported relative to NT control and relative to MHT control. Although MHT does impact HEYA8 cell viability, it does not affect DOX-induced cytotoxicity in either cell line. .............. 144
Figure 5-1: IC50 values for cell lines after 72 h single-drug treatment with DOX (left) or OLP (right) alone. ............................................................................................................................... 172
xvii
Figure 5-2: IC50 values for cell lines, grouped as BRCA deficient or proficient, following treatment with DOX (left) or OLP (right). When the IC50 values for BRCA deficient (i.e. mutated or methylated) cell lines are considered together and compared to values for the BRCA proficient cell lines, no significant difference is identified. ...................................... 173
Figure 5-3: Summary of CI values for the combinations of DOX and OLP following 72h incubation in nine ovarian cancer cells lines at the indicated molar ratios required for Fa = 0.5 (A) and Fa = 0.75 (B). Each CI value was calculated and a heat map was generated as outlined in Materials and Methods on the basis of three independent IC50 experiments performed on different days for DOX, OLP and each combination ratio. ........ 174
Figure 5-4: Representative confocal images of cellular organization of OVCAR8, OV-90 and HEYA8 MCTS. MCTS were incubated for 1 h with 1X CellMaskTM Green plasma membrane stain in PBS, and the fluorescent stain was detected using confocal microscopy. Scale bars represent 100 µm. .................................................................................. 180
Figure 5-5: MCTS volume (left panels) and growth inhibition upon DOX treatment (right panels) for OVCAR8 (A, B), OV-90 (C, D) and HEYA8 (E, F) ovarian cancer cell lines. MCTS volume as a function of time is shown on a log plot fit to the Gompertz equation of tumor growth following seeding of 1000, 3000 and 5000 cells per well (A, C, E). The dotted line indicates a diameter of 500 µm, which was chosen as baseline for further studies. MCTS growth inhibition was assessed following treatment with DOX for a 72 h period (B, D, F). Every other day, 50 % of the media was replaced prior to imaging and determination of MCTS volume. ..................................................................................................... 181
Figure 5-6 (Next page): Growth inhibition studies for OVCAR8 (A), HEYA8 (B) and OV-90 (C) MCTS in response to treatment with DOX:OLP at select molar ratios in comparison to DOX alone. The initial DOX concentration was chosen as described above to be 1 µM for OVCAR8 MCTS and 0.2 µM for OV-90 and HEYA8 MCTS. OLP concentrations were calculated for each molar ratio based on each DOX concentration. Following the 72 h treatment period, 50 % of media was replaced every other day prior to imaging and determination of MCTS volume. A (*) indicates significant differences between the indicated treatment group pairs, while a (#) represents significant differences between DOX and each ratio (p 0.05). ......................................................................................................... 184
Figure 5-7: Effect of OLP treatment on growth inhibition of MCTS formed from OVCAR8 (A), OV-90 (B) and HEYA8 (C) ovarian cancer cells. MCTS growth inhibition was assessed following a 72h treatment with OLP. Every other day, 50 % of the media was replaced prior to imaging and determination of MCTS volume ................................................ 196
Figure 6-1: Formulation stability of the DOX:OLP solution incubated at 37 °C over a period of 14 days. Concentration of both drugs reduces slightly, however both remain over 85 % of initial levels after a two week period. ...................................................................... 211
Figure 6-2: Representative peaks for OLP, DOX and DAN (internal standard for DOX) after extraction from in vitro release media containing BSA. ............................................................. 212
xviii
Figure 6-3: In vitro release study of OLP and DOX using an ALZET pump immersed in physiological BSA solution at 37 °C. Both drugs release at the same percentage and therefore keep their relative ratio over the course of the study. ............................................... 214
xix
List of Abbreviations
ACN Acetonitrile
ACS American Chemical Society
ADP Adenosine diphosphate
APH Acid Phosphatase
BCM Block Copolymer Micelle
BCM-DOX BCM formulation containing DOX
BER Base Excision Repair
BSA Bovine Serum Albumin
CDDP Cis-DichloroDiamminePlatinum(II)
CEA Anti-CarcinoEmbryonic Antigen
CI Combination Index
CMC Critical Micelle Concentration
CT Computed Tomography
DAN Daunorubicin
DDSB DNA Double Strand Break
DLS Dynamic Light Scattering
DMF Dimethylformamide
DOX Doxorubicin
DSB Double Strand Break
DSSB DNA Single Strand Break
DTX Docetaxel
EE Encapsulation Efficiency
EGF Epidermal Growth Factor
EGFR Epidermal Growth Factor Receptor
EOC Epithelial Ovarian Cancer
EPR Enhanced Permeability and Retention
FBS Fetal Bovine Serum
H&E Haematoxylin and Eosin Tissue Staining
HGSOC High Grade Serous Ovarian Cancer
HIPEC Hyperthermic Intraperitoneal Chemotherapy
HRR Homologous Recombination Repair Pathway
ICP-MS Inductively Coupled Plasma Mass Spectrometry
IF ImmunoFluorescence
IFP Interstitial Fluid Pressure
ILSCM Intravital Laser Scanning Confocal Fluorescence Microscopy
LC Loading Content
MCTS Multicellular Tumor Spheroids
MHT Mild HyperThermia
MPS Mononuclear Phagocyte System
MRI Magnetic Resonance Imaging
MTD Maximum Tolerated Dose
xx
MVD Mean Vessel Density
MW Molecular Weight
NCI National Cancer Institute
NDDS Nano Drug Delivery System
NER Nucleotide Excision Repair
NHEJ Non-Homologous End Joining Pathway
NLS Nuclear Localization Signal
NT Normothermia
OC Ovarian Cancer
OLP Olaparib
ORR Overall Response Rate
OS Overall Survival
PARP Poly ADP ribose polymerase
PBS Phosphate Buffered Saline
PCL Polycaprolactone
PEG Polyethylene Glycol
PET Positron Emission Tomography
PFS Progression Free Survival
PK Pharmacokinetics
PLA Polylactide
PPE Palmar-Plantar Erythrodysesthesia
PTX Paclitaxel
ROI Region of Interest
RR Response Rate
SD Standard Deviation
SF Surviving Fraction
STR Short Tandem Repeat
TEA Triethylamine
TEM Transmission Electron Microscopy
TGF-β Transforming Growth Factor
UA Uranyl Acetate
VEGF Vascular Endothelial Growth Factor
1
Introduction, Hypothesis and Overview
Sina Eetezadi and Christine Allen
Written by S. Eetezadi. Edited by C. Allen.
Chapter 1: Introduction, Hypothesis and Overview 2
1.1. Introduction
1.1.1. Ovarian Cancer Treatment in Clinical Practice
Worldwide over 220 000 women are diagnosed with ovarian cancer every year.1 It is
predicted that in 2015 alone there will be 14,180 deaths from ovarian cancer in the United
States.2 In Canada, while ovarian cancer ranks 7th in incidence among all cancers in women,
it is 5th in mortality accounting for the highest mortality rate of all gynaecological
malignancies.3 Ovarian cancer is staged based on the FIGO system which consists of 4 main
stages with subcategories.4 Stage I is disease limited to the ovaries only, stage II indicates
disease extending to the pelvis, stage III abdominal disease or affected lymph nodes and
finally stage IV distant metastases. As a consequence of the asymptomatic nature of the
disease only one fifth of patients are diagnosed at stage I when the cancer is still confined to
the ovaries.1,2 The vast majority of patients present at stage III and IV with distant spread into
the abdomen and beyond at the time of diagnosis, facing a five-year survival rate post-
diagnosis of only 27 %.2,3 About 90 % of ovarian cancers are epithelial ovarian cancers (EOCs)
which are recognized as a heterogeneous disease and are classified as follows according to
histological subtypes: mucinous (3 %), low-grade serous (5 %), endometrioid (11 %), clear cell
(11 %) and high-grade serous (70 %).5,6 High-grade serous ovarian cancer (HGSOC) also
represents the most aggressive form of the disease and is characterized by almost ubiquitous
p53 mutation. As many as half of diagnosed HGSOCs have deficiencies in DNA repair
pathways.7 Most common are mutations in the BRCA1 and BRCA2 genes of the homologous
recombination repair (HRR) pathway, which is the high-fidelity repair mechanism for DNA
double strand breaks.8
First line intervention for advanced stage ovarian cancer is cytoreductive surgery, followed
by platinum-based chemotherapy. The goal of surgery is to remove as much of the tumor
Chapter 1: Introduction, Hypothesis and Overview 3
burden as possible, which has been shown to be beneficial for long-term survival.9 Achieving
residual tumor lesions, with each measuring < 1 cm in diameter, is commonly defined as
“optimal” cytoreduction whereas further reduction is deemed “ideal” where only microscopic
residual disease remains.10 Over the last decades, the standard follow-up chemotherapy
regimen has been and continues to be based on six cycles of carboplatinum together with
paclitaxel.1,11 Docetaxel has been evaluated as an alternative to paclitaxel and was found to
be equivalent; however, is not generally used in clinical practice as first-line therapy.12
Chemotherapy is generally administered intravenously, although clinical trials indicated
significant survival improvements with intraperitoneal chemotherapy.13,14 Yet, in a widely
discussed randomised clinical trial toxic side-effects limited applicability to only 42 % of
patients recruited and so far general uptake of this regimen has been poor in clinical
practice.15
Although most patients achieve complete clinical response following first-line therapy,
over 75 % will relapse within a median interval of 18 months, develop partial or full resistance
to platinum-based therapy and progressively establish a chronic disease.1,8 These patients
are classified into three categories depending on their time of recurrence after initial therapy
(i.e. the platinum-free interval). Platinum sensitive patients are those whose disease recurs in
more than 6 months and are generally re-challenged with another cycle of platinum-based
chemotherapy. This regimen is further continued until the patient eventually develops
resistance to platinum.1 Platinum-refractory patients are those whose disease progresses
during platinum treatment; while, in platinum resistant patients the disease recurs in less than
6 months after initial therapy.
To date, there is no consensus on the management of platinum refractory and resistant
patients.11,16 Commonly, a non-platinum single agent is given including topotecan or
Chapter 1: Introduction, Hypothesis and Overview 4
PEGylated liposomal doxorubicin (PLD, Doxil® / Caelix®). Objective response of platinum
resistant patients to all second-line agents is 10 – 15 %.17 Response is not superior with PLD,
although less hematologic and GI toxicities ensue.18,19 Currently, drugs targeting the VEGF
pathway such as bevacizumab (BEV, Avastin®) are also under intense clinical investigation;
however, their role in ovarian cancer remains unclear.20 BEV is approved by the FDA and the
European authorities for recurrent, platinum resistant disease in combination with
chemotherapy. In the EU approval of BEV was also granted for first-line therapy and platinum-
sensitive recurrence. However, given the high cost and unclear benefit, widespread clinical
uptake has so far not occurred.1,20 Finally, inhibitors of the poly(ADP-ribose) polymerase
(PARP) enzymes have shown promise in platinum-resistant ovarian cancer. Olaparib (OLP,
Lynparza®) is the first member of this class, which has recently been approved by the FDA for
refractory, advanced OC associated with germline BRCA mutations.21 Currently, PARP
inhibitors are mostly investigated for long-term maintenance therapy, since dose-limiting,
additive myelotoxicity has limited applications in combination therapy with cytotoxic
agents.1,22
1.1.2. The Role of PEGylated Liposomal Doxorubicin in Ovarian Cancer
PLD is a nanoformulation of DOX wherein the drug is loaded within PEGylated liposomes
with an average diameter of 90 nm.23 In recent years, it has become a major component in
the routine management of recurrent ovarian cancer and is approved for second-line therapy
in Canada as well as the United States for patients whose disease is refractory to the
platinum-based regimen.24 Because of its improved toxicity profile in comparison to other
agents, it is said to be considered by many clinicians as the first choice non-platinum agent
for relapsed ovarian cancer.24-28
Chapter 1: Introduction, Hypothesis and Overview 5
Of the various mechanisms of action proposed for the antineoplastic activity of DOX, DNA
intercalation and interaction with topoisomerase II appear to be the predominant ones at
clinically relevant drug concentrations of 1 µM to 2 µM.29 Crystal structures have shown that
the DOX molecule intercalates with its aromatic chromophore portion between two base pairs
of DNA while its daunosamine sugar unit interacts with bordering base pairs.30 It is
hypothesized that this results in a stabilized intermediary state of the topoisomerase II
enzyme, after it has split the DNA strand. The enzyme remains covalently attached to the DNA
preventing resealing and resulting in a DNA double strand break (DDSB).31 At sufficiently high
drug concentrations the cell’s ability to repair such damage is surpassed and apoptosis is
induced. Other mechanisms of action for DOX such as lipid peroxidation, direct effects on the
cell membrane as well as radical formation have been proposed, but appear to be only
relevant at concentrations exceeding the clinically relevant ones.29
The primary benefits of PLD are significant improvements in the pharmacokinetic and
toxicological profiles of DOX, in comparison to administration of the drug alone in the
conventional formulation. In preclinical models, it has been reported that encapsulation of
DOX in liposomes, leads to a 35-fold increase in half-life resulting in more than a 10-fold
increase in drug accumulation at the tumor site.32 Subsequent clinical trials have further
confirmed these results.33,34 This increase in tumor accumulation is driven by the enhanced
permeability and retention (EPR) effect, put forward by Matsumura and Maeda in 1986,
wherein it was reported that macromolecules with a size of 40 - 800 kDa accumulate within
solid tumors due to leaky vasculature and are retained at the site owing to poor lymphatic
drainage.35,36
Treatment with conventional DOX is often accompanied by toxicity that limits the
therapeutic benefit, most notably congestive heart failure and chronic cardiomyopathy.37 The
Chapter 1: Introduction, Hypothesis and Overview 6
causes of DOX cardiotoxicity are still not fully elucidated but likely different from its tumor cell
killing mechanism, because cardiomyocytes are minimally replicating cells.38 Commonly, iron
oxidation and reactive oxygen species generation have been proposed, however the failure of
anti-oxidants to alleviate these effects in clinical trials may suggest other mechanisms as the
root cause.29,38 While PLD leads to significant improvements in cardiotoxicity, clinical studies
have also revealed that PLD is associated with the incidence of dose-limiting palmar-plantar
erythrodysesthesia (PPE) also known as “hand-foot syndrome”.39
1.1.3. Targeting DNA Repair in Ovarian Cancer
Mutations in the DNA repair genes BRCA1 and BRCA2 of the homologous recombination
repair (HRR) pathway are a common trait of ovarian cancers, especially of the high-grade
serous histotype.40 HRR facilitates accurate repair of DNA double strand breaks (DDSB) by
gap-filling DNA synthesis using the homologous sister chromatid as a template.41,42
Mutations in the BRCA1/2 genes are associated on one hand with an increased risk of
developing ovarian cancer,43 however, on the other hand also with significantly better survival
and improved chemotherapy response once the disease has developed.44,45 This is expected,
since HRR deficiency confers sensitivity of cancer cells to DDSB inducing agents such as
platinum and DOX.45
Overall, it is suggested that over 50 % of HGSOC present with DNA repair deficiencies.8,17
About 10 to 15 % of patients have inherited germline BRCA mutations, which is particularly
common in the HGSOC phenotype.8,40 Additionally, somatic mutations of both genes occur,
such that about 20 % of all ovarian cancers show BRCA-related deficiency in HRR.46,47
Moreover, some tumors express an epigenetic profile or mutations in BRCA related genes
that have been described by the term “BRCAness” and result in phenotypic traits similar to
Chapter 1: Introduction, Hypothesis and Overview 7
mutation carriers. These cells are more susceptible to DNA damage, which can arise for
example due to BRCA1 silencing, EMSY amplification, BRCA2 transcriptional silencing, and
deficiencies in RAD51, ATM, ATR, Chk1 and 2, and the FA protein complex.45,47,48
Besides HRR there are various other DNA repair pathways that together form a
sophisticated defence mechanism against genotoxic damage, which is vital for cell survival.
Non-homologous end-joining (NHEJ) is an error-prone DDSB repair pathway which takes part
in the repair of DDSBs; however, unlike HRR without use of a sister chromatid template. The
mismatch repair pathway recognizes and repairs mismatched bases. The nucleotide excision
repair (NER) pathway removes adducts induced by ultraviolet radiation which leads to DNA
distortion. Finally, base excision repair (BER) is the predominant mechanism in the repair of
DNA single strand breaks (DSSB).42 Poly (ADP-ribose) polymerases (PARP), a family of
nuclear protein enzymes plays a key role in BER. Specifically, PARP localizes to DSSBs and
synthesizes poly (ADP) ribose chains as a signal to recruit DNA repair proteins which mount
the enzymatic machinery of BER.49,50 However, if left unrepaired, DSSBs can give rise to
DDSBs that in turn would be primarily repaired by HRR, providing the cell has functional, non-
BRCA mutated HRR. Therefore, BRCA and PARP are central to DNA repair and the ability to
maintain normal cell integrity and function.
If DNA damage is extensive and irreparable, induction of cell death occurs. Consequently,
for BRCA mutated tumors this can be exploited by using inhibitors of PARP, which represent
a novel class of drugs targeting the BER DNA repair pathway. They impair the ability of PARP
to recruit BER enzymes and stabilize PARP at the sites of DNA damage, thereby impeding
DSSB repair, which results in DDSB and ultimately cell death. This concept known as ‘synthetic
lethality’ has shown promise in ovarian cancer, as it leaves the HRR impaired cancer cell
without a reliable defense mechanism against DNA damage.51,52 PARP inhibition renders BER
Chapter 1: Introduction, Hypothesis and Overview 8
non-functional and as a result, HRR defective tumor cells depend on error-prone NHEJ,
yielding unrepaired DNA damage and ultimately inducing apoptosis.49,53 Furthermore, a low
toxicity profile can be expected from 'synthetic lethality' as a treatment strategy since healthy
non-mutated tissues should not be affected.53 Based on the concept of synthetic lethality
PARP inhibitors such as olaparib (OLP, Lynparza® early names: AZD- 2281/KU-0059436) have
been extensively studied as monotherapy in clinical trials.54-56. While significant
improvements in progression-free survival (PFS) has been shown, no improvement in overall
survival (OS) was detected.57,58 Still, based on the clinical trial results, Lynparza® became the
first-in-class drug that was approved in December 2014 by the EU and US authorities for
recurrent BRCA-mutated ovarian cancer following at least 3 cycles of platinum-based
therapy.58 Lynparza® is available as an oral capsule formulation of 400 mg twice daily and is
approved to be given until disease progression.
PARP inhibitors have also been investigated as a strategy to potentiate the cytotoxic
effects of DNA-damaging agents by inhibiting the DNA repair process following treatment.
Chemotherapy drugs, such as DOX rely on the induction of DNA damage in rapidly
proliferating tumor cells and consequently the inhibition of DNA repair should increase their
treatment efficacy or permit the administration of lower doses without compromising
efficacy. Despite pre-clinical evidence demonstrating tumor sensitization using PARP
inhibitors in combination with various cytotoxic agents, the optimal usage in cancer treatment
remains unclear, since outcomes vary depending on drug combination and the animal model
used.59-62 In the clinical setting, various combinations that include OLP and a cytotoxic drug
have been evaluated, but in all cases the addition of a PARP inhibitor to chemotherapy
increased the degree of myelosuppression necessitating dose reductions. As such, dose-
limiting myelosupression was observed when using OLP together with gemcitabine63,64 and
Chapter 1: Introduction, Hypothesis and Overview 9
topotecan65 requiring dose reductions of both agents that resulted in the administration of
sub-therapeutic drug concentrations. The etiology of the myelosuppression could not be
elucidated in these trials, since gemcitabine and topotecan both induce a high degree of
myelosuppression already as single-agents and therefore the contribution of the PARP
inhibitor is unclear. In a carefully conducted dose-escalation study of OLP together with PLD,
the combination was found to be more tolerable in comparison to the studies mentioned
above, yet still 27 % of patients suffered serious adverse events.66
In all trials it was assumed that maximal therapeutic effect is achieved when the maximum
tolerated dose (MTD) of both drugs is employed. However, clinical efficacy may also depend
on the molar ratios of the combined drugs and as shown in a number of studies specific ratios
may be more beneficial than others.67,68 Furthermore, by means of nanotechnology the
exposure of healthy tissue to cytotoxic drug may be reduced, which could in turn reduce the
dose-limiting toxicities observed in clinical trials. For instance, whereas myelosuppression is
an acute dose-limiting toxicity of free DOX, it is rather mild in patients treated with PLD.69
Indeed, the Phase 1 clinical trial mentioned above reported improved tolerability of the
combination of OLP with PLD and proposed the combination for further investigation.66
Currently, a number of Phase I clinical trials are investigating combinations of OLP and
platinum or paclitaxel (NCT01650376, NCT01081951, NCT01237067, NCT02418624) or
doxorubicin (NCT00819221).
Chapter 1: Introduction, Hypothesis and Overview 10
1.2. Rationale
The work presented herein aims to improve the therapeutic outcome associated with
recurrent ovarian cancer by (1) creating a 3D cell screening method as an in vitro model of
the disease (2) developing a nanoformulation of DOX that is more efficacious than PLD and
(3) evaluating additional strategies to enhance treatment efficacy such as mild hyperthermia
(MHT) and combination therapy with PARP inhibitors.
As outlined above, small avascular tumor nodules of less than 1 cm in diameter remain
following optimal cytoreductive surgery.70 Moreover, aggregates of cells can be found in the
patient’s ascites fluid promoting the metastatic dissemination of the disease, especially
during recurrence.71 Previously, it has been reported in the literature that such cell aggregates
termed multicellular tumor spheroids (MCTS) can be created and cultured in vitro.72,73 MCTS
of ovarian cancer cells that are cultured in vitro and of similar size to that of tumor nodules in
vivo, may serve as an accurate model of ovarian cancer post-surgery.70 As such, an in vitro
model of intermediate complexity between oversimplified 2D cell culture and in vivo animal
models may be a promising tool to evaluate the efficacy of treatment modalities in a high-
throughput manner. Specifically, in contrast to in vivo tumor models, in vitro cultures are
believed to be better suited for systematic studies of formulation parameters in a highly
controlled environment.
For treatment of recurrent disease, the introduction of PLD into the clinical setting has
brought amelioration in patients’ quality of life, yet better therapeutic options are still in need
given that response rates (RRs) to PLD in platinum-resistant, recurrent ovarian cancer are only
9 % - 16 %.74 From a formulation scientist’s perspective, there are two key elements of the
PLD nanoformulation that are in need of improvement. Firstly, PLD and liposomes in general
are stable nanoparticles that are known to retain their cargo in particular if this includes
Chapter 1: Introduction, Hypothesis and Overview 11
hydrophilic or amphiphilic drugs. This is beneficial in terms of toxicity, given that the drug is
well retained within the liposomes until reaching the target site. Yet, the high stability of the
nanoparticles also limits efficacy since the drug may not be fully released from the liposomes
at the tumor site.75 Therefore while PLD results in a significant enhancement in tumor
accumulation of DOX, it has been found that only half of the drug is released from the
liposomes at the tumor site and thus available to elicit its effect.75 Secondly, the relatively
large size of the liposomes impedes penetration into the tumor tissue and consequently
cancer cells deeper within the tissue may not be sufficiently exposed to therapeutic drug
concentrations.76,77
By designing a small-sized, stable block copolymer micelle (BCM) formulation it is
hypothesized that great improvements in tumor bioavailability and tumor penetration over
PLD can be achieved, while gaining high tumor accumulation and maintaining a relatively
favourable toxicity profile that is comparable to that for PLD. Indeed, it has been found that
liposomes of 90 nm in diameter only penetrate the tumor interstitium up to about 30 µm away
from blood vessels,78 whereas studies indicate that 50 nm micelles penetrate up to 100 µm
into the tissue.79,80 Block copolymer micelles represent a relatively simple formulation
strategy, and are small-sized nanomedicines that can result in high tumor bioavailability of
hydrophobic drugs.81,82 In aqueous media, block copolymer micelles are formed by self-
assembly of amphiphilic copolymers and consist of a hydrophobic core surrounded by a
hydrophilic shell. BCMs can be prepared to range in size from 10 to 100 nm. The hydrophobic
core offers a suitable environment for encapsulation of hydrophobic drugs thereby enhancing
the drug’s solubility in aqueous media and shielding the drug from the biological milieu.83,84
While optimization of the delivery of DOX would benefit ovarian cancer therapy,
improvements may also be achievable by modulating and enhancing the therapeutic effect
Chapter 1: Introduction, Hypothesis and Overview 12
of the drug itself. Therefore, this project also involved the evaluation of two strategies to
enhance drug efficacy: (1) addition of mild hyperthermia and (2) combination therapy with
PARP inhibitors. Hyperthermic intraperitoneal chemotherapy (HIPEC) defined as heating of a
drug solution to around 42 °C for 1 h during administration of chemotherapy may improve the
tissue penetration of nanomedicines. This administration of chemotherapy under mild
hyperthermic conditions has been reported to enhance tumor penetration and intracellular
uptake of drugs in clinical trials of ovarian cancer.85,86 In particular, DOX is one of the agents
commonly used for HIPEC, due to its stability to heat and high molecular weight.87 The second
strategy that was pursued to improve the efficacy of DOX in the treatment of recurrent ovarian
cancer, and especially HGSOC, was integration of a PARP inhibitor. DOX inhibits DNA
polymerase activity and topoisomerase II causing DSSB and DDSB breaks.74 Further,
inhibition of PARP enhances DOX antitumor activity in p53-deficient cancers,59,88 which is
promising for HGSOC, as 90 - 100 % of cases show p53 mutations.8,17,89 Additionally, PARP
inhibition has been reported to increase topoisomerase II expression, thereby specifically
potentiating the action of DOX.60 In this work OLP was selected as the PARP inhibitor to be
combined with DOX, given that it was most advanced in clinical trials. However, optimal
strategies for dosing and administration of this combination have not been elucidated,90 and
it is believed that determining optimal molar drug ratios of both drugs will increase efficacy,
while minimizing dose-limiting toxicities.
Chapter 1: Introduction, Hypothesis and Overview 13
1.3. Hypothesis
Delivery of DOX in BCMs will enhance both the intratumoral distribution and tumor
bioavailability of DOX resulting in a significant improvement in growth inhibition, relative to
PLD, in clinically relevant 3D cell models of ovarian cancer. An additional goal of this research
is to examine the impact of combined administration of OLP and DOX on the therapeutic
outcomes in ovarian cancer +/- BRCA mutations.
In order to address this hypothesis the following specific objectives have been identified:
Objective 1:
Development of 3D cell culture models as a tool to evaluate tissue penetration and growth
inhibition of nanomedicines
Objective 2:
Formulation and physico-chemical characterization of a polymeric micelle formulation of
doxorubicin (BCM-DOX).
Objective 3:
Evaluation of tissue penetration and growth inhibition of BCM-DOX in comparison to PLD in
in vitro 3D MCTS cultures.
Objective 4:
In vitro evaluation of combined effects of DOX and OLP in monolayers and 3D cell culture
models formed from different ovarian cancer cell lines +/- BRCA mutations.
Chapter 1: Introduction, Hypothesis and Overview 14
1.4. Overview of Thesis Chapters
The thesis presented herein is divided into six chapters. The first chapter includes an
introduction and the last chapter contains conclusions and a description of possible future
directions.
The second chapter provides an extensive overview of the current state of clinical
development of BCMs and the challenges associated with the effective design and
development of this class of nanomedicines. The impact of critical features, such as the size,
stability and functionalization of BCMs is discussed, while key pre-clinical endpoints and
models are critiqued. Given clinical considerations, this chapter also proposes an “ultra-EPR”
framework to guide future development of micellar nano drug delivery systems.
The third chapter describes the development of MCTS as a 3D cell culture model. For this
purpose two model cell lines, HeLa and HT29 were selected based on their ability to reliably
form homogenous MCTS. A series of complementary assays were established and validated
for evaluation of the in vitro efficacy of docetaxel (DTX) -loaded block copolymer micelles
(BCM+DTX) based on PEG-b-PCL in comparison to Taxotere® as corresponding model
systems. This work led to the validation of the acid phosphatase assay (APH) for use in MCTS
and development of the growth inhibition MCTS assay, which were both heavily employed in
future chapters. Overall, this work demonstrates the use of MCTS as a viable platform for the
evaluation of nanomedicines under conditions that more closely reflect the in vivo tumor
microenvironment relative to traditional monolayer cultures.
The fourth chapter applies the knowledge gained in chapters two and three to the
development and evaluation of a treatment strategy for recurrent ovarian cancer. Based on
the same copolymer, namely, PEG-b-PCL, a novel micellar formulation for DOX was developed
and characterized (BCM-DOX). Using MCTS as a model of residual disease post-surgery,
Chapter 1: Introduction, Hypothesis and Overview 15
BCM-DOX was compared to PLD in terms of in vitro release, monolayer cytotoxicity as well as
DOX tissue penetration and efficacy in MCTS. Additionally, the study examines the impact of
mild hyperthermia (MHT) on the cytotoxicity of DOX. BCM-DOX achieved the benefits of an
extended release formulation of DOX and resulted in improvements in free drug accumulation
over PLD, while yielding drug levels approaching that achievable through exposure to DOX
alone. In comparison to PLD, this translated into superior MCTS growth inhibition in the short-
term and comparable inhibition in the long-term.
Finally, the fifth chapter examines the combination of DOX and OLP with an emphasis on
identification of the optimal molar combination ratio of these agents that yields synergistic
effects in vitro. The combination effects of DOX and OLP were evaluated in 2D and 3D cell
culture models in a unique panel of 9 ovarian cancer cell lines with or without HRR
deficiencies, such as those with BRCA1/2 mutations or epigenetic modifications. Synergistic,
additive or antagonistic effects of DOX and OLP were determined on the basis of combination
index values obtained using the Chou and Talalay method. With the cell lines that reliably grew
as MCTS, growth inhibition assays were performed for select molar ratios. Overall, this study
demonstrates the importance of examining multiple drug ratios in a significant number of cell
lines to determine maximally synergistic ratios of combinations of a PARP inhibitor together
with a DNA-damaging agent in HGSOC. As such, response to treatment was found to be highly
heterogeneous from one cell line to another and not predictable by BRCA mutation status.
Furthermore, while ratios that yield synergistic effect could be determined, others resulted in
additive and even antagonistic effects.
Chapter 1: Introduction, Hypothesis and Overview 16
1.5. References
1. Jayson GC, Kohn EC, Kitchener HC, Ledermann JA. Ovarian cancer. Lancet. Oct 11 2014;384(9951):1376-1388.
2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA: a cancer journal for clinicians. Jan-Feb 2015;65(1):5-29.
3. Canadian Cancer Society's Advisory Committee on Cancer Statistics. Canadian Cancer Statistics 2015. Canadian Cancer Society,. 2015.
4. Prat J, Oncology FCoG. FIGO's staging classification for cancer of the ovary, fallopian tube, and peritoneum: abridged republication. Journal of gynecologic oncology. Apr 2015;26(2):87-89.
5. Vargas AN. Natural history of ovarian cancer. Ecancermedicalscience. 2014;8:465.
6. Groen RS, Gershenson DM, Fader AN. Updates and emerging therapies for rare epithelial ovarian cancers: One size no longer fits all. Gynecologic oncology. Feb 2015;136(2):373-383.
7. Bell D, Berchuck A, Birrer M, et al. Integrated genomic analyses of ovarian carcinoma. Nature. Jun 30 2011;474(7353):609-615.
8. Cooke SL, Brenton JD. Evolution of platinum resistance in high-grade serous ovarian cancer. The lancet oncology. Nov 2011;12(12):1169-1174.
9. du Bois A, Reuss A, Pujade-Lauraine E, Harter P, Ray-Coquard I, Pfisterer J. Role of Surgical Outcome as Prognostic Factor in Advanced Epithelial Ovarian Cancer: A Combined Exploratory Analysis of 3 Prospectively Randomized Phase 3 Multicenter Trials By the Arbeitsgemeinschaft Gynaekologische Onkologie Studiengruppe Ovarialkarzinom (AGO-OVAR) and the Groupe d'Investigateurs Nationaux Pour les Etudes des Cancers de l'Ovaire (GINECO). Cancer-Am Cancer Soc. Mar 15 2009;115(6):1234-1244.
10. Elattar A, Bryant A, Winter-Roach BA, Hatem M, Naik R. Optimal primary surgical treatment for advanced epithelial ovarian cancer. Cochrane Database Syst Rev. 2011(8):CD007565.
11. Coleman RL, Monk BJ, Sood AK, Herzog TJ. Latest research and treatment of advanced-stage epithelial ovarian cancer. Nature reviews. Clinical oncology. Apr 2013;10(4):211-224.
12. Vasey PA, Jayson GC, Gordon A, et al. Phase III randomized trial of docetaxel-carboplatin versus paclitaxel-carboplatin as first-line chemotherapy for ovarian carcinoma. Journal of the National Cancer Institute. Nov 17 2004;96(22):1682-1691.
13. Teo MCC. Update on the management and the role of intraperitoneal chemotherapy for ovarian cancer. Current opinion in obstetrics & gynecology. Feb 2014;26(1):3-8.
14. Armstrong DK, Bundy B, Wenzel L, et al. Intraperitoneal cisplatin and paclitaxel in ovarian cancer. The New England journal of medicine. Jan 5 2006;354(1):34-43.
15. Grosso G, Rossetti D, Coccolini F, Bogani G, Ansaloni L, Frigerio L. Intraperitoneal chemotherapy in advanced epithelial ovarian cancer: a survey. Arch Gynecol Obstet. Sep 2014;290(3):425-434.
16. Banerjee S, Kaye SB. New strategies in the treatment of ovarian cancer: current clinical perspectives and future potential. Clinical cancer research : an official journal of the American Association for Cancer Research. Mar 1 2013;19(5):961-968.
Chapter 1: Introduction, Hypothesis and Overview 17
17. Davis A, Tinker AV, Friedlander M. "Platinum resistant" ovarian cancer: What is it, who to treat and how to measure benefit? Gynecologic oncology. Jun 2014;133(3):624-631.
18. Cronin B, Robison K, Raker C, Moore R, Granai C, Dizon D. Pegylated Liposomal Doxorubicin in Recurrent Ovarian Cancer: Is There a Role for Maintenance Therapy? Clinical Ovarian and Other Gynecologic Cancer. 2014;6(1/2):17-20.
19. Gibson JM, Alzghari S, Ahn C, Trantham H, La-Beck NM. The role of pegylated liposomal doxorubicin in ovarian cancer: a meta-analysis of randomized clinical trials. The oncologist. 2013;18(9):1022-1031.
20. Shu CA, Konner JA. Breaking down the evidence for bevacizumab in ovarian cancer. The oncologist. Feb 2015;20(2):91-93.
21. Deeks ED. Olaparib: first global approval. Drugs. Feb 2015;75(2):231-240.
22. Syrios J, Banerjee S, Kaye SB. Advanced epithelial ovarian cancer: from standard chemotherapy to promising molecular pathway targets--where are we now? Anticancer Res. May 2014;34(5):2069-2077.
23. Barenholz Y. Doxil (R) - The first FDA-approved nano-drug: Lessons learned. Journal of Controlled Release. Jun 10 2012;160(2):117-134.
24. Markman M. Pegylated liposomal doxorubicin: appraisal of its current role in the management of epithelial ovarian cancer. Cancer management and research. 2011;3:219-225.
25. Lawrie TA, Bryant A, Cameron A, Gray E, Morrison J. Pegylated liposomal doxorubicin for relapsed epithelial ovarian cancer. Cochrane Database Syst Rev. 2013;7:CD006910.
26. Mutch DG, Orlando M, Goss T, et al. Randomized phase III trial of gemcitabine compared with pegylated liposomal doxorubicin in patients with platinum-resistant ovarian cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. Jul 1 2007;25(19):2811-2818.
27. Ferrandina G, Ludovisi M, Lorusso D, et al. Phase III trial of gemcitabine compared with pegylated liposomal doxorubicin in progressive or recurrent ovarian cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. Feb 20 2008;26(6):890-896.
28. Gordon AN, Fleagle JT, Guthrie D, Parkin DE, Gore ME, Lacave AJ. Recurrent epithelial ovarian carcinoma: a randomized phase III study of pegylated liposomal doxorubicin versus topotecan. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. Jul 15 2001;19(14):3312-3322.
29. Gewirtz DA. A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol. Apr 1 1999;57(7):727-741.
30. Frederick CA, Williams LD, Ughetto G, et al. Structural comparison of anticancer drug-DNA complexes: adriamycin and daunomycin. Biochemistry. Mar 13 1990;29(10):2538-2549.
31. Tacar O, Sriamornsak P, Dass CR. Doxorubicin: an update on anticancer molecular action, toxicity and novel drug delivery systems. The Journal of pharmacy and pharmacology. Feb 2013;65(2):157-170.
32. Laginha KM, Verwoert S, Charrois GJR, Allen TM. Determination of doxorubicin levels in whole tumor and tumor nuclei in murine breast cancer tumors. Clinical Cancer Research. Oct 1 2005;11(19):6944-6949.
Chapter 1: Introduction, Hypothesis and Overview 18
33. Symon Z, Peyser A, Tzemach D, et al. Selective delivery of doxorubicin to patients with breast carcinoma metastases by stealth liposomes. Cancer-Am Cancer Soc. Jul 1 1999;86(1):72-78.
34. Gabizon A, Catane R, Uziely B, et al. Prolonged circulation time and enhanced accumulation in malignant exudates of doxorubicin encapsulated in polyethylene-glycol coated liposomes. Cancer Res. Feb 15 1994;54(4):987-992.
35. Maeda H. Macromolecular therapeutics in cancer treatment: the EPR effect and beyond. Journal of controlled release : official journal of the Controlled Release Society. Dec 10 2012;164(2):138-144.
36. Matsumura Y, Maeda H. A new concept for macromolecular therapeutics in cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor agent smancs. Cancer Res. Dec 1986;46(12 Pt 1):6387-6392.
37. Minotti G, Menna P, Salvatorelli E, Cairo G, Gianni L. Anthracyclines: Molecular advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacological Reviews. Jun 2004;56(2):185-229.
38. Shi Y, Moon M, Dawood S, McManus B, Liu PP. Mechanisms and management of doxorubicin cardiotoxicity. Herz. Jun 2011;36(4):296-305.
39. Waterhouse DN, Tardi PG, Mayer LD, Bally MB. A comparison of liposomal formulations of doxorubicin with drug administered in free form: changing toxicity profiles. Drug safety. 2001;24(12):903-920.
40. Risch HA, McLaughlin JR, Cole DE, et al. Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer. Am J Hum Genet. Mar 2001;68(3):700-710.
41. Jekimovs C, Bolderson E, Suraweera A, Adams M, O'Byrne KJ, Richard DJ. Chemotherapeutic compounds targeting the DNA double-strand break repair pathways: the good, the bad, and the promising. Frontiers in oncology. 2014;4:86.
42. Bouwman P, Jonkers J. The effects of deregulated DNA damage signalling on cancer chemotherapy response and resistance. Nat Rev Cancer. Sep 2012;12(9):587-598.
43. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet. May 2003;72(5):1117-1130.
44. Konstantinopoulos PA, Spentzos D, Karlan BY, et al. Gene Expression Profile of BRCAness That Correlates With Responsiveness to Chemotherapy and With Outcome in Patients With Epithelial Ovarian Cancer. Journal of Clinical Oncology. Aug 1 2010;28(22):3555-3561.
45. Muggia F, Safra T. 'BRCAness' and Its Implications for Platinum Action in Gynecologic Cancer. Anticancer Res. Feb 2014;34(2):551-556.
46. Hennessy BTJ, Timms KM, Carey MS, et al. Somatic Mutations in BRCA1 and BRCA2 Could Expand the Number of Patients That Benefit From Poly (ADP Ribose) Polymerase Inhibitors in Ovarian Cancer. Journal of Clinical Oncology. Aug 1 2010;28(22):3570-3576.
47. Rigakos G, Razis E. BRCAness: finding the Achilles heel in ovarian cancer. The oncologist. 2012;17(7):956-962.
Chapter 1: Introduction, Hypothesis and Overview 19
48. O'Sullivan CC, Moon DH, Kohn EC, Lee JM. Beyond Breast and Ovarian Cancers: PARP Inhibitors for BRCA Mutation-Associated and BRCA-Like Solid Tumors. Frontiers in oncology. 2014;4:42.
49. Do K, Chen AP. Molecular pathways: targeting PARP in cancer treatment. Clinical cancer research : an official journal of the American Association for Cancer Research. Mar 1 2013;19(5):977-984.
50. De Vos M, Schreiber V, Dantzer F. The diverse roles and clinical relevance of PARPs in DNA damage repair: current state of the art. Biochem Pharmacol. Jul 15 2012;84(2):137-146.
51. Curtin NJ. DNA repair dysregulation from cancer driver to therapeutic target. Nat Rev Cancer. Dec 2012;12(12):801-817.
52. Martin SA, Hewish M, Lord CJ, Ashworth A. Genomic instability and the selection of treatments for cancer. The Journal of pathology. Jan 2010;220(2):281-289.
53. Eskander RN, Tewari KS. PARP inhibition and synthetic lethality in ovarian cancer. Expert review of clinical pharmacology. Sep 2014;7(5):613-622.
54. Audeh MW, Carmichael J, Penson RT, et al. Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and recurrent ovarian cancer: a proof-of-concept trial. Lancet. Jul 24 2010;376(9737):245-251.
55. Kaye SB, Lubinski J, Matulonis U, et al. Phase II, open-label, randomized, multicenter study comparing the efficacy and safety of olaparib, a poly (ADP-ribose) polymerase inhibitor, and pegylated liposomal doxorubicin in patients with BRCA1 or BRCA2 mutations and recurrent ovarian cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. Feb 1 2012;30(4):372-379.
56. Syrios J, Banerjee S, Kaye SB. Advanced Epithelial Ovarian Cancer: From Standard Chemotherapy to Promising Molecular Pathway Targets - Where Are we Now? Anticancer Res. May 2014;34(5):2069-2077.
57. Ledermann J, Harter P, Gourley C. Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial (vol 15, pg 852, 2014). Lancet Oncology. Apr 2015;16(4):E158-E158.
58. Deeks ED. Olaparib: First Global Approval. Drugs. Feb 2015;75(2):231-240.
59. Munoz-Gamez JA, Martin-Oliva D, Aguilar-Quesada R, et al. PARP inhibition sensitizes p53-deficient breast cancer cells to doxorubicin-induced apoptosis. The Biochemical journal. Feb 15 2005;386(Pt 1):119-125.
60. Magan N, Isaacs RJ, Stowell KM. Treatment with the PARP-inhibitor PJ34 causes enhanced doxorubicin-mediated cell death in HeLa cells. Anticancer Drugs. Jul 2012;23(6):627-637.
61. Rottenberg S, Jaspers JE, Kersbergen A, et al. High sensitivity of BRCA1-deficient mammary tumors to the PARP inhibitor AZD2281 alone and in combination with platinum drugs. Proceedings of the National Academy of Sciences of the United States of America. Nov 4 2008;105(44):17079-17084.
62. Donawho CK, Luo Y, Luo YP, et al. ABT-888, an orallyactive poly(ADP-ribose) polymerase inhibitor that potentiates DNA-damaging agents in preclinical tumor models. Clinical Cancer Research. May 1 2007;13(9):2728-2737.
Chapter 1: Introduction, Hypothesis and Overview 20
63. Rajan A, Carter CA, Kelly RJ, et al. A phase I combination study of olaparib with cisplatin and gemcitabine in adults with solid tumors. Clinical cancer research : an official journal of the American Association for Cancer Research. Apr 15 2012;18(8):2344-2351.
64. Bendell J, O'Reilly EM, Middleton MR, et al. Phase I study of olaparib plus gemcitabine in patients with advanced solid tumours and comparison with gemcitabine alone in patients with locally advanced/metastatic pancreatic cancer. Ann Oncol. Apr 2015;26(4):804-811.
65. Samol J, Ranson M, Scott E, et al. Safety and tolerability of the poly(ADP-ribose) polymerase (PARP) inhibitor, olaparib (AZD2281) in combination with topotecan for the treatment of patients with advanced solid tumors: a phase I study. Invest New Drugs. Aug 2012;30(4):1493-1500.
66. Del Conte G, Sessa C, von Moos R, et al. Phase I study of olaparib in combination with liposomal doxorubicin in patients with advanced solid tumours. Br J Cancer. Aug 12 2014;111(4):651-659.
67. Carol H, Fan MMY, Harasym TO, et al. Efficacy of CPX-351, (Cytarabine: Daunorubicin) Liposome Injection, Against Acute Lymphoblastic Leukemia (ALL) Xenograft Models of the Pediatric Preclinical Testing Program. Pediatr Blood Cancer. Jan 2015;62(1):65-71.
68. Dicko A, Mayer LD, Tardi PG. Use of nanoscale delivery systems to maintain synergistic drug ratios in vivo. Expert Opin Drug Del. Dec 2010;7(12):1329-1341.
69. Gordon AN, Granai CO, Rose PG, et al. Phase II study of liposomal doxorubicin in platinum- and paclitaxel-refractory epithelial ovarian cancer. Journal of Clinical Oncology. Sep 2000;18(17):3093-3100.
70. Elattar A, Bryant A, Winter-Roach BA, Hatem M, Naik R. Optimal primary surgical treatment for advanced epithelial ovarian cancer. Cochrane Db Syst Rev. 2011(8).
71. Shield K, Ackland ML, Ahmed N, Rice GE. Multicellular spheroids in ovarian cancer metastases: Biology and pathology. Gynecologic oncology. Apr 2009;113(1):143-148.
72. Hirschhaeuser F, Menne H, Dittfeld C, West J, Mueller-Klieser W, Kunz-Schughart LA. Multicellular tumor spheroids: An underestimated tool is catching up again. Journal of biotechnology. Jul 1 2010;148(1):3-15.
73. Friedrich J, Seidel C, Ebner R, Kunz-Schughart LA. Spheroid-based drug screen: considerations and practical approach. Nature protocols. 2009;4(3):309-324.
74. Adams SF, Marsh EB, Elmasri W, et al. A high response rate to liposomal doxorubicin is seen among women with BRCA mutations treated for recurrent epithelial ovarian cancer. Gynecologic oncology. Dec 2011;123(3):486-491.
75. Ekdawi SN, Stewart JM, Dunne M, et al. Spatial and temporal mapping of heterogeneity in liposome uptake and microvascular distribution in an orthotopic tumor xenograft model. Journal of controlled release : official journal of the Controlled Release Society. Apr 8 2015;207:101-111.
76. Uster PS, Working PK, Vaage J. Pegylated liposomal doxorubicin (DOXIL (R), CAELYX (R)) distribution in tumour models observed with confocal laser scanning microscopy. International journal of pharmaceutics. Mar 20 1998;162(1-2):77-86.
77. Unezaki S, Maruyama K, Hosoda J, et al. Direct measurement of the extravasation of polyethyleneglycol-coated liposomes into solid tumor tissue by in vivo fluorescence microscopy. International journal of pharmaceutics. Nov 22 1996;144(1):11-17.
Chapter 1: Introduction, Hypothesis and Overview 21
78. Yuan F, Leunig M, Huang SK, Berk DA, Papahadjopoulos D, Jain RK. Microvascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograft. Cancer Res. Jul 1 1994;54(13):3352-3356.
79. Cabral H, Matsumoto Y, Mizuno K, et al. Accumulation of sub-100 nm polymeric micelles in poorly permeable tumours depends on size. Nat Nanotechnol. Dec 2011;6(12):815-823.
80. Mikhail AS, Eetezadi S, Ekdawi SN, Stewart J, Allen C. Image-based analysis of the size- and time-dependent penetration of polymeric micelles in multicellular tumor spheroids and tumor xenografts. International journal of pharmaceutics. Apr 10 2014;464(1-2):168-177.
81. Lu Y, Park K. Polymeric micelles and alternative nanonized delivery vehicles for poorly soluble drugs. International journal of pharmaceutics. Aug 30 2013;453(1):198-214.
82. Mohamed S, Parayath NN, Taurin S, Greish K. Polymeric nano-micelles: versatile platform for targeted delivery in cancer. Therapeutic delivery. Oct 2014;5(10):1101-1121.
83. Torchilin VP. Micellar nanocarriers: Pharmaceutical perspectives. Pharmaceutical research. Jan 2007;24(1):1-16.
84. Mikhail AS, Allen C. Block copolymer micelles for delivery of cancer therapy: transport at the whole body, tissue and cellular levels. Journal of controlled release : official journal of the Controlled Release Society. Sep 15 2009;138(3):214-223.
85. van Driel WJ, Lok CA, Verwaal V, Sonke GS. The role of hyperthermic intraperitoneal intraoperative chemotherapy in ovarian cancer. Current treatment options in oncology. Apr 2015;16(4):329.
86. Eskander RN, Cripe J, Bristow RE. Intraperitoneal chemotherapy from Armstrong to HIPEC: challenges and promise. Current treatment options in oncology. Mar 2014;15(1):27-40.
87. Chiva LM, Gonzalez-Martin A. A critical appraisal of hyperthermic intraperitoneal chemotherapy (HIPEC) in the treatment of advanced and recurrent ovarian cancer. Gynecologic oncology. Jan 2015;136(1):130-135.
88. Mason KA, Valdecanas D, Hunter NR, Milas L. INO-1001, a novel inhibitor of poly(ADP-ribose) polymerase, enhances tumor response to doxorubicin. Invest New Drugs. Feb 2008;26(1):1-5.
89. Lee JM, Ledermann JA, Kohn EC. PARP Inhibitors for BRCA1/2 mutation-associated and BRCA-like malignancies. Ann Oncol. Jan 2014;25(1):32-40.
90. Zorn KK. PARP inhibition in epithelial ovarian cancer: high hopes undergo a reality check. Oncology. Feb 2012;26(2):128-136.
22
The challenges facing block copolymer micelles for cancer therapy: in vivo barriers and clinical translation
Sina Eetezadi, Sandra N. Ekdawi and Christine Allen
Reprint from Advanced Drug Delivery Reviews (October 2014)
DOI: 10.1016/j.addr.2014.10.001
Written by S. Eetezadi and S. N. Ekdawi. Figures by S. Eetezadi. Edited by
C. Allen.
The copyright of this article belongs to Elsevier B.V., the publisher of
Advanced Drug Delivery Reviews. Permission had to be requested for
publishing the article as part of this dissertation, which was obtained.
Chapter 2: Challenges facing Block Copolymer Micelles 23
2.1. Abstract
The application of block copolymer micelles (BCMs) in oncology has benefitted from
advances in polymer chemistry, drug formulation and delivery as well as in vitro and in vivo
biological models. While great strides have been made in each of these individual areas, there
remains some disappointment overall, citing, in particular, the absence of more BCM
formulations in clinical evaluation and practice. In this review, we aim to provide an overview
of the challenges presented by in vivo systems to the effective design and development of
BCMs. In particular, the barriers posed by systemic administration and tumor properties are
examined. The impact of critical features, such as the size, stability and functionalization of
BCMs is discussed, while key pre-clinical endpoints and models are critiqued. Given clinical
considerations, we present this work as a means to stimulate a renewed focus on the unique
chemical versatility bestowed by BCMs and a measured grasp of representative in vitro and
in vivo models.
Chapter 2: Challenges facing Block Copolymer Micelles 24
2.2. Introduction
The medical application of micelle-based nanotechnology dates back to the pioneering
work of Speiser at the ETH in Zurich with the development of drug delivery systems for
controlled release. In 1976, he explored the use of solidified micelles, termed ‘nanoparts’, and
put forth that “the partition of drugs in such nanoparts seemed to be promising as a new
parenteral drug delivery system for long-term therapy”.1,2 The principles of this seminal
publication indeed led to the first drug delivery application of block copolymer micelles
(BCMs) by Ringsdorf.3 Decades later, impactful contributions by Kataoka, Kabanov and others
have resulted in BCM formulations that have now reached late-stage clinical development.4-6
BCMs are nano-sized aggregates of amphiphilic copolymers with a size range of about 10
to 100 nm (Figure 2-1). They consist of a hydrophobic core that serves as loading space for
hydrophobic drugs and an outer shell, or corona, comprised of hydrophilic material that
provides a protective interface between the micelle core and external medium. In aqueous
media, at copolymer concentrations at or above the critical micelle concentration (CMC), self-
assembly results in micelles possessing greater thermodynamic and kinetic stability than that
achieved using small-molecule surfactants.7,8 Indeed, the copolymers can be tailored to result
in stable micelles that are optimized for tumor-selective delivery of therapeutic agents.9-11 By
formulating small-molecule chemotherapeutics in BCMs, their solubility can be enhanced
while their pharmacokinetic, as well as biodistribution profiles, can be favorably altered. Such
modulation of the in vivo distribution of small-molecule agents enables a reduction in often
dose-limiting normal tissue toxicities and can yield significant improvements in their
therapeutic index.12,13 Therefore, BCMs provide a functional platform for the design of nano-
sized drug delivery systems (NDDSs) to overcome the challenges faced by conventional
chemotherapeutics. The chemical diversity of monomers that form the copolymer building
Chapter 2: Challenges facing Block Copolymer Micelles 25
blocks offers synthetic versatility enabling customization at the molecular level, and control
of the physico-chemical properties of the BCMs (i.e. size, morphology, stability, and surface
properties) 14. The ease of chemical modification of the copolymers also allows for
optimization of drug loading (via physical encapsulation or conjugation) and release, as well
as surface functionalization with radionuclides and/or targeting moieties (Figure 2-1).15-17
Initially regarded as “pharmaceutical curiosities” 2, BCMs presently have the potential to offer
three key advantages over conventional formulation strategies: (1) increased solubility of the
encapsulated drug 18, (2) high adaptability of the physico-chemical properties of the BCM
system 11,19, and (3) improved biodistribution of drug and thereby reduced systemic toxicity.20
However, despite intense research activity on NDDSs such as BCMs, and subsequently, an
extensive number of publications generated on this topic over the past several decades
(Figure 2-2), clinical translation has proven challenging. In particular, the achievement of
significant improvements in efficacy, characterized by concomitant reductions in tumor
burden, disease recurrence and metastatic progression, remains an elusive goal.21-25 Today,
NDDS-based cancer therapy finds itself at a crossroads, challenging the unique promise and
ultimate clinical relevance of nanomedicines.22,23,26 Of note, its disputed state brings into
question the future of BCMs, nearly four decades following their emergence as a drug delivery
platform. In particular, it has become increasingly evident that the realization of substantial
clinical benefits requires clear elucidation of the biological complexity of the drug delivery
process and its use as a driving force to guide the development of future nanomedicines. The
major challenge remains overcoming the physiological and biophysical barriers imposed by
the host, tumor and host-tumor interactions, while integrating due recognition of the vast
extent of inter-patient and intra-tumoral heterogeneity.7,27 BCMs, in particular, stand out
among the advanced NDDSs owing to their potential versatility. Yet, their viability as a
Chapter 2: Challenges facing Block Copolymer Micelles 26
successful drug delivery platform relies on the design and implementation of novel
approaches that exploit and/or overcome pathophysiological mechanisms.
Figure 2-1: Versatility of BCM chemistry. BCMs provide a flexible platform for the design of
NDDSs given the synthetic versatility that enables customization at the molecular level.
Among a wide range of factors which influence key physico-chemical properties and in vivo
performance, we illustrate select parameters which have been varied as a means to control
size, stability, loading and labeling of BCMs.
Chapter 2: Challenges facing Block Copolymer Micelles 27
This review aims to provide a discussion of the attributes and shortcomings of BCM-
mediated cancer therapy, particularly within the context of biological barriers and, ultimately
clinical translation. The design of new and more effective cancer therapy strategies calls for
a comprehensive understanding of the underlying pathological mechanisms hindering and/or
helping the targeted delivery of BCMs to tumors, the factors driving the success of NDDSs, as
well as an awareness of the significant discrepancies observed between pre-clinical and
clinical outcomes. First, we aim to provide a review of the key in vivo barriers impeding the
effective delivery of chemotherapeutics via BCMs, and present strategies that have been
employed in an attempt to overcome these barriers. Finally, the goal of translating BCM
technology to the clinic is examined, paying due attention to significant hurdles to its
achievement; namely, the design, evaluation and interpretation of in vivo studies. Throughout
this review, concepts are illustrated using examples obtained from literature on BCMs.
However, in some cases, studies on other NDDSs (e.g. liposomes) are highlighted as a means
to enhance our general understanding of the factors that influence the in vivo behavior and
performance of these systems.
Chapter 2: Challenges facing Block Copolymer Micelles 28
2.3. Biology vs. Block Copolymer Micelles
Effective drug delivery remains an outstanding feat to be achieved, encompassing
challenges such as NDDS instability, limited tumor accumulation and/or penetration. Further,
a third challenge consists of enhancing tumor drug bioavailability, which we describe here as
the amount of drug available to tumor cells with the potential to elicit a cytotoxic effect. In
this section, our goal is to establish the key in vivo barriers to the effective transport of
therapeutics via BCM-mediated delivery, with due appreciation of the tumor
microenvironment (Figure 2-3). Specifically, both the exploitable and resistant mechanisms
proper to solid tumors will be examined relative to existing nanomedicine formulations and
their design. With a creative grasp of the positive complexity conferred through BCM
properties, innovation may arise as we gain a firmer understanding of the negative complexity
afflicting tumors. Specifically, we postulate that the complexity present in designing BCMs
can be exploited to produce more effective formulations, while that present in tumors
exposes the biophysical barriers and ensuing limitations to effective drug delivery.
2.3.1. First line of defense: the blood compartment
The introduction of BCMs to the dynamic blood compartment typically engenders a host
of endogenous obstacles before reaching their destination. A distinguishing feature of NDDSs
generally is their ability to deliver a larger and more sustainable payload of drugs, relative to
conventional drug administration.28-30, leading to increased tumor accumulation and reduced
systemic toxicities. However, BCMs, like many NDDSs, are susceptible to increased clearance
as a result of opsonization and uptake by the mononuclear phagocyte system (MPS), as well
as variations in size and stability of the construct. Nevertheless, the ability to control BCM
properties for specific applications is worthy of efforts to design long-circulating and effective
drug release systems.
Chapter 2: Challenges facing Block Copolymer Micelles 29
2.3.1.1. Physico-chemical properties of BCMs: impact on their pharmacokinetics and biodistribution
The ability of BCMs to accumulate at sites of tumor growth is owed to the enhanced
permeability and retention (EPR) effect resulting from the characteristically permeable blood
vessels and defective lymphatic drainage system of tumors.31 This phenomenon, recognized
over the past three decades as the main conduit for NDDS-mediated drug delivery, favors the
enhanced tumor deposition of therapeutic agents, shuttled by NDDSs measuring above the 5
nm threshold for renal filtration.29,30,32 Indeed, it is known that increasing NDDS size can lead
to prolonged residence time in the blood.7,33 Importantly, a balance must be struck between a
sufficiently large NDDS diameter that can extend circulation time of the drug-carrier system,
and a permissively small construct that can penetrate dense tumor tissue. NDDSs bearing a
hydrodynamic diameter above 200 nm have been reported to undergo faster clearance from
the blood circulation than their smaller counterparts.34 Further, the potential for greater serum
protein adsorption increases with increasing NDDS size 33 which may also reduce blood
circulation times via enhanced opsonization and deposition within the liver and spleen.35,36
The use of poly(ethylene glycol) (PEG) as the hydrophilic shell, or corona-forming block of the
copolymers, has provided a means to evade serum protein-mediated clearance or
opsonization. In general, this technique promotes the evasion of plasma proteins (i.e.
opsonins) which bind to NDDSs and can signal uptake by phagocytic cells as well as
sequestration of the entrapped drug within organs of the MPS.37,38 Early studies by Kwon et
al. using doxorubicin-conjugated PEG-b-poly(aspartate) copolymers for a BCM formulation
revealed that high molecular weight PEG blocks of 12000 in combination with low molecular
weight poly(aspartate) blocks showed increased blood stability, in comparison to lower
molecular weight PEG blocks of 1000 or 5000 and demonstrated the ability of PEG to shield
the drug-loaded BCM core from biological components.39 While PEG remains the most
Chapter 2: Challenges facing Block Copolymer Micelles 30
commonly used hydrophilic polymer building block there are a number of limitations
associated with this material.40,41 As a result there has been investigation into the use of other
hydrophilic polymers.42,43
It is recognized that BCM properties such as stability, size, size distribution, morphology
and surface properties impact their pharmacokinetic profile and tissue distribution.13,14
However, the lack of standardized methods of evaluation weakens our ability to predict the
corresponding change in in vivo behavior as a function of variation in a given BCM property.7,44
Nevertheless, continued investigation into the in vivo fate of BCMs provides invaluable insight
into the impact of certain properties, and their potential modulation of biodistribution. For
instance, a slightly negative charge (-10.6 mV) on the surface of 30 nm PEG-b-poly(D,L-lactide)
(PEG-b-PDLLA) copolymer micelles has been shown to reduce non-specific uptake by the liver
and spleen relative to uncharged micelles, and consequently favors prolonged residence
within the blood compartment.45
Important consideration of the size and surface chemistry of BCMs is also reflected in their
ability to effectively target tumors. While the issue of tumor accumulation and extravasation
will be examined in more depth in section 2.3.2, we highlight here more recent examples of
the relationship between transport within the blood compartment and accumulation at the
tumor site.46-49 Indeed, the physico-chemical properties of the designed BCM system will likely
dictate its ability to withstand the barriers associated with blood-borne transport and
ultimately, exert its intended therapeutic effect at the tumor lesion. The impact of varying
copolymer molecular weight and/or molecular targeting of BCMs on transport at the whole
body level have been examined, along with their associated effects on tissue and cellular
uptake. Lee et al. have shown that the distinct BCM sizes arising from varying the block
lengths of PEG and poly(caprolactone) (PCL) (i.e. PEG2000-b-PCL2000 or PEG5000-b-PCL5000) pose
Chapter 2: Challenges facing Block Copolymer Micelles 31
a great influence on the plasma clearance of BCMs. Indeed, BCMs measuring 25 nm in
diameter exhibited a circulation half-life of only 13 h relative to one of 29 h for BCMs with a
diameter of 60 nm. In contrast, surface functionalization of the BCMs with epidermal growth
factor (EGF) had no impact on the pharmacokinetics (PK) of the BCMs of either size, or on
the clearance of the formulations via the MPS.46 Similarly, Fonge et al. investigated the impact
of multiple BCM formulation variables; namely, BCM size (i.e. diameter of 60 nm vs. 15 nm),
surface density of EGF and nature of the bifunctional chelator used for chelation of Indium-
111 (111In) – all of which impacted the PK and biodistribution of the various formulations.47
While it is recognized that optimal densities of surface ligands are likely to be a function of
both the ligand and the NDDS, the authors showed that inclusion of 5 mol % EGF was
detrimental to the systemic circulation of BCMs, while 0.2 mol % EGF resulted in a PK profile
that was undistinguishable from that of the non-targeted BCMs.46,50 It was postulated that the
25-fold greater molar surface density of EGF promoted MPS-mediated opsonization of the
BCMs and deposition in the EGF receptor (EGFR)-rich environment of the liver. The PK of the
15 nm BCMs, on the other hand, was largely controlled by their small size, with all other
variables resulting in a negligible effect.
An additional parameter capable of modulating circulation lifetime and/or biodistribution
lies in the morphology of the nanoconstructs. Geng et al. have demonstrated that BCMs
bearing a filamentous shape, termed ‘filomicelles’, are capable of circulating for up to 1 week
following systemic administration relative to the 2-day vascular residence time of their
spherical counterparts 51. The rod-like shape of filomicelles is believed to facilitate fluid flow
in the blood and promote evasion of uptake by phagocytic cells. Discher’s group later followed
up this work by examining the biodistribution and tumor accumulation of filomicelles labeled
with a near-infrared fluorophore. In this study, the authors confirmed delayed MPS-mediated
Chapter 2: Challenges facing Block Copolymer Micelles 32
clearance of the filomicelles and further observed greater tumor selectivity with the
filamentous shape relative to spherical micelles identical in composition. Not only were the
filomicelles able to circumvent phagocytosis by virtue of their shape as shown previously,51
but they were also found to permeate tumor tissue more readily than healthy tissue, thus
elucidating their tumor selectivity.52
Therefore, the size, morphology and surface functionalization of BCMs remain important
factors in optimizing the delivery and transport of chemotherapeutic agents to solid tumors.
2.3.1.2. In vivo drug retention within BCMs
The stability of BCMs holds importance for the purpose of formulating toxic and/or
unstable therapeutics into NDDSs for enhanced tumor accumulation with reduced systemic
exposure. This goal is compromised if the vehicle does not serve its purpose as a long-
circulating, stable entity but rather as that of a drug solubilizer. In the latter case, the
formulated drug dissociates rapidly from the carrier and exhibits a pharmacokinetic profile
similar to that of the free drug, negating the role of the carrier as a tumor-targeted delivery
system. Nevertheless, an important distinction between the two systems pertains to the
ability to effectively deliver and release the drug at the target site; indeed, a balance in carrier
design is required such that the drug-loaded BCM system results in favorable PK and
biodistribution profiles, yet adequate drug release at the tumor. The high protein content in
the blood can disrupt micelle stability by mechanisms of drug extraction from the core as well
as through protein adsorption onto the corona.8,53 The aforementioned issues are exemplified
by comparison of human clinical data obtained from two different formulations of the drug
doxorubicin (DOX), namely SP1049C (comprised of Pluronic® copolymer micelles)54 and
NK911 (comprised of PEG-b-poly(aspartate) copolymer micelles).55 SP1049C, which
physically incorporates DOX, resulted in a drug clearance rate of 12.6 mL/min/kg in humans,
Chapter 2: Challenges facing Block Copolymer Micelles 33
which was similar to that obtained for the free drug (i.e. 14.4 mL/min/kg). Consequently, this
yielded only a slight increase of 12.5 % in the plasma AUC of the drug (i.e. from 1.6 µg/h/mL
to 1.8 µg/h/mL). On the other hand, in the NK911 BCM formulation, a fraction of the drug is
chemically conjugated to the core-forming block of the copolymer to enhance loading and
retention of further physically entrapped drug.56 This strategy enabled the formulation to act
as a true carrier resulting in a reduced drug clearance rate (6.7 mL/min/kg) and more than a
2-fold increase in the drug plasma AUC (3.3 µg/h/mL) in comparison to free drug, following
administration in humans.57
In another study examining the drug paclitaxel (PTX), Burt’s group demonstrated the rapid
dissociation of the hydrophobic drug from the poly(lactide) (PLA) core of amphiphilic BCMs.58
In this case, the dissociation may have resulted due to the lack of preferential interaction
between PTX and the core-forming copolymer. Indeed, a more recently designed formulation
of PTX, known as NK105, has been shown to act as a true carrier due to derivatization of the
core-forming poly(aspartate) block with 4-phenyl-1-butanol. The resulting copolymer
promotes hydrophobic interactions and retention of the physically entrapped drug.59 These
observations underline the importance of polymer-drug interaction, or compatibility.14
Bronich’s group recently proposed a core cross-linked micelle platform for combined
delivery of cisplatin (CDDP) and PTX in a xenograft model of ovarian cancer 60. In this study,
the authors recognized and exploited the capacity of proteases to selectively degrade the
polypeptide chains of multi-compartment, triblock PEG-b-PGlu-b-PPhe cross-linked (cl)
micelles resulting in effective release of their drug cargo. The physical stability of cl-micelles
was found to be superior to that of non-cross-linked micelles and proteolytic degradation,
monitored in vitro using cathepsin B, promoted drug release from the cl-micelles. Indeed, a
significantly greater anti-tumor effect was observed following treatment with the combined
Chapter 2: Challenges facing Block Copolymer Micelles 34
(CDDP+PTX)/cl-micelles due to their systemic stability, selective drug release and synergistic
therapeutic effect. The optimal drug ratio of CDDP to PTX required to achieve a synergistic
effect in vitro was loaded into the cl-micelles, retained in the NDDS and found to be maintained
at the tumor site. This enabled superior tumor inhibition and survival relative to chemotherapy
administered in free form or via BCMs as single drug formulations.
2.3.2. Second line of defense: tumor extravasation and accumulation
The degree of physico-chemical control in designing BCM systems is challenged
following their introduction into the bloodstream. However, while the challenges imparted by
protein adsorption, opsonization, and clearance mechanisms can be fairly well predicted
given the constancy of systemic response (cf. section 2.3.1), the remaining in vivo barriers
constitute a less predictable threat to NDDS-mediated drug delivery. In particular, spatio-
temporal heterogeneity in tumor structure and function impedes the transport of NDDSs
and/or their associated drug cargo throughout the tumor.
2.3.2.1. The EPR effect
Neoplastic growth may provoke the recruitment of existing blood vessels that supply
neighboring tissue.61 However, tumor progression causes an increased demand for nutrients
which exceed the supply provided by the existing blood vessels, and therefore,
neovascularization is adopted as a strategy to meet the malignant regimen.62 The extreme
production of angiogenic stimulators leading to the creation of a disorganized and tortuous
blood vessel architecture 63 also hinders the ability to reproduce the branching hierarchy of
arterioles, capillaries and venules found in normal tissues. This elicits differences in the
degree of vascularization of the tumor and thereby gives rise to hypoxic regions resulting from
irregular blood perfusion.64,65 Furthermore, abnormal maturation of the vessels,66 reflected in
a discontinuous basement membrane, defective endothelial monolayer and pericyte
Chapter 2: Challenges facing Block Copolymer Micelles 35
coverage 67,68 leads to hyperpermeability brought on by widened vessel fenestrations (i.e. 100
times greater than in normal vessels).69 Overall, the poorly regulated and inherently
heterogeneous vasculature found within different tumor types,70 as well as within the same
tumor type,70,71 inevitably leads to the heterogeneous intratumoral distribution of
administered therapeutics.72,73 The lack of a functional lymphatic system in tumors also
contributes to the limited distribution of macromolecular and small-molecule
chemotherapeutics due to increased interstitial fluid pressure (IFP).73 The impaired
mechanism of tissue clearance effectively facilitates the retention of these agents within the
tumor. Thus, the combined ability for enhanced extravasation and retention of larger
macromolecules in particular,74 is loosely defined as the EPR effect. As will be treated in
further detail in section 2.4, the status and exploitation of the EPR effect continues to be
explored as pre-clinical models aim to integrate the growing insight gained from clinical
data.75
2.3.2.2. Chemical offense: adapting BCM properties for enhanced tumor uptake
In 1994, Kwon et al. reported the first example of enhanced tumor accumulation of DOX
due to administration in BCMs via exploitation of the EPR effect. Indeed, the authors found
that the extended circulation lifetime achieved with the drug-conjugated BCMs resulted in
greater tumor accumulation of DOX relative to administration of free drug.76 Since then, it has
been found that the rate and extent of extravasation or transvascular transport of NDDSs
from the bloodstream into tumor tissue is not only dependent on their circulation lifetime, but
also on their tumor vascular permeability.74 Overall an increase in the size of the
macromolecule or NDDS can decrease tumor vascular permeability, but increase the blood
residence time.77 The increase in blood residence time can, in turn, provide more time for
extravasation; however, in some cases this only partially compensates for the decrease in
Chapter 2: Challenges facing Block Copolymer Micelles 36
tumor vascular permeability.74 Therefore, an optimal size or hydrodynamic diameter must be
found that enables passive accumulation at the tumor through strategic exploitation of the
EPR effect. Indeed, a stimulating discussion on the role of the EPR effect has been raised in
a recent review by Nichols and Bae, also discussed further in section 2.4. Importantly, the
authors convincingly highlight the heterogeneity associated with the EPR effect in pre-clinical
models.78 Nonetheless, assessment of an EPR effect or tumor accumulation provides a
means of comparing the relative performance of NDDSs, particularly when conducting head-
to-head evaluations of NDDS properties (e.g. size, targeting). As introduced previously, Lee et
al. evaluated the distribution of 25 nm and 60 nm BCMs in xenograft models of human breast
cancer. In accordance with their shorter vascular residence time, the smaller BCMs also
exhibited lower tumor deposition 48 hours post-administration relative to their larger
counterparts.50 Interestingly, the conjugation of EGF compensated for the significantly shorter
half-life of the 25 nm BCMs by resulting in comparable tumor uptake levels to the non-
targeted 60 nm BCMs in the EGFR-overexpressing MDA-MB-468 xenograft model at 48 hours
post-administration. As such, successful tumor retention via active targeting may present an
attractive strategy for enhancing the tumor accumulation of BCMs without the need for size
amplification.50 Despite being commonly sought out as a strategy to enhance the tumor
accumulation and/or intracellular uptake of the shuttled chemotherapeutic, receptor-
mediated, or frequently termed ‘active’, targeting remains a highly investigated, and at times
divisive, approach.22,24,26,79 Due to the reporting of both possible outcomes as a result of
receptor-mediated targeting (i.e. enhanced tumor accumulation and/or cell uptake), there is
much debate surrounding the benefit of functionalizing NDDSs with targeting ligands.80-82
While some studies have shown no significant difference in tumor accumulation between
targeted and non-targeted NDDSs,80,83 others have achieved greater tumor accumulation
from the addition of a targeting moiety.84 A number of factors have emerged that can
Chapter 2: Challenges facing Block Copolymer Micelles 37
potentially influence the outcome of receptor-mediated targeting. Selection of a target
receptor effectively favors one that is over- and/or selectively expressed in cancerous
tissue.85 While the targeting ligand must possess a sufficiently high affinity for the target
receptor and be capable of inducing receptor-mediated endocytosis.86 However, as
mentioned previously, surface functionalization of NDDSs can have important repercussions
on their circulation lifetime, and consequently, total tumor uptake. As such, it is imperative
that the addition of a targeting ligand does not lead to a reduction in the circulation lifetime of
the NDDS and its entrapped therapeutic cargo.87 In addition, the expression of the target
antigen plays a key role in determining the success of the receptor-mediated targeting
strategy. In particular, the accessibility and distribution of the target antigen within the tumor
microenvironment has been found to have a significant impact on the anti-tumor efficacy of
targeted therapeutics. This has been clearly highlighted through the use of antibodies for
radioimmunotherapy.88,89 For example, El Emir et al. evaluated the tumor uptake and
corresponding therapeutic efficacy of the anti-carcinoembryonic antigen (CEA) antibody,
A5B7, in two contrasting tumor xenograft models of human colorectal cancer, LS174T and
SW1222. This study aimed to represent two different clinical conditions; indeed, one model in
which the tissue exhibited moderate-to-poor differentiation and another model which was
well differentiated and bore a greater resemblance to healthy colon. Despite equivalent levels
of tumor accumulation of the antibody attained in both models, vastly different responses to
radiotherapy were observed following intravenous (i.v.) administration of the 131I-labeled
antibody,131I-A5B7. Indeed, the LS174T tumor model exhibiting poorer tissue differentiation,
reflected in a heterogeneous blood vessel architecture and distribution, required a 3-fold
increase in 131I-A5B7 dose to match the tumor growth inhibition achieved in the second, better
differentiated tumor model. Fluorescence microscopy of whole tumor sections revealed
perivascular retention of A5B7 in the LS174T model principally brought on by the limited
Chapter 2: Challenges facing Block Copolymer Micelles 38
spatial accessibility of the CEA antigen from an inefficient vascular network. Interestingly, the
authors showed that both tumor models displayed a similar degree of vessel permeability
and antigen expression, while no significant hindrance was imparted from IFP or intercellular
gap junctions. This study highlighted several influencing microenvironmental parameters on
the therapeutic efficacy of targeted therapies, demonstrating that the spatial distribution of a
target antigen in relation to that of functional blood vessels plays a key role in ensuring that
most targets, and the tumor itself, are hit.88 It should be noted that the physico-chemical
properties of the NDDS, in particular its size, may also influence the targeting efficiency of the
system as a whole. As such, there is great value in evaluating the targeting capability of a
given ligand-receptor pair as a function of NDDS size. Still, studies investigating the same
ligand-receptor pair are growing, as evidenced by several examples employing the NGR ligand
further discussed below.
While some tumor or stromal cell antigens may be relatively more abundant and/or
accessible than others, vascular targeting as a whole may be perceived as distinct from its
counterpart tumor cell targeting by means of relatively more uniform spatial accessibility.
Simnick et al. have demonstrated the enhanced targeting capability of elastin-like polypeptide
block copolymer micelles modified with the tumor vasculature-specific NGR targeting
ligand.90 Intravital laser scanning confocal fluorescence microscopy (ILSCM) and
immunofluorescence (IF) revealed the preferential in vivo tumor localization of NGR-modified
micelles within regions populated with endothelial and perivascular cells. Again, extravasation
was similar for both targeted and non-targeted micelles, however, the pattern of accumulation
seemed to differ, favoring the retention of NGR-micelles in the vascular compartment, as well
as the extravascular space, relative to the non-targeted micelles. Yet, Simnick et al. did
highlight the need for further optimization of their targeting system, citing the requisite
Chapter 2: Challenges facing Block Copolymer Micelles 39
characterization of the density and clustering of the target receptor present on the tumor
vasculature and accordingly, the design of appropriate targeting systems and tumor
models.90 Kataoka’s group also recently examined the in vivo performance of cyclic RGD
(cRGD)-linked BCMs incorporating (1,2-diaminocyclohexane)platinum(II) (DACHPt) as a
therapeutic strategy for treating glioblastoma.91 cRGD-linked (20% surface density) DACHPt-
incorporated BCMs resulted in significant tumor growth inhibition relative to free drug or
DACHPt administered in non-targeted BCMs or micelles functionalized with the cRAD
mismatched peptide. Strikingly, both inductively coupled plasma mass spectrometry (ICP-
MS) and ILSCM confirmed that tumor uptake was more than 2-fold greater with the cRGD-
BCMs as early as 4 hours post-injection (hpi) relative to the cRAD-BCMs, and continued to
increase up to 24 hpi.91 The fact that both the cRGD- and cRAD-targeted BCMs were the same
size (i.e. approximately 30 nm) suggests that a phenomenon distinct from, or additive to, the
EPR effect may be at play, enabling the specific binding and uptake of cRGD-targeted BCMs
in angiogenic tumors overexpressing integrin receptors. In fact, contrary to the cRGD-BCMs,
the mismatched peptide-functionalized, cRAD-BCMs, did not extravasate from blood vessels
or penetrate into tumor tissue as revealed by quantitative permeation profiles. According to
the authors, this receptor-mediated targeting strategy effectively accelerated tumor uptake
and penetration by inducing “active transport-mediated pathways”, such as transcytosis,
which provided a means of passage across the “vascular barrier”, by virtue of cRGD binding.91
Therefore, this study contributes an additional strategy wherein the end goal of receptor-
mediated targeting is not necessarily to promote cellular internalization of the drug but rather
to aid in accessing more of the tumor (i.e. for greater tumor penetration). Such alternative
targeting strategies also extend the margins imposed by the EPR effect, and support
analogous studies evaluating the tumor targeting capability of vasculature-targeted NDDSs.
For instance, a study by Dunne et al. employed NGR-targeted liposomes and demonstrated
Chapter 2: Challenges facing Block Copolymer Micelles 40
the ability to circumvent the limitations associated with passive targeting of NDDSs by
homing to the tumor endothelium.87 The authors argue that “active targeting to tumor cells in
the extravascular compartment relies on passive targeting as a prerequisite”, and therefore,
targeting NDDSs to tumor vascular receptors presents a potential means to gain some
autonomy amid the insecurity of relying on a pronounced, EPR-dependent enhancement of
tumor accumulation.
Figure 2-2: Number of annual publications for select NDDSs. Despite the exponential growth
in research activity over the last several decades, very few NDDSs have achieved translation
from “bench to bedside”. Data show the total number of publications compiled as of March
2014 using Scopus® search engine with search terms "micelle", "liposome" or "polymer-drug
conjugate” paired separately with "drug delivery”.
Chapter 2: Challenges facing Block Copolymer Micelles 41
2.3.2.3. Biological retaliation: exploiting tumor properties for enhanced BCM uptake
Although the EPR effect has been clearly shown to enable accumulation of
macromolecules and NDDSs in solid tumors, it is important to note that the extent of this
effect is also largely dependent on the physiological properties of the vasculature at the tumor
site 92,93. Notably, vascular permeability, perfusion and density are not only abnormal in
tumors but also present both spatial and temporal heterogeneity in the extent to which they
supply the lesion.26 For instance, while vascular permeability will vary as a function of NDDS
size,27,74 it is also highly influenced by the site of tumor growth, ‘normalizing’ treatment 94 (e.g.
anti-angiogenic therapy) and experimental design used for measurement of this parameter
(e.g. dextrans vs. proteins74). Vascular perfusion is compromised when an excess of fluid
escapes from the vascular compartment due to vessel hyperpermeability, when fluid flow is
impaired as a result of vessel tortuosity, and/or due to vessel compression.94,95 Vascular
density can suffer when the uncontrolled proliferation of tumor cells, forces vessels apart and
increases inter-vessel distances. Nutrient supply is then limited within these poorly
vascularized regions of the tumor.96 As mentioned previously, the degree of vascularization
and endothelial permeability have been known to vary between tumor types, between blood
vessels perfusing the same tumor and within the same tumor “during tumor growth,
regression and relapse”.37,97 Given the critical role the vascular conduit plays in delivering
nanomedicines to solid tumors, the characterization and modulation of tumor vascular
properties presents an essential aspect in enhancing NDDS-based drug delivery.
A sobering clinical example of the impact of tumor properties on the accumulation of
NDDSs is found in the key study by Harrington et al. in which the authors examined the PK
and biodistribution of 111In-labeled PEGylated liposomes, a formulation equivalent to the
clinically employed therapeutic Doxil®. The study was performed in 17 patients with locally
advanced cancers (i.e. breast, head and neck, lung, cervical and glioma),98 whereby region of
Chapter 2: Challenges facing Block Copolymer Micelles 42
interest (ROI) analysis of gamma camera images of those patients revealed a broad level of
liposome uptake at the tumor, ranging from 2.7 to 53 % ID/kg tumor (i.e. percentage injected
dose per kg tumor). Liposome uptake was found to be greatest in head and neck tumors (33.0
± 15.8 % ID/kg), with intermediate levels found in lung tumors (18.3 ± 5.7 % ID/kg) and finally,
the lowest levels were recorded in breast tumors (5.3 ± 2.6 % ID/kg). These results were, in
part, attributed to the significant variability present in the structural and functional integrity of
the tumor vasculature across these tumor types. As such, others have found a direct
correlation between microvessel density (MVD) and tumor accumulation of liposomes,99 and
recently, with BCMs.48
The function of tumor blood vessels has garnered much attention as a source of both peril
and potential in NDDS-based drug delivery. Vascular normalization has been sought after as
a means to improve the delivery of chemotherapy through functional and structural maturing
of the tumor blood vessels, which would in turn mitigate the extent of elevated IFP which
presents a significant barrier to the transport of NDDSs.94 The paradox of enhancing NDDS-
based drug delivery by reducing the size of vessel fenestrations was elucidated in a study by
Chauhan et al. whereby the tumor distribution of two clinically approved nanomedicines,
Doxil® and Abraxane®, was evaluated in mammary tumors.100 Following the pharmacologic
induction of vascular normalization by anti-VEGFR2 antibody DC101, it was found that the 10-
nm nanomedicine, Abraxane®, was able to better permeate tumors relative to Doxil®, and thus
this vascular modulation strategy was deemed favorable for nanomedicines of 10 nm or less
in size. Conversely, vascular permeabilization, or “abnormalization”, has been shown to
augment the permeability of tumors to NDDSs by increasing the number of permeable
vessels. This strategy was effectively explored using low-dose TNF-ɑ as a vasoactive
modulator and resulted in a 5 - to 6-fold increase in tumor accumulation of Doxil®.101 Physical
Chapter 2: Challenges facing Block Copolymer Micelles 43
means have also been investigated for permeabilization of tumor vasculature. As reported by
Lammers and colleagues, hyperthermia and radiotherapy have been applied in conjunction
with poly(hydroxypropyl methacrylate) (pHPMA) copolymer-based drug delivery systems as
a means to promote vasoactivity and overall tumor accumulation. Specifically, hyperthermia
was found to result in an increase in pHPMA copolymer tumor accumulation, albeit in one out
of the three tumor models evaluated.102 This report highlights the variable impact
hyperthermia can impart as a function of tumor type and experimental conditions.103-105 On
the other hand, the administration of radiotherapy prior to that of copolymers of different
molecular weights was found to yield an increase in tumor accumulation in all tumor models
under investigation.102 This result was believed to be due to production of the permeabilizing
factors vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF)
following radiation treatment,106-108 as well as a decrease in IFP 109 and also cellular depletion
from the tumor 110 and vascular wall.111,112.
2.3.3. Third line of defense, and the Achilles’ heel of nanomedicines: tumor penetration and tumor drug bioavailability
The poor penetration of NDDSs and their limited ability to deliver drugs to their sites of
action within tumors remain significant challenges obstructing their successful clinical
translation 49. Despite the often unyielding features of solid tumors, a better understanding of
the influential properties of NDDSs, as well as of tumors themselves, has emerged. Indeed,
complementary studies revealing the impact of the physico-chemical properties of NDDSs,
and that of tumor properties and/or their manipulation, contribute to our mission to improve
NDDS-based drug delivery by adding incremental gains to our body of knowledge.
Chapter 2: Challenges facing Block Copolymer Micelles 44
2.3.3.1. Factors affecting tumor penetration
It is known that therapeutic agents are transported from the vasculature to the tumor and
through the tumor interstitium by diffusion according to a concentration gradient and/or by
convection according to hydrostatic and colloid osmotic pressure gradients. The contribution
of diffusion and/or convection to transport is dependent on the molecular weight of the agent
as well as the status of transvascular and intratumoral gradients.27 Following NDDS
extravasation from the bloodstream to the tumor interstitium, they must penetrate the tissue
to reach cancer cells.27 The successful delivery and homogeneous distribution of NDDSs
within tumors is particularly challenging due to both the abnormal blood flow and interstitial
hypertension (i.e. high IFP). Blood vessels are known to be ‘unevenly distributed’ throughout
tumors which gives rise to well, poorly and even un-perfused regions.27 As well, given that
lymphatic drainage is compromised at tumor sites, due to compression of lymphatic vessels
by cancer cells, and leakage of excess fluid from the tumor vasculature, the IFP is increased,
especially at the tumor core.27 The osmotic and hydrostatic pressures in normal human
capillaries are reported to be about 28 and 20 mm Hg, respectively. The interstitial fluid
osmotic and hydrostatic pressures are said to be approximately 8 and -1 to -3 mm Hg,
respectively, in normal tissues.113 This results in the net movement of fluid from the capillaries
into normal tissue and throughout the interstitium. In contrast, in tumor tissue, the interstitial
fluid osmotic and hydrostatic pressures are known to increase and depending on their values
this can either result in the net movement of fluid into the tumor (as is the case for normal
tissues) or the movement of fluid out of the tumor.113 The IFP in tumor tissue is at a minimum
at the tumor periphery and reaches levels as high as 10 - 60 mmHg at the core of human
tumors.114 For this reason, the convective transport of NDDSs in solid tumors is believed to
be largely limited to the tumor periphery where peritumoral lymphatics are known to reduce
the IFP.100 Throughout the remainder of the tumor volume it is believed that diffusion serves
Chapter 2: Challenges facing Block Copolymer Micelles 45
as the primary mechanism of transvascular and interstitial transport owing to high IFP levels
that can approach or exceed the capillary pressure.115
Insight into the transport of macromolecules and NDDSs has continued to expand, as
studies evaluate their distribution at increasing resolutions in vivo as a function of size 9,74
and/or receptor-mediated targeting,48,50 as well as time and tumor properties.49 Recently, the
spatio-temporal distribution of BCMs of different sizes (i.e. 15 and 55 nm) was examined in
multicellular tumor spheroids (MCTS) as well as their corresponding tumor xenografts in vivo.
In particular, micelle penetration was quantitatively assessed in HeLa and HT29 tumor
models and was found to vary as a function of time, micelle size and tumor tissue porosity.
Mikhail et al. found that smaller, 15 nm BCMs were cleared more quickly from tumors relative
to their larger counterparts. MCTS-based penetration, representative of tumor extravasation
and interstitial transport, revealed a homogenous distribution of the 15 nm BCMs as early as
1 h post-incubation, with the larger micelles only achieving a similar spread by 24 h post-
incubation. Interestingly, greater micelle accumulation was observed in HeLa tissue, both in
vitro and in vivo, owing to the lower cell density and greater extracellular volume relative to the
tissue architecture found in HT29. This finding was supported by similar reports of greater
macromolecule penetration as a result of low interstitial cellular volume or resulting depletion
from chemical treatment.116 As such, manipulation of tumor components holds promise in
enhancing the extravasation, accumulation and penetration of NDDSs.
The modulation of the tumor vascular network has been shown to bear implications on
anti-tumor efficacy, beyond the level of accumulation. Kataoka’s group recently evaluated the
tumor microdistribution and therapeutic effect of fluorescently-labeled BCMs of different
diameters and containing the anti-cancer agent 1,2-diaminocyclohexane-platinum (II)
(DACHPt) in the human pancreatic adenocarcinoma BxPC3 xenograft model in mice.9 At 24
Chapter 2: Challenges facing Block Copolymer Micelles 46
hours post-administration, the 30 nm BCMs were shown to have penetrated into the dense
BxPC3 tumor network while the 70 nm BCMs remained close to blood vessels. The 30 nm
BCMs containing DACHPt were found to completely suppress tumor growth and the 70 nm
BCMs showed no anti-tumor effect. Conversely, an increase in permeability through
administration of a low-dose transforming growth factor (TGF)-β inhibitor rendered the
BxPC3 tumors more conducive to the penetration of the larger 70 nm micelles, and thus
further enhancing their efficacy. As performed in this study, several reports have employed
quantitative analysis for determining the tumor penetration of BCMs,50 antibodies,88,117 and
drugs.116 Nevertheless, further analysis is warranted in investigating the impact of select
modulators on the intratumoral transport of NDDSs. For instance, the tumor penetration and
extravascular accumulation of BCMs as a function of active targeting has, to the best of our
knowledge, only been quantified in two studies.50,90 Introduced previously, Lee et al.
investigated the in vivo fate of EGF-conjugated (i.e. actively targeted, T-BCMs) and non-
targeted BCMs (NT-BCMs) in the EGFR overexpressing MDA-MB 468 human breast tumor
model in mice at 48 hpi. BCMs were prepared from methoxyPEG-b-PCL copolymers, labeled
with Alexa-Fluor 647, and measured either 25 or 60 nm in diameter.50 Confocal fluorescence
microscopy combined with MATLAB image analysis of tumor sections revealed that the NT-
BCMs of 25 nm in diameter diffused further away from blood vessels (mean localization
distance from nearest blood vessel, Dmean = 42 9 µm) following extravasation, relative to the
60 nm NT-BCMs (Dmean = 23 4 µm). However, the introduction of molecular targeting
imposed a “binding site barrier” effect which retarded the tumor penetration of the T-BCMs of
25 nm in diameter (Dmean = 29 7 µm), while it showed no impact on the penetration of the
already size-restricted 60 nm BCMs (Dmean = 21 9 µm). Nevertheless, active targeting of the
Chapter 2: Challenges facing Block Copolymer Micelles 47
25 nm BCMs was found to result in a significant increase in both cellular and nuclear uptake
of the BCMs in vivo.
The goal of achieving improved tumor penetration of nanomedicines is a valid one, given
the compelling evidence linking limited tumor penetration to poor therapeutic efficacy.73,74 Yet,
cellular uptake has also been found to play a critical role in achieving better efficacy with
nanomedicines. The observation that increased cellular internalization could outweigh the
benefits of enhanced tumor accumulation was first noted by Kirpotin et al..83 Can we assert
the same for tumor penetration? Evidently, a number of answers present themselves – albeit
not necessarily solutions. For instance, the advanced drug delivery system in question may
be capable of enhanced interstitial or transcellular transport, but may not optimally release its
drug cargo at a therapeutic dose at the right time. As well, enhanced transport properties of
NDDSs do not preclude their ability to similarly populate healthy tissues. On the other hand,
delivery systems designed to target a specific receptor facilitate selective uptake of potent
chemotherapeutics, but as mentioned previously, can be limited by the “binding site barrier”.
As reported by El Emir et al., “binding site barrier” may constitute either an advantage or an
impediment depending on the pattern of antigen expression.88 As such, a spatio-temporal
characterization of the target antigen’s expression in relevant tumor models would likely be
valuable in optimizing targeted NDDSs. Further understanding of tumor and vascular
permeability, as well as tumor cell sensitivity to therapy as a function of the immediate
microenvironment within the tumor would aid in the selection and delivery of the appropriate
therapy.
2.3.3.2. Tumor Drug Bioavailability: improving site-specific drug exposure
While poor tumor drug bioavailability due to limited drug release at the tumor has plagued
some liposome-based formulations,118 it is less of an issue for BCMs. As mentioned
Chapter 2: Challenges facing Block Copolymer Micelles 48
previously in section 2.3.1.2, drug retention in BCMs constitutes a major challenge impeding
their success. However, the delivery of drug from BCMs to its cellular or subcellular site(s) of
action remains a crucial step in ensuring maximal therapeutic efficacy. While primarily
qualitative methods have been frequently employed in examining the intracellular fate of
BCMs and/or drugs,119-121 there is a shortage of quantitative data reporting the localization of
these materials within cellular fractions (e.g. nucleus and cytoplasm). Despite the wealth of
strategies that have been devised to achieve truly targeted drug delivery (e.g. active targeting,
pH-sensitive polymers),122 the design of in vitro studies should anticipate corresponding in
vivo studies in order to gauge the clinical potential of the micellar system under investigation.
In fact, Hoang et al. have demonstrated both the in vitro and in vivo targeting performance
of a radiolabeled BCM formulation actively targeted with the Fab fragment of the monoclonal
antibody trastuzumab (TmAb-Fab) and a nuclear localization signal (NLS). 111In-labeled
TmAb-Fab-NLS-BCMs were successfully internalized via receptor-mediated endocytosis in
HER2-expressing breast cancer cells. In particular, nuclear localization of approximately 43 %
of the internalized BCM population was obtained, enabling site-specific and controlled cell
damage induced by the emission of short range Auger electrons.123 This same BCM system
was further challenged in vivo whereby the PK, and biodistribution, as well as the tumor,
cellular and subcellular levels of BCM accumulation were quantified in BT-474 (HER2+) and
MDA-MB-231 (HER2-) tumor xenograft models. In vivo cell fractionation of the tumors
confirmed increases in both cell association and nuclear uptake of the actively targeted BCMs
as a result of TmAb-Fab and NLS functionalization, respectively, in the HER2-overexpressing
BT-474 model. This study effectively demonstrated the potential of a multifunctional BCM
platform to provide selective in vivo targeting and delivery of therapeutics (i.e. targeting the
cytoplasm and/or nucleus). Further insight was also gained in the intratumoral distribution of
Chapter 2: Challenges facing Block Copolymer Micelles 49
the radiolabeled BCMs using autoradiography, revealing a positive correlation between BCM
localization and perfused blood vessels.48 Finally, in many cases multidrug resistance
remains an outstanding impediment to the effective cellular delivery of drug. Seminal
research by Kabanov’s group has shown that formulation of drugs in BCMs may serve as a
strategy to overcome multidrug resistance.124 Specifically, Pluronic® copolymer unimers have
been shown to increase the sensitivity of P-glycoprotein (Pgp) overexpressing multidrug
resistant cells by a factor of 20 to 1000 for Pgp substrates such as doxorubicin and by a factor
of 2-5 for non-Pgp substrates such as cisplatin.125
Figure 2-3: Challenges for BCM-based drug delivery. For the successful translation of BCM
technology into the clinical space, three main delivery challenges are presented: I) stability in
the blood compartment and prolonged circulation, II) penetration into deep layers of the tumor
tissue and III) tumor bioavailability of the drug within cancer cells. I) Following systemic
administration, BCMs must remain intact for effective delivery of the drug payload to the
target site (a). However, a challenge to this feat is leakage of drug from the nanocarrier (b) or
disruption of the integrity of the BCM and subsequent release of drug in its free form (c). II)
Within the tumor, a dynamic network of blood vessels exists wherein perfusion of certain
vessels may be deficient (e.g. transient or non-existent) (a). Following extravasation, tumor
penetration of NDDSs is possible by means of diffusion (b), albeit hindered by the dense extra-
cellular matrix and/or sequestration via stromal cells (c). III) Upon delivery of the shuttled
therapy via BCMs at the tumor site, three scenarios may present themselves: the carrier
system need not penetrate the tumor interstitium and the drug is released in the extracellular
space (a), the carrier system penetrates into the tumor and drug is released into the
extracellular space (b) and finally, the carrier system penetrates through the tissue and is
internalized by a cell where the drug is released (c).
Chapter 2: Challenges facing Block Copolymer Micelles 50
Chapter 2: Challenges facing Block Copolymer Micelles 51
The tumor penetration of nanomedicines and their potential for tumor cell-specific drug
delivery remain limiting factors in their ability to withstand pre-clinical evaluation and attain
clinical status. As we will discuss in more detail in section 2.4, the breadth of pre-clinical
models available to challenge the performance of new and existing candidates constitutes
both a help and a hindrance; the characterization of tumor properties that can feed into the
design and/or modulation of NDDS-based strategies is crucial.
Chapter 2: Challenges facing Block Copolymer Micelles 52
2.4. Translatability: best practices and lessons learned
As outlined in the previous section, a number of significant in vivo barriers remain to be
overcome for effective anti-cancer drug delivery. Among a variety of nanosystems, BCMs
offer a relatively greater degree of control over their physico-chemical characteristics. The
question remains whether we can maintain this ‘chemical’ leverage in the face of increasingly
complex biological challenges in order to fully exploit this technology, making it a viable
treatment option for an increasing number of patients. Recently, a number of reviews have
critically examined the progress achieved to date with nanomedicines developed for use in
oncology.21-23
The seminal work performed by Maeda et al. on the EPR effect was published in 1986,31
with Doxil® being approved in 1995 as the first nanomedicine for cancer treatment.126 These
milestone events, only decades old, may appear distant in the midst of the vast array of
research that has grown exponentially in this area, reflected in the number of publications on
several NDDSs for applications in oncology (Figure 2-2). Despite our growing repertoire of
candidate formulations – a true testament to the inventive and dynamic nature of our
research community – it is important to examine and reflect on the few formulations that
have graduated from pre-clinical study, in the hopes of extracting value from both positive
and negative findings. At this time, there are five BCM formulations of chemotherapeutic
drugs in clinical development (Table 1) and only one of these formulations, Genexol-PM, has
been approved for human use (albeit exclusively in South Korea). What major impediments
may have delayed or obstructed the achievement of more substantial contributions to the
advancement of BCM-based cancer treatment? From our perspective, two critical concepts
merit further investigation in support of the successful clinical translation of BCMs and other
nanomedicines: (1) EPR status in pre-clinical models and in the clinical setting, and whether
Chapter 2: Challenges facing Block Copolymer Micelles 53
its operation is critical and/or present to allow for successful nanomedicine-based therapy;
and (2) the dynamic heterogeneity present in cancer, and the degree to which it can and
cannot be harnessed. Herein, we aim to provide a discussion of strategies that may help to
address both challenges.
2.4.1. Towards clinically relevant nanoformulations
Grislain et al. first observed NDDS accumulation in tumors in 1983,127 while in 1986,
Matsumura and Maeda proposed the EPR effect as the driving principle for this
phenomenon.31 Since this time, the EPR effect has become the cornerstone of design for
countless nanomedicines. BCMs are one of the more advanced technologies within this
toolbox. For the vast majority of publications in the NDDS-based drug delivery field, a common
refrain has been to achieve a long circulation lifetime with the aim of exploiting the EPR effect
as the sole driving force for ensuring therapeutic efficacy. This particular view has led to many
publications and a considerable number of pre-clinical success stories, but very few tangible
outcomes for patients. Despite being perceived as a universal trait of solid tumors, the
variability observed in the delivery of NDDSs to tumors has dubbed the EPR effect as
somewhat of a “moving target”.128 Just recently, Nichols and Bae denounced the chronic over-
simplification of the concept of the EPR effect, which may be responsible, among other things,
for the inadequate therapeutic outcomes obtained with some nanomedicines.78 This may
limit the widespread use of a passive targeting approach on the basis of EPR.26,75,129 In
response, we put forward the need to design and evaluate BCM formulations beyond
exploitation of the EPR effect alone (Figure 2-4). Specifically, we must also aim to maximize
the tumor bioavailability of the drug which includes drug release and intratumoral distribution,
as well as intended cellular and subcellular delivery of the therapeutic. A comprehensive
Chapter 2: Challenges facing Block Copolymer Micelles 54
investigation of leading nanoformulations, originating from NDDS design to in-human trials,
may shed valuable light on promising avenues as well as blind alleys.
Two model BCM-based formulations that have achieved profound improvements in
prolonging the circulation lifetime of the entrapped drug in comparison to that of the free drug
were developed by Nanocarrier (Japan); namely, NK105 and NC-6004. NC-6004 was
developed as a micelle formulation for cisplatin (cis-dichlorodiammineplatinum(II); CDDP)
and is based on the self-assembly of PEG-b-poly(glutamic acid) copolymer (i.e. PEG-b-P(Glu)).
This formulation employed CDDP to drive micelle formation by cross-linking the poly(glutamic
acid) core, resulting in highly stable BCMs.130 Of note, poly(aspartic acid) was initially used as
the core polymer, which also increased blood AUC of the drug by a factor of 5 relative to the
BCMs bearing a poly(glutamic acid) core; however, the micelles were found to be less stable
and CDDP accumulated in the liver and spleen resulting in significant toxicity.130 In a recent
publication, Mochida et al. showed that the spatial arrangement of the amino acid secondary
structure within the PEG-b-P(Glu) BCMs has a significant influence on micelle stability.19 Using
only optically active D- or L-glutamic acid units resulted in the formation of α-helices within
the micelle core which in turn improved stability under in vivo conditions, whereas use of a
combination of D- and L-glutamate did not. These studies further demonstrate that even very
small changes in the polymer backbone (i.e. addition of a methylene group or change in
stereoisomerism) can have a significant impact on drug retention and formulation stability in
vivo. The first pre-clinical evaluation of NC-6004 in CDF1 mice bearing subcutaneous tumors
of murine C26 colon carcinoma reported a significant increase in plasma AUC of drug,
resulting in a 20-fold increase in tumor accumulation relative to administration of free CDDP.
These results yielded complete tumor eradication in two thirds of the treated animals,130 while
the typical dose-limiting nephrotoxicity, neurotoxicity and ototoxicity associated with
Chapter 2: Challenges facing Block Copolymer Micelles 55
administration of free CDDP were found to be significantly reduced.20,131 In the Phase I clinical
trial of NC-6004, a 230-fold increase in the plasma half-life of CDDP was reported with an 8.5-
fold increase in plasma AUC in comparison to administration of free drug.132 Consistent with
observations from the pre-clinical animal studies, toxicities were less frequent and less severe
than those observed with free drug; a finding which was attributed to the controlled release
of the platinum agent. In this study, stable disease was achieved in 40 % of a heavily pre-
treated patient population presenting with a wide range of advanced stage cancers.132
Recently, NC-6004 successfully completed a Phase I/II trial for locally advanced and
metastatic pancreatic cancer in combination with gemcitabine (NCT00910741) and is now
recruiting for a Phase III trial (NCT02043288, Table 1).
As discussed previously in section 2.3.1.2, NK105 is a micellar formulation formed from
diblock copolymers wherein PEG functions as the hydrophilic component of the copolymer
and a modified polyaspartate core is employed for increased compatibility with the
incorporated drug PTX.59 In pre-clinical evaluation, administration of the highly hydrophobic
drug PTX in this formulation resulted in more than a 125-fold increase in the plasma AUC in
comparison to administration of a conventional PTX formulation, which included drug
solubilized in Cremophor EL and ethanol.133 Whereas PTX in its conventional formulation was
rapidly cleared from the plasma and the tumor site, NK105-formulated PTX was retained
much longer demonstrating a greater than 50-fold increase in plasma AUC and an over 10-
fold increase in tumor AUC.59 Correspondingly, much greater efficacy, as well as reduced
toxicity, were observed in mice which was attributed to a decrease in distribution to normal
tissues.59,133 In a Phase II clinical trial of NK105 in 57 patients with previously treated
advanced gastric cancer, the PK observed at a 150 mg/m2 dose resulted in a 9-fold increase
in plasma AUC in comparison to conventional PTX delivered at a dose of 210 mg/m2. The
Chapter 2: Challenges facing Block Copolymer Micelles 56
overall response rate (ORR) observed was 25 %, with 30.4 % of patients achieving stable
disease for several months. Since there was no free drug control arm in this study, further
interpretation of the results is difficult at this time.134
The detailed review of these two examples illustrates the fact that long circulation can be
upheld using BCM technology, particularly following clinical translation, and highlights the
specific design requirements needed to engineer a BCM system that is suitable to load and
retain a particular drug of interest. Indeed, long circulation remains a prerequisite for achieving
tumor accumulation.128 However, attaining a sufficient level of tumor deposition constitutes
merely one step in the long itinerary of a NDDS from the point of injection to its target site
(Figure 2-3). It is also noteworthy that although we have highlighted two remarkable cases of
clinically successful BCM formulations, the tumor uptake level for NC-6004 reaches only
about 10 % of the injected dose, meaning that the remaining 90 % of the administered drug
distributes to off-target sites. Nonetheless, 10 % remains significantly higher than levels
achieved with conventional CDDP formulations (i.e. < 1 %).130
Renewed efforts are merited in increasing the tumor exposure of the shuttled drug and
ensuring a lethal effect takes place at the target cancer cells. Tumor drug bioavailability
effectively hinges on three main factors: drug release, tissue penetration and cellular uptake.
NDDSs are faced with the daunting challenge of achieving tumor penetration coupled with
cellular delivery of the drug. In general, there are three drug release scenarios which can occur
upon reaching the tumor site: (1) the carrier system does not penetrate into the tumor and
releases the drug into the extracellular space from where it diffuses into cells, (2) the carrier
system penetrates into the tumor and releases the drug into the extracellular space and finally
(3) the carrier system penetrates through the tissue and is internalized by a cell where it
releases the drug (Figure 2-3). It is important to note that these three scenarios illustrate single
Chapter 2: Challenges facing Block Copolymer Micelles 57
phenomena that are likely not exclusive of one another but rather can occur simultaneously,
with a distinct probability for each. Very few studies have examined the tumor penetration of
BCMs in vivo as outlined above.9,46,49 Furthermore, it is important to recognize that even small
molecules such as doxorubicin are limited in terms of tumor penetration and in general are
not able to reach all cancer cells. This is a serious issue that is most likely exacerbated for
NDDSs given their increased size.71, 130, 131 As a result tumor penetration is of relevance even
for avascular metastatic lesions or remnants from cytoreductive surgery that may be only a
few millimeters in diameter. Additionally, most chemotherapeutic agents act within the cell
and require subcellular localization. It is also worth mentioning that the common narrative of
adding a receptor-mediated, or “active”, targeting moiety to a NDDS to enable “homing” to the
tumor target is a broad simplification of the actual phenomenon. In fact, our current
knowledge on the potential advantages of active targeting is incomplete, as outlined in the
previous section (cf. section 2.3.3.1) as well as in a number of excellent reviews on this
topic22,24,26,79.
At this time in the field, we require more studies that examine the distribution of NDDSs
and/or drug administered in NDDSs at the whole body, tissue and cellular levels. It is
recognized that these in vivo distribution studies are resource-intensive and challenging;
however, they are instrumental in elucidating the ultimate fate and thus therapeutic potential
of the NDDS. Findings from these studies can then be related to the physico-chemical
properties of the NDDS and associated efficacy data.
2.4.2. The management of heterogeneity at the pre-clinical and clinical levels
As outlined in this article, there are three factors which can influence the success of NDDS-
based therapy: properties of the delivery system, biological features of the tumor, and finally,
Chapter 2: Challenges facing Block Copolymer Micelles 58
translatability of pre-clinical data to the clinical setting. Cancer arises from different organs
and reveals itself in many different subtypes originating from cells that proliferate in an
uncontrolled manner as a result of genetic and epigenetic aberrations. Consequently, the
disease manifests with tremendous heterogeneity given the manifold cell types and the
complex underlying mechanisms of mutation it derives from.135 The observed heterogeneity
includes variability between patients and tumor types as well as within the same patient. As
a result, tumor heterogeneity influences all aspects of the transport of nanomedicines.7 In
order to gain a better understanding, pre-clinical models of cancer play a pivotal role in
formulation optimization and selection. Typically, evaluation is conducted in models of
increasing complexity with initial assessment of efficacy in monolayer cultures of cancer cells
using chromogenic assays such as MTT.136 The US National Cancer Institute (NCI) has
developed a panel of 60 cell lines representative of breast, colon, CNS, leukemia, lung,
melanoma, ovarian, prostate and renal cancers.137 This NCI60 panel was originally developed
in the 1980’s and has been characterized extensively by the NCI and others. In general, it
provides guidance on cell lines that are reflective of a particular cancer and a means of
standardizing in vitro assessment of drug activity.137,138 Evaluation of the efficacy of drug
formulations in monolayer cultures is a first step; yet, given the simplicity of this model, it is
largely unable to predict efficacy in vivo.139,140 The major limitation of cell monolayers is their
inability to reflect transport restrictions and other mechanisms of drug resistance associated
with the three dimensional tumor microenvironment.141 In contrast to in vivo tumor models,
in vitro cultures are well suited for systematic evaluation of drug activity and uptake in a highly
controlled environment. As such, three-dimensional cell culture systems may serve as an
intermediary between the oversimplified structure of monolayer cultures and the highly
complex nature of in vivo models.141-143 Spheroid cultures possess a complex network of cell-
cell contacts and extracellular matrix, as well as pH, oxygen, metabolic and proliferative
Chapter 2: Challenges facing Block Copolymer Micelles 59
gradients analogous to the conditions in poorly vascularized and avascular regions of solid
tumors. We and others have established methods for the evaluation of BCM-based
formulations in spheroids (i.e. MCTS) 141,144,145 and have shown that such models are able to
predict relative trends in the tissue penetration of BCMs in vivo.49
As a next step, PK and tumor drug bioavailability become the most relevant formulation
parameters to assess in animal models. For this purpose, either cancer cells of murine or
human origin are introduced into immune suppressed mice or genetically engineered mouse
models (GEEMs) are used to develop spontaneous tumors.146,147 Most commonly,
immortalized cell lines are employed and are implanted either subcutaneously or at the
relevant orthotopic site. Therefore, despite the many possibilities available, there is
tremendous simplicity associated with the vast majority of the animal models that we employ
routinely in our research. Indeed, numerous reports have questioned the physiological
relevance of subcutaneous models and in particular their ability to predict human response
to therapy.146-148 Furthermore, in a study conducted by the NCI on 39 small-molecule drugs, it
was found that “in vivo activity in a particular histology in a tumor model did not closely
correlate with activity in the same human cancer histology, casting doubt on the
correspondence of the pre-clinical models to clinical results”.139 One of the major issues
associated with these models is the relative ratio of tumor-to-body weight. As discussed in a
recent review article by Taurin et al., a typical subcutaneous tumor weighs about 1 g which is
equivalent to a weight percentage of 4 % in a typical laboratory mouse with a 25 g body
weight.75 For a normal adult of 70 kg, this would result in a tumor mass of about 2.8 kg,
whereas most often surgical resection would be attempted prior to the administration of any
chemo- or targeted biological therapy. In the same way, a tumor of this size will develop within
weeks in a mouse model; while, in humans, it would generally take years to develop.26,75
Chapter 2: Challenges facing Block Copolymer Micelles 60
Lammers et al. observed that the tumor accumulation of HPMA-doxorubicin conjugates was
2-fold lower in a slow growing tumor model in mice in comparison to that achieved in a fast
growing model of prostate carcinoma.102 Whereas the fast growing model developed a tumor
of 1 cm in size within two weeks, for the slow growing model, this process took approximately
one year. As expected, histology of the slow growing model indicated a much more mature
and well-differentiated vascular system with a reduced number and size of endothelial gaps
thereby limiting extravasation of the polymer-drug conjugate. In comparison to the fast
growing subcutaneous models, more complex animal models wherein the tumor is grown at
the orthotopic site have been proposed to afford more insight.146,147 These models benefit
from the integration of image-based methods, such as bioluminescence imaging to assess
tumor burden.149 Moreover, orthotopic models often metastasize and provide an opportunity
to assess the effect of a formulation on treating metastatic burden. Patient-derived primary
tumor models preserve many of the phenotypic and genotypic features of the patient tumor
which makes them one of the most predictive experimental models.150 Yet, access to tissue
samples from patient biopsies is very limited. Alternatively, GEEMs allow monitoring the
gradual development of a cancer and the effect of treatment for prolonged periods under a
vital immune system.151 As such, these models can yield insight into the potential of NDDSs
capable of targeting early stages of tumor progression and/or allow the investigation of
immunostimulatory NDDSs.152,153
Overall, the various models should be regarded as tools that offer the possibility to gain
specific insights into the complex relationships between nanomedicines and cancer.
Therefore, it is imperative to have a clear clinical question in mind and then to seek out the
appropriate model to address this specific question in the best possible way. Every model is
by its own definition an imperfect replication of the true nature of the disease. At this time, no
Chapter 2: Challenges facing Block Copolymer Micelles 61
one model accurately depicts all cancers given the heterogeneity and complexity associated
with the clinical presentation of the disease. However, pre-clinical tumor models
characterized for specific parameters that are reproduced in the clinical setting (e.g. vascular
density, permeability; tumor interstitial permeability) and known to influence the efficacy of
BCM-drug formulations may increase the predictive power of pre-clinical studies and expedite
the translation of such formulations to the clinic.
Chapter 2: Challenges facing Block Copolymer Micelles 62
Figure 2-4: Beyond the EPR effect. For many years, the EPR-centric paradigm has dominated
the development of NDDSs including BCMs. In this paradigm, NDDS stability, drug retention
and long circulation in the bloodstream have been sought after in order to exploit the EPR
effect which has been assumed to be directly predictive of therapeutic effect. However, given
the tremendous heterogeneity in cancer, it is now understood that we must design NDDSs
beyond the EPR effect alone and that parameters such as tumor penetration, tumor drug
bioavailability and, in particular, patient EPR status (i.e. extent to which EPR is operative), have
a combined impact on therapeutic efficacy. The proposed ultra-EPR paradigm incorporates
this deeper understanding of biological response to NDDS mediated drug delivery and may
enable patient stratification based on EPR status which could potentially not only increase
therapeutic efficacy, but also reduce toxicities.
* = The role of receptor-mediated targeting remains disputed in enhancing tumor
accumulation.
# = “ultra” to be interpreted as the original Latin meaning “beyond”.
Chapter 2: Challenges facing Block Copolymer Micelles 63
The design of clinical studies could hence benefit from the identification of select
pathophysiological traits, and the extent to which they are operative. As such, it has been
suggested to pre-select patients that are likely to respond to a specific therapy prior to
commencing a treatment cycle.154 In 2012, the FDA released a draft guidance on “Enrichment
Strategies for Clinical Trials to Support Approval of Human Drugs and Biological Products”
wherein enrichment is defined as “the prospective use of any patient characteristic to select
a study population in which detection of a drug effect is more likely than it would be in an
unselected population”. As a means to achieve patient enrichment, this document also
highlighted the need for co-development of diagnostic tests in parallel with drug development
activities.155 For this purpose, theranostic nanoformulations, or therapeutic nanosystems
possessing a diagnostic equivalent, have been proposed for image-guided selection and
treatment of patients that are likely to respond to a specific therapy.156-158 The significance of
this strategy was exemplified in the seminal work conducted by Harrington and colleagues,
alluded to earlier, in which the vast heterogeneity in liposome uptake was exposed between
tumor types, as well as between patients presenting with the same tumor type. As a result, a
pre-treatment scan using radiolabeled 111In-liposomes was proposed as a future strategy to
“predict the likelihood of a response to treatment [to liposomal therapy] in different groups of
patients”.98 In keeping with this approach, Merrimack Pharmaceuticals recently reported
completion of a Phase I clinical trial of MM-302, a HER2-targeted liposome formulation of
doxorubicin (NCT01304797) for treatment of patients with advanced HER2-positive breast
cancer. In this trial, “tumor deposition and biodistribution of MM-302 was assessed in a
subset of patients using 64Cu-labeled MM-302 and positron emission tomography
(PET)/computed tomography (CT) imaging”.159 Pending results from the trial, such a
diagnostic could potentially be employed to pre-select patients that are likely to respond to
MM-302.
Chapter 2: Challenges facing Block Copolymer Micelles 64
The selective success of liposomal cancer therapy has been attributed to the specific
nature of the tumor types that have benefitted (e.g. AIDS-related Kaposi’s sarcoma, multiple
myeloma).78 According to Nichols and Bae, Doxil® showed greatest efficacy in tumors bearing
little solid tissue and/or in which EPR was highly operative.78 At the same time, it was
suggested by the authors that an “anti-EPR effect” could effectively sabotage the efforts of
macro- and NDDSs due to the transport restrictions resulting from increasing levels of IFP.78
As such, there may be no definitive conclusion at this time regarding the seemingly capricious
nature of the EPR effect. While prudent interpretation is earnestly encouraged,75,78 so is the
incorporation of complementary and informative imaging studies.160
Indeed, the advent of theranostics has brought with it the capacity to transform the
capabilities of BCMs in oncology. The concept to “diagnose and deliver” 161 is underway in the
form of various formulations in pre-clinical development. The ability to identify likely
responders to nanomedicines based on their tumor uptake of the NDDS and/or tumor
properties (e.g. vascular permeability, density and blood flow) is made possible through
imaging, and allows us to positively exploit the inherent heterogeneities present in cancer.
Furthermore, real-time and continued monitoring of therapies facilitates their adjustment on
a per-patient basis in order to maintain a maximal therapeutic index.160 BCMs have been
employed in conjunction with imaging modalities for image-guided drug delivery. For
instance, theranostic BCMs prepared from PEG-b-PDLLA have been loaded with
superparamagnetic iron oxide nanoparticles (SPIONs) and doxorubicin, for magnetic
resonance imaging (MRI) and therapy, respectively, with added receptor-mediated targeting
to endothelial 162 or tumor cells.163 Similarly, optical imaging has been harnessed as a
complementary modality to the administration of chemotherapy using BCMs.164,165 However,
the use of therapeutic-diagnostic pairs, such as the MM-302 and 64Cu-labeled MM-302 pair
Chapter 2: Challenges facing Block Copolymer Micelles 65
introduced above seems to be prevalent among NDDSs wherein an active agent is neither
complexed, conjugated nor chelated. Indeed, such modification to the initial NDDS would
result in a new chemical entity and ensuing alterations in in vivo performance (i.e. PK and
biodistribution). Thus, the use of therapeutic-diagnostic pairs appears suitable for NDDSs
whereby the added drug bears no appreciable effect on the nanosystem. Further, the
diagnostic component of the pair is expected to serve a true purpose as a tool for patient pre-
selection and stratification, whereas a theranostic nanosystem would effectively introduce a
dose of therapy in conjunction with imaging. The latter scenario would more likely satisfy the
aim of monitoring response to therapy, potentially in real-time.
The selectivity in patient response to NDDS-based therapy highlights the need for more
sophisticated strategies in stratifying responders. In particular, EPR status and function have
proven elusive parameters that can be characterized in patients. For instance, whereas
interpretation of a binary test such as the mutation of a gene may be less complicated, it
becomes more challenging with continuous tests where thresholds need to be defined. This
may indeed be the case in determining a measure of EPR.154 As such, the clinical translation
of NDDS-based therapies may effectively benefit from improved criteria for patient selection.
In the FDA draft guidance for clinical trials (cf. section 2.4.2), it is stated that “enrichment
strategies should enable smaller populations of patients to be used in trials and also enable
a larger effect to be observed”.155 Therefore, the validation of a diagnostic test is crucial in
order to determine cut-off points to include or exclude patients from NDDS-based therapy. It
is conceivable that a NDDS-based ‘standard’ may serve to identify the extent to which EPR
operates in a given patient; accordingly, a set of such standards may be devised in order to
represent various NDDSs (e.g. ~100 nm liposomes or sub-100 nm BCM systems) and thus
provide some calibration of EPR. Such a strategy would contribute to a more standardized
Chapter 2: Challenges facing Block Copolymer Micelles 66
approach in NDDS-based drug delivery and more successful clinical translation. In summary,
the winning strategy to address heterogeneity may be less the pursuit of a “magic bullet” type
of nanomedicine that has the capability to treat any type of cancer, but rather matching the
patient’s tumor profile with the most promising treatment option.
Chapter 2: Challenges facing Block Copolymer Micelles 67
2.5. Conclusions
Today, scientists are presented with a wide array of models and tools at their disposal to
assess the performance of nanomedicines. However, an acute awareness and prudence of
the capabilities of our pre-clinical models has grown in significance, as we gain a better
understanding of those parameters (formulation- or tumor-related) which are most clinically
relevant and influential. The ability to design BCMs varying in size, shape and functionalization
offers considerable potential to achieve clinical translation through the possible
permutations. Furthermore, beyond pursuit of the right balance between attaining drug
retention in the NDDS while in the circulation and release at the tumor site, the
pharmacodynamic effects of the drug must also be considered in order to ensure therapeutic
levels of drug are reached. However, progress has yet to be made as we apprehend the
diverse nature of cancer, and accordingly, the likely diverse solutions that are bound to
emerge. Efforts must be undertaken to work towards finding optimal parameters and
combinations as a function of cancer properties and how they may be expressed within each
individual host. While incremental progress will certainly arise from promising and conclusive
reports, equal consideration of negative findings is bound to contribute significantly to our
advancement. Furthermore, given the challenges the heterogeneity of cancer poses, the focus
should be on providing solutions that are specifically aimed at a clearly defined patient
population. If initial success can be achieved, this will propel the field and unleash new
excitement as well as funding opportunities that can widen the range of applicability of the
technology to a larger population.
Chapter 2: Challenges facing Block Copolymer Micelles 68
2.6. Acknowledgements
S.E. is funded by the NSERC CREATE Biointerfaces Training Program and holds an Ontario
Trillium Scholarship. S.N.E. is the recipient of the Pfizer Canada Graduate Fellowship in
Pharmaceutical Sciences and a fellowship from the CIHR Strategic Training Program in
Biological Therapeutics. C. A. acknowledges GlaxoSmithKline for an endowed Chair in
Pharmaceutics and Drug Delivery and research funding from NSERC, the CIHR Breast Cancer
Initiative and the Avon Foundation.
Chapter 2: Challenges facing Block Copolymer Micelles 69
2.7. References
1. Birrenbach G, Speiser PP. Polymerized micelles and their use as adjuvants in immunology. Journal of pharmaceutical sciences. Dec 1976;65(12):1763-1766.
2. Kreuter J. Nanoparticles--a historical perspective. International journal of pharmaceutics. Feb 22 2007;331(1):1-10.
3. Bader H, Ringsdorf H, Schmidt B. Watersoluble Polymers in Medicine. Angew Makromol Chem. 1984;123(Aug):457-485.
4. Alakhov V, Klinski E, Li SM, et al. Block copolymer-based formulation of doxorubicin. From cell screen to clinical trials. Colloid Surface B. Nov 1999;16(1-4):113-134.
5. Matsumura Y. The drug discovery by nanomedicine and its clinical experience. Jpn J Clin Oncol. Jun 2014;44(6):515-525.
6. Cabral H, Kataoka K. Progress of drug-loaded polymeric micelles into clinical studies. Journal of controlled release : official journal of the Controlled Release Society. Sep 28 2014;190:465-476.
7. Mikhail AS, Allen C. Block copolymer micelles for delivery of cancer therapy: Transport at the whole body, tissue and cellular levels. Journal of Controlled Release. Sep 15 2009;138(3):214-223.
8. Owen SC, Chan DPY, Shoichet MS. Polymeric micelle stability. Nano Today. Feb 2012;7(1):53-65.
9. Cabral H, Matsumoto Y, Mizuno K, et al. Accumulation of sub-100 nm polymeric micelles in poorly permeable tumours depends on size. Nat Nanotechnol. Dec 2011;6(12):815-823.
10. Murakami M, Cabral H, Matsumoto Y, et al. Improving Drug Potency and Efficacy by Nanocarrier-Mediated Subcellular Targeting. Sci Transl Med. Jan 5 2011;3(64).
11. Torchilin VP. Micellar nanocarriers: Pharmaceutical perspectives. Pharmaceutical research. Jan 2007;24(1):1-16.
12. Zhang L, Gu FX, Chan JM, Wang AZ, Langer RS, Farokhzad OC. Nanoparticles in medicine: Therapeutic applications and developments. Clin Pharmacol Ther. May 2008;83(5):761-769.
13. Kwon GS, Kataoka K. Block-Copolymer Micelles as Long-Circulating Drug Vehicles. Advanced drug delivery reviews. Sep 1995;16(2-3):295-309.
14. Allen C, Maysinger D, Eisenberg A. Nano-engineering block copolymer aggregates for drug delivery. Colloid Surface B. Nov 1999;16(1-4):3-27.
15. Nicolas J, Mura S, Brambilla D, Mackiewicz N, Couvreur P. Design, functionalization strategies and biomedical applications of targeted biodegradable/biocompatible polymer-based nanocarriers for drug delivery. Chemical Society Reviews. 2013;42(3):1147-1235.
16. Jones MC, Leroux JC. Polymeric micelles - a new generation of colloidal drug carriers. Eur J Pharm Biopharm. Sep 1999;48(2):101-111.
17. Rapoport N. Physical stimuli-responsive polymeric micelles for anti-cancer drug delivery. Prog Polym Sci. Aug-Sep 2007;32(8-9):962-990.
18. Mikhail AS, Allen C. Poly(ethylene glycol)-b-poly(epsilon-caprolactone) Micelles Containing Chemically Conjugated and Physically Entrapped Docetaxel: Synthesis, Characterization, and the Influence of the Drug on Micelle Morphology. Biomacromolecules. May 2010;11(5):1273-1280.
Chapter 2: Challenges facing Block Copolymer Micelles 70
19. Mochida Y, Cabral H, Miura Y, et al. Bundled Assembly of Helical Nanostructures in Polymeric Micelles Loaded with Platinum Drugs Enhancing Therapeutic Efficiency against Pancreatic Tumor. Acs Nano. Jul 2014;8(7):6724-6738.
20. Baba M, Matsumoto Y, Kashio A, et al. Micellization of cisplatin (NC-6004) reduces its ototoxicity in guinea pigs. Journal of Controlled Release. Jan 10 2012;157(1):112-117.
21. Lammers T. SMART drug delivery systems: Back to the future vs. clinical reality. International journal of pharmaceutics. Sep 15 2013;454(1):527-529.
22. Park K. Facing the Truth about Nanotechnology in Drug Delivery. Acs Nano. Sep 2013;7(9):7442-7447.
23. Ruenraroengsak P, Cook JM, Florence AT. Nanosystem drug targeting: Facing up to complex realities. Journal of Controlled Release. Feb 15 2010;141(3):265-276.
24. Crommelin DJ, Florence AT. Towards more effective advanced drug delivery systems. International journal of pharmaceutics. Sep 15 2013;454(1):496-511.
25. Bae YH, Mrsny RJ, Park K. Cancer Targeted Drug Delivery An Elusive Dream. [S.l.]: Springer New York; 2013: http://myaccess.library.utoronto.ca/login?url=http://link.springer.com/openurl?genre=book&isbn=978-1-4614-7875-1.
26. Lammers T, Kiessling F, Hennink WE, Storm G. Drug targeting to tumors: principles, pitfalls and (pre-) clinical progress. Journal of controlled release : official journal of the Controlled Release Society. Jul 20 2012;161(2):175-187.
27. Jain RK, Stylianopoulos T. Delivering nanomedicine to solid tumors. Nature reviews. Clinical oncology. Nov 2010;7(11):653-664.
28. Seymour LW, Miyamoto Y, Maeda H, et al. Influence of molecular weight on passive tumour accumulation of a soluble macromolecular drug carrier. European journal of cancer. 1995;31A(5):766-770.
29. Li SD, Huang L. Pharmacokinetics and biodistribution of nanoparticles. Mol Pharm. Jul-Aug 2008;5(4):496-504.
30. Choi HS, Liu W, Misra P, et al. Renal clearance of quantum dots. Nature biotechnology. Oct 2007;25(10):1165-1170.
31. Matsumura Y, Maeda H. A new concept for macromolecular therapeutics in cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor agent smancs. Cancer Res. Dec 1986;46(12 Pt 1):6387-6392.
32. Fang J, Nakamura H, Maeda H. The EPR effect: Unique features of tumor blood vessels for drug delivery, factors involved, and limitations and augmentation of the effect. Advanced drug delivery reviews. Mar 18 2011;63(3):136-151.
33. Alexis F, Pridgen E, Molnar LK, Farokhzad OC. Factors affecting the clearance and biodistribution of polymeric nanoparticles. Mol Pharm. Jul-Aug 2008;5(4):505-515.
34. Moghimi SM, Hedeman H, Muir IS, Illum L, Davis SS. An Investigation of the Filtration Capacity and the Fate of Large Filtered Sterically-Stabilized Microspheres in Rat Spleen. Biochimica et biophysica acta. Jul 11 1993;1157(3):233-240.
35. Moghimi SM, Hunter AC, Murray JC. Long-circulating and target-specific nanoparticles: Theory to practice. Pharmacological Reviews. Jun 2001;53(2):283-318.
36. Owens DE, Peppas NA. Opsonization, biodistribution, and pharmacokinetics of polymeric nanoparticles. International journal of pharmaceutics. Jan 3 2006;307(1):93-102.
Chapter 2: Challenges facing Block Copolymer Micelles 71
37. Moghimi SM, Hunter AC, Andresen TL. Factors controlling nanoparticle pharmacokinetics: an integrated analysis and perspective. Annu Rev Pharmacol Toxicol. Jan 17 2012;52:481-503.
38. Chrastina A, Massey KA, Schnitzer JE. Overcoming in vivo barriers to targeted nanodelivery. Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology. Jul-Aug 2011;3(4):421-437.
39. Kwon GS, Yokoyama M, Okano T, Sakurai Y, Kataoka K. Biodistribution of micelle-forming polymer-drug conjugates. Pharmaceutical research. Jul 1993;10(7):970-974.
40. Verhoef JJF. Questioning the use of PEGylation for drug delivery. Drug delivery and translational research. 2013;3(6).
41. Shiraishi K, Yokoyama M. Polymeric Micelles Possessing Polyethyleneglycol as Outer Shell and Their Unique Behaviors in Accelerated Blood Clearance Phenomenon. Biol Pharm Bull. Jun 2013;36(6):878-882.
42. Barz M, Luxenhofer R, Zentel R, Vicent MJ. Overcoming the PEG-addiction: well-defined alternatives to PEG, from structure-property relationships to better defined therapeutics. Polym Chem-Uk. 2011;2(9):1900-1918.
43. Knop K, Hoogenboom R, Fischer D, Schubert US. Poly(ethylene glycol) in Drug Delivery: Pros and Cons as Well as Potential Alternatives. Angew Chem Int Edit. 2010;49(36):6288-6308.
44. Hunter AC, Moghimi SM. Therapeutic synthetic polymers: a game of Russian roulette? Drug discovery today. Oct 1 2002;7(19):998-1001.
45. Yamamoto Y, Nagasaki Y, Kato Y, Sugiyama Y, Kataoka K. Long-circulating poly(ethylene glycol)-poly(D,L-lactide) block copolymer micelles with modulated surface charge. Journal of controlled release : official journal of the Controlled Release Society. Nov 9 2001;77(1-2):27-38.
46. Lee H, Fonge H, Hoang B, Reilly RM, Allen C. The Effects of Particle Size and Molecular Targeting on the Intratumoral and Subcellular Distribution of Polymeric Nanoparticles. Mol Pharmaceut. Jul-Aug 2010;7(4):1195-1208.
47. Fonge H, Huang H, Scollard D, Reilly RM, Allen C. Influence of formulation variables on the biodistribution of multifunctional block copolymer micelles. Journal of Controlled Release. Feb 10 2012;157(3):366-374.
48. Hoang B, Ekdawi SN, Reilly RM, Allen C. Active Targeting of Block Copolymer Micelles with Trastuzumab Fab Fragments and Nuclear Localization Signal Leads to Increased Tumor Uptake and Nuclear Localization in HER2-Overexpressing Xenografts. Mol Pharmaceut. Nov 2013;10(11):4229-4241.
49. Mikhail AS, Eetezadi S, Ekdawi SN, Stewart J, Allen C. Image-based analysis of the size- and time-dependent penetration of polymeric micelles in multicellular tumor spheroids and tumor xenografts. International journal of pharmaceutics. Apr 10 2014;464(1-2):168-177.
50. Lee H, Hoang B, Fonge H, Reilly RM, Allen C. In Vivo Distribution of Polymeric Nanoparticles at the Whole-Body, Tumor, and Cellular Levels. Pharmaceutical research. Nov 2010;27(11):2343-2355.
51. Geng Y, Dalhaimer P, Cai S, et al. Shape effects of filaments versus spherical particles in flow and drug delivery. Nat Nanotechnol. Apr 2007;2(4):249-255.
Chapter 2: Challenges facing Block Copolymer Micelles 72
52. Christian DA, Cai S, Garbuzenko OB, et al. Flexible filaments for in vivo imaging and delivery: persistent circulation of filomicelles opens the dosage window for sustained tumor shrinkage. Mol Pharm. Sep-Oct 2009;6(5):1343-1352.
53. Karmali PP, Simberg D. Interactions of nanoparticles with plasma proteins: implication on clearance and toxicity of drug delivery systems. Expert Opin Drug Del. Mar 2011;8(3):343-357.
54. Valle JW, Armstrong A, Newman C, et al. A phase 2 study of SP1049C, doxorubicin in P-glycoprotein-targeting pluronics, in patients with advanced adenocarcinoma of the esophagus and gastroesophageal junction. Invest New Drug. Oct 2011;29(5):1029-1037.
55. Matsumura Y, Hamaguchi T, Ura T, et al. Phase I clinical trial and pharmacokinetic evaluation of NK911, a micelle-encapsulated doxorubicin. Brit J Cancer. Nov 15 2004;91(10):1775-1781.
56. Nakanishi T, Fukushima S, Okamoto K, et al. Development of the polymer micelle carrier system for doxorubicin. Journal of Controlled Release. Jul 6 2001;74(1-3):295-302.
57. Sutton D, Nasongkla N, Blanco E, Gao JM. Functionalized micellar systems for cancer targeted drug delivery. Pharmaceutical research. Jun 2007;24(6):1029-1046.
58. Burt HM, Zhang XC, Toleikis P, Embree L, Hunter WL. Development of copolymers of poly(D,L-lactide) and methoxypolyethylene glycol as micellar carriers of paclitaxel. Colloid Surface B. Nov 1999;16(1-4):161-171.
59. Hamaguchi T, Matsumura Y, Suzuki M, et al. NK105, a paclitaxel-incorporating micellar nanoparticle formulation, can extend in vivo antitumour activity and reduce the neurotoxicity of paclitaxel. Brit J Cancer. Apr 11 2005;92(7):1240-1246.
60. Desale SS, Cohen SM, Zhao Y, Kabanov AV, Bronich TK. Biodegradable hybrid polymer micelles for combination drug therapy in ovarian cancer. Journal of controlled release : official journal of the Controlled Release Society. Nov 10 2013;171(3):339-348.
61. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature. Sep 14 2000;407(6801):249-257.
62. Folkman J, Watson K, Ingber D, Hanahan D. Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature. May 4 1989;339(6219):58-61.
63. De Bock K, Cauwenberghs S, Carmeliet P. Vessel abnormalization: another hallmark of cancer? Molecular mechanisms and therapeutic implications. Curr Opin Genet Dev. Feb 2011;21(1):73-79.
64. Kamoun WS, Chae SS, Lacorre DA, et al. Simultaneous measurement of RBC velocity, flux, hematocrit and shear rate in vascular networks. Nature methods. Aug 2010;7(8):655-660.
65. Fukumura D, Duda DG, Munn LL, Jain RK. Tumor Microvasculature and Microenvironment: Novel Insights Through Intravital Imaging in Pre-Clinical Models. Microcirculation. 2010;17(3):206-225.
66. Jain RK. Molecular regulation of vessel maturation. Nature medicine. Jun 2003;9(6):685-693.
67. Baluk P, Hashizume H, McDonald DM. Cellular abnormalities of blood vessels as targets in cancer. Curr Opin Genet Dev. Feb 2005;15(1):102-111.
Chapter 2: Challenges facing Block Copolymer Micelles 73
68. Morikawa S, Baluk P, Kaidoh T, Haskell A, Jain RK, McDonald DM. Abnormalities in pericytes on blood vessels and endothelial sprouts in tumors. The American journal of pathology. Mar 2002;160(3):985-1000.
69. Hobbs SK, Monsky WL, Yuan F, et al. Regulation of transport pathways in tumor vessels: role of tumor type and microenvironment. Proceedings of the National Academy of Sciences of the United States of America. Apr 14 1998;95(8):4607-4612.
70. Konerding MA, Malkusch W, Klapthor B, et al. Evidence for characteristic vascular patterns in solid tumours: quantitative studies using corrosion casts. Brit J Cancer. May 1999;80(5-6):724-732.
71. Vakoc BJ, Lanning RM, Tyrrell JA, et al. Three-dimensional microscopy of the tumor microenvironment in vivo using optical frequency domain imaging. Nature medicine. Oct 2009;15(10):1219-U1151.
72. Narang AS, Varia S. Role of tumor vascular architecture in drug delivery. Advanced drug delivery reviews. Jul 18 2011;63(8):640-658.
73. Minchinton AI, Tannock IF. Drug penetration in solid tumours. Nat Rev Cancer. Aug 2006;6(8):583-592.
74. Dreher MR, Liu W, Michelich CR, Dewhirst MW, Yuan F, Chilkoti A. Tumor vascular permeability, accumulation, and penetration of macromolecular drug carriers. Journal of the National Cancer Institute. Mar 1 2006;98(5):335-344.
75. Taurin S, Nehoff H, Greish K. Anticancer nanomedicine and tumor vascular permeability; Where is the missing link? Journal of controlled release : official journal of the Controlled Release Society. Dec 28 2012;164(3):265-275.
76. Kwon G, Suwa S, Yokoyama M, Okano T, Sakurai Y, Kataoka K. Enhanced Tumor Accumulation and Prolonged Circulation Times of Micelle-Forming Poly(Ethylene Oxide-Aspartate) Block Copolymer-Adriamycin Conjugates. Journal of Controlled Release. Feb 1994;29(1-2):17-23.
77. Charrois GJR, Allen TM. Rate of biodistribution of STEALTH((R)) liposomes to tumor and skin: influence of liposome diameter and implications for toxicity and therapeutic activity. Bba-Biomembranes. Jan 10 2003;1609(1):102-108.
78. Nichols JW, Bae YH. EPR: Evidence and fallacy. Journal of controlled release : official journal of the Controlled Release Society. Sep 28 2014;190:451-464.
79. Bertrand N, Wu J, Xu X, Kamaly N, Farokhzad OC. Cancer nanotechnology: the impact of passive and active targeting in the era of modern cancer biology. Advanced drug delivery reviews. Feb 2014;66:2-25.
80. Pirollo KF, Chang EH. Does a targeting ligand influence nanoparticle tumor localization or uptake? Trends Biotechnol. Oct 2008;26(10):552-558.
81. Gray BP, McGuire MJ, Brown KC. A liposomal drug platform overrides peptide ligand targeting to a cancer biomarker, irrespective of ligand affinity or density. PloS one. 2013;8(8):e72938.
82. Cheng Z, Al Zaki A, Hui JZ, Muzykantov VR, Tsourkas A. Multifunctional nanoparticles: cost versus benefit of adding targeting and imaging capabilities. Science. Nov 16 2012;338(6109):903-910.
83. Kirpotin DB, Drummond DC, Shao Y, et al. Antibody targeting of long-circulating lipidic nanoparticles does not increase tumor localization but does increase internalization in animal models. Cancer Res. Jul 1 2006;66(13):6732-6740.
Chapter 2: Challenges facing Block Copolymer Micelles 74
84. Lammers T, Hennink WE, Storm G. Tumour-targeted nanomedicines: principles and practice. Br J Cancer. Aug 5 2008;99(3):392-397.
85. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science. May 12 1989;244(4905):707-712.
86. Wang M, Thanou M. Targeting nanoparticles to cancer. Pharmacol Res. Aug 2010;62(2):90-99.
87. Dunne M, Zheng J, Rosenblat J, Jaffray DA, Allen C. APN/CD13-targeting as a strategy to alter the tumor accumulation of liposomes. Journal of controlled release : official journal of the Controlled Release Society. Sep 25 2011;154(3):298-305.
88. El Emir E, Qureshi U, Dearling JL, et al. Predicting response to radioimmunotherapy from the tumor microenvironment of colorectal carcinomas. Cancer Res. Dec 15 2007;67(24):11896-11905.
89. Blumenthal RD, Sharkey RM, Kashi R, Natale AM, Goldenberg DM. Physiological factors influencing radioantibody uptake: a study of four human colonic carcinomas. Int J Cancer. Jul 30 1992;51(6):935-941.
90. Simnick AJ, Amiram M, Liu W, et al. In vivo tumor targeting by a NGR-decorated micelle of a recombinant diblock copolypeptide. Journal of controlled release : official journal of the Controlled Release Society. Oct 30 2011;155(2):144-151.
91. Miura Y, Takenaka T, Toh K, et al. Cyclic RGD-linked polymeric micelles for targeted delivery of platinum anticancer drugs to glioblastoma through the blood-brain tumor barrier. Acs Nano. Oct 22 2013;7(10):8583-8592.
92. Carmeliet P, Jain RK. Principles and mechanisms of vessel normalization for cancer and other angiogenic diseases. Nature reviews. Drug discovery. Jun 2011;10(6):417-427.
93. Fukumura D, Yuan F, Monsky WL, Chen Y, Jain RK. Effect of host microenvironment on the microcirculation of human colon adenocarcinoma. The American journal of pathology. Sep 1997;151(3):679-688.
94. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science. Jan 7 2005;307(5706):58-62.
95. Orcutt KD, Rhoden JJ, Ruiz-Yi B, Frangioni JV, Wittrup KD. Effect of small-molecule-binding affinity on tumor uptake in vivo: a systematic study using a pretargeted bispecific antibody. Molecular cancer therapeutics. Jun 2012;11(6):1365-1372.
96. Valencia CA, Rhodenizer D, Bhide S, et al. Assessment of target enrichment platforms using massively parallel sequencing for the mutation detection for congenital muscular dystrophy. J Mol Diagn. May-Jun 2012;14(3):233-246.
97. Munn LL. Aberrant vascular architecture in tumors and its importance in drug-based therapies. Drug discovery today. May 1 2003;8(9):396-403.
98. Harrington KJ, Mohammadtaghi S, Uster PS, et al. Effective targeting of solid tumors in patients with locally advanced cancers by radiolabeled pegylated liposomes. Clinical cancer research : an official journal of the American Association for Cancer Research. Feb 2001;7(2):243-254.
99. Koukourakis MI, Koukouraki S, Giatromanolaki A, et al. Liposomal doxorubicin and conventionally fractionated radiotherapy in the treatment of locally advanced non-small-cell lung cancer and head and neck cancer. Journal of Clinical Oncology. Nov 1999;17(11):3512-3521.
Chapter 2: Challenges facing Block Copolymer Micelles 75
100. Chauhan VP, Stylianopoulos T, Martin JD, et al. Normalization of tumour blood vessels improves the delivery of nanomedicines in a size-dependent manner. Nat Nanotechnol. Jun 2012;7(6):383-388.
101. Seynhaeve AL, Hoving S, Schipper D, et al. Tumor necrosis factor alpha mediates homogeneous distribution of liposomes in murine melanoma that contributes to a better tumor response. Cancer Res. Oct 1 2007;67(19):9455-9462.
102. Lammers T, Peschke P, Kuhnlein R, et al. Effect of radiotherapy and hyperthermia on the tumor accumulation of HPMA copolymer-based drug delivery systems. Journal of controlled release : official journal of the Controlled Release Society. Feb 26 2007;117(3):333-341.
103. Reinhold H, Endrich B. Tumour microcirculation as a target for hyperthermia. International Journal of Hyperthermia. 1986;2(2):111-137.
104. Jain RK, Ward-Hartley K. Tumor blood flow- characterization, modifications, and role in hyperthermia. IEEE TRANS. SONICS ULTRASONICS. 1984;31(5):504-526.
105. Song C, Park H, Lee C, Griffin R. Implications of increased tumor blood flow and oxygenation caused by mild temperature hyperthermia in tumor treatment. International Journal of Hyperthermia. 2005;21(8):761-767.
106. Park JS, Qiao L, Su ZZ, et al. Ionizing radiation modulates vascular endothelial growth factor (VEGF) expression through multiple mitogen activated protein kinase dependent pathways. Oncogene. May 31 2001;20(25):3266-3280.
107. Chung YL, Jian JJ, Cheng SH, et al. Sublethal irradiation induces vascular endothelial growth factor and promotes growth of hepatoma cells: implications for radiotherapy of hepatocellular carcinoma. Clinical cancer research : an official journal of the American Association for Cancer Research. May 1 2006;12(9):2706-2715.
108. Lee YJ, Galoforo SS, Berns CM, et al. Effect of ionizing radiation on AP-1 binding activity and basic fibroblast growth factor gene expression in drug-sensitive human breast carcinoma MCF-7 and multidrug-resistant MCF-7/ADR cells. The Journal of biological chemistry. Dec 1 1995;270(48):28790-28796.
109. Ianiski FR, Alves CB, Souza AC, et al. Protective effect of meloxicam-loaded nanocapsules against amyloid-beta peptide-induced damage in mice. Behavioural brain research. Apr 21 2012;230(1):100-107.
110. Hatch NW, Srodulski SJ, Chan HW, Zhang X, Tannock LR, King VL. Endogenous androgen deficiency enhances diet-induced hypercholesterolemia and atherosclerosis in low-density lipoprotein receptor-deficient mice. Gend Med. Oct 2012;9(5):319-328.
111. Tang T, Thompson JC, Wilson PG, Nelson C, Williams KJ, Tannock LR. Decreased body fat, elevated plasma transforming growth factor-beta levels, and impaired BMP4-like signaling in biglycan-deficient mice. Connect Tissue Res. 2013;54(1):5-13.
112. Seruga B, Pond GR, Hertz PC, Amir E, Ocana A, Tannock IF. Comparison of absolute benefits of anticancer therapies determined by snapshot and area methods. Ann Oncol. Nov 2012;23(11):2977-2982.
113. Heldin CH, Rubin K, Pietras K, Ostman A. High interstitial fluid pressure - An obstacle in cancer therapy. Nat Rev Cancer. Oct 2004;4(10):806-813.
114. Lunt SJ, Fyles A, Hill RP, Milosevic M. Interstitial fluid pressure in tumors: therapeutic barrier and biomarker of angiogenesis. Future Oncol. Dec 2008;4(6):793-802.
Chapter 2: Challenges facing Block Copolymer Micelles 76
115. Chauhan VP, Stylianopoulos T, Boucher Y, Jain RK. Delivery of molecular and nanoscale medicine to tumors: transport barriers and strategies. Annu Rev Chem Biomol Eng. 2011;2:281-298.
116. Fung AS, Jonkman J, Tannock IF. Quantitative Immunohistochemistry for Evaluating the Distribution of Ki67 and Other Biomarkers in Tumor Sections and Use of the Method to Study Repopulation in Xenografts after Treatment with Paclitaxel. Neoplasia. Apr 2012;14(4):324-+.
117. Rhoden JJ, Wittrup KD. Dose dependence of intratumoral perivascular distribution of monoclonal antibodies. Journal of pharmaceutical sciences. Feb 2012;101(2):860-867.
118. White SC, Lorigan P, Margison GP, et al. Phase II study of SPI-77 (sterically stabilised liposomal cisplatin) in advanced non-small-cell lung cancer. Brit J Cancer. Oct 9 2006;95(7):822-828.
119. Rapoport N, Marin A, Luo Y, Prestwich GD, Muniruzzaman M. Intracellular uptake and trafficking of pluronic micelles in drug-sensitive and MDR cells: Effect on the intracellular drug localization. Journal of pharmaceutical sciences. Jan 2002;91(1):157-170.
120. Yoo HS, Park TG. Biodegradable polymeric micelles composed of doxorubicin conjugated PLGA-PEG block copolymer. Journal of Controlled Release. Jan 29 2001;70(1-2):63-70.
121. Du WJ, Xu ZQ, Nystrom AM, Zhang K, Leonard JR, Wooley KL. F-19- and Fluorescently Labeled Micelles as Nanoscopic Assemblies for Chemotherapeutic Delivery. Bioconjugate chemistry. Dec 2008;19(12):2492-2498.
122. Talelli M, Rijcken CJF, Hennink WE, Lammers T. Polymeric micelles for cancer therapy: 3 C's to enhance efficacy. Curr Opin Solid St M. Dec 2012;16(6):302-309.
123. Hoang B, Reilly RM, Allen C. Block copolymer micelles target Auger electron radiotherapy to the nucleus of HER2-positive breast cancer cells. Biomacromolecules. Feb 13 2012;13(2):455-465.
124. Batrakova EV, Kabanov AV. Pluronic block copolymers: Evolution of drug delivery concept from inert nanocarriers to biological response modifiers. Journal of Controlled Release. Sep 10 2008;130(2):98-106.
125. Alakhov VY, Moskaleva EY, Batrakova EV, Kabanov AV. Hypersensitization of multidrug resistant human ovarian carcinoma cells by pluronic P85 block copolymer. Bioconjugate chemistry. Mar-Apr 1996;7(2):209-216.
126. Barenholz Y. Doxil (R) - The first FDA-approved nano-drug: Lessons learned. Journal of Controlled Release. Jun 10 2012;160(2):117-134.
127. Grislain L, Couvreur P, Lenaerts V, Roland M, Deprezdecampeneere D, Speiser P. Pharmacokinetics and Distribution of a Biodegradable Drug-Carrier. International journal of pharmaceutics. 1983;15(3):335-345.
128. Ekdawi SN, Mikhail AS, Stapelton S, et al. Long Circulation and Tumor Accumulation. In: Bae YH, Randall JM, Park K, eds. Cancer Targeted Drug Delivery: Springer New York; 2013:543-571.
129. Prabhakar U, Maeda H, Jain RK, et al. Challenges and key considerations of the enhanced permeability and retention effect for nanomedicine drug delivery in oncology. Cancer Res. Apr 15 2013;73(8):2412-2417.
Chapter 2: Challenges facing Block Copolymer Micelles 77
130. Nishiyama N, Okazaki S, Cabral H, et al. Novel cisplatin-incorporated polymeric micelles can eradicate solid tumors in mice. Cancer Res. Dec 15 2003;63(24):8977-8983.
131. Uchino H, Matsumura Y, Negishi T, et al. Cisplatin-incorporating polymeric micelles (NC-6004) can reduce nephrotoxicity and neurotoxicity of cisplatin in rats. Brit J Cancer. Sep 19 2005;93(6):678-687.
132. Plummer R, Wilson RH, Calvert H, et al. A Phase I clinical study of cisplatin-incorporated polymeric micelles (NC-6004) in patients with solid tumours. Brit J Cancer. Feb 15 2011;104(4):593-598.
133. Emoto S, Yamaguchi H, Kishikawa J, Yamashita H, Ishigami H, Kitayama J. Antitumor effect and pharmacokinetics of intraperitoneal NK105, a nanomicellar paclitaxel formulation for peritoneal dissemination. Cancer Sci. Jul 2012;103(7):1304-1310.
134. Kato K, Chin K, Yoshikawa T, et al. Phase II study of NK105, a paclitaxel-incorporating micellar nanoparticle, for previously treated advanced or recurrent gastric cancer. Invest New Drug. Aug 2012;30(4):1621-1627.
135. Hanahan D, Weinberg RA. Hallmarks of Cancer: The Next Generation. Cell. Mar 4 2011;144(5):646-674.
136. Martin A, Clynes M. Comparison of 5 microplate colorimetric assays for in vitro cytotoxicity testing and cell proliferation assays. Cytotechnology. 1993;11(1):49-58.
137. Shoemaker RH. The NCI60 human tumour cell line anticancer drug screen. Nat Rev Cancer. Oct 2006;6(10):813-823.
138. Lorenzi PL, Reinhold WC, Varma S, et al. DNA fingerprinting of the NCI-60 cell line panel. Molecular cancer therapeutics. Apr 2009;8(4):713-724.
139. Johnson JI, Decker S, Zaharevitz D, et al. Relationships between drug activity in NCI preclinical in vitro and in vivo models and early clinical trials. Brit J Cancer. May 18 2001;84(10):1424-1431.
140. Phillips RM, Bibby MC, Double JA. A critical appraisal of the predictive value of in vitro chemosensitivity assays. Journal of the National Cancer Institute. Sep 19 1990;82(18):1457-1468.
141. Mikhail AS, Eetezadi S, Allen C. Multicellular Tumor Spheroids for Evaluation of Cytotoxicity and Tumor Growth Inhibitory Effects of Nanomedicines In Vitro: A Comparison of Docetaxel-Loaded Block Copolymer Micelles and Taxotere (R). PloS one. Apr 23 2013;8(4).
142. Olive PL, Durand RE. Drug and radiation resistance in spheroids: cell contact and kinetics. Cancer metastasis reviews. Jun 1994;13(2):121-138.
143. Hirschhaeuser F, Menne H, Dittfeld C, West J, Mueller-Klieser W, Kunz-Schughart LA. Multicellular tumor spheroids: An underestimated tool is catching up again. Journal of biotechnology. Jul 1 2010;148(1):3-15.
144. Kim TH, Mount CW, Gombotz WR, Pun SH. The delivery of doxorubicin to 3-D multicellular spheroids and tumors in a murine xenograft model using tumor-penetrating triblock polymeric micelles. Biomaterials. Oct 2010;31(28):7386-7397.
145. Tsukioka Y, Matsumura Y, Hamaguchi T, Koike H, Moriyasu F, Kakizoe T. Pharmaceutical and biomedical differences between micellar doxorubicin (NK911) and liposomal doxorubicin (Doxil). Japanese Journal of Cancer Research. Oct 2002;93(10):1145-1153.
Chapter 2: Challenges facing Block Copolymer Micelles 78
146. Ruggeri BA, Camp F, Miknyoczki S. Animal models of disease: Pre-clinical animal models of cancer and their applications and utility in drug discovery. Biochem Pharmacol. Jan 1 2014;87(1):150-161.
147. Das Thakur M, Pryer NK, Singh M. Mouse tumour models to guide drug development and identify resistance mechanisms. Journal of Pathology. Jan 2014;232(2):103-111.
148. Singh M, Ferrara N. Modeling and predicting clinical efficacy for drugs targeting the tumor milieu. Nature biotechnology. Jul 2012;30(7):648-657.
149. Vassileva V, Moriyama EH, De Souza R, et al. Efficacy assessment of sustained intraperitoneal paclitaxel therapy in a murine model of ovarian cancer using bioluminescent imaging. Brit J Cancer. Dec 9 2008;99(12):2037-2043.
150. Malaney P, Nicosia SV, Dave V. One mouse, one patient paradigm: New avatars of personalized cancer therapy. Cancer Lett. Mar 1 2014;344(1):1-12.
151. Sharpless NE, DePinho RA. Model organisms - The mighty mouse: genetically engineered mouse models in cancer drug development. Nature Reviews Drug Discovery. Sep 2006;5(9):741-754.
152. Hamzah J, Altin JG, Herringson T, et al. Targeted Liposomal Delivery of TLR9 Ligands Activates Spontaneous Antitumor Immunity in an Autochthonous Cancer Model. Journal of Immunology. Jul 15 2009;183(2):1091-1098.
153. Cabral H, Murakami M, Hojo H, et al. Targeted therapy of spontaneous murine pancreatic tumors by polymeric micelles prolongs survival and prevents peritoneal metastasis. Proceedings of the National Academy of Sciences of the United States of America. Jul 9 2013;110(28):11397-11402.
154. Fridlyand J, Simon RM, Walrath JC, et al. Considerations for the successful co-development of targeted cancer therapies and companion diagnostics. Nature reviews. Drug discovery. Oct 2013;12(10):743-755.
155. US Food & Drug Administration. Guidance for industry: enrichment strategies for clinical trials to support approval of human drugs and biological products (draft duidance). 2012; http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM332181.pdf.
156. Lanza GM, Moonen C, Baker JR, Jr., et al. Assessing the barriers to image-guided drug delivery. Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology. Jan-Feb 2014;6(1):1-14.
157. Kim K, Kim JH, Park H, et al. Tumor-homing multifunctional nanoparticles for cancer theragnosis: Simultaneous diagnosis, drug delivery, and therapeutic monitoring. Journal of Controlled Release. Sep 1 2010;146(2):219-227.
158. Mura S, Couvreur P. Nanotheranostics for personalized medicine. Advanced drug delivery reviews. Oct 2012;64(13):1394-1416.
159. Merrimack Pharmaceuticals. Merrimack Announces Encouraging Clinical Data From Expanded Phase 1 Study of MM-302 for the Treatment of Advanced HER2-Positive Breast Cancer. 2013; http://investors.merrimackpharma.com/releasedetail.cfm?ReleaseID=813850. Accessed 17 Feb, 2014.
160. Lammers T, Rizzo LY, Storm G, Kiessling F. Personalized Nanomedicine. Clinical Cancer Research. Sep 15 2012;18(18):4889-4894.
Chapter 2: Challenges facing Block Copolymer Micelles 79
161. Cabral H, Kataoka K. Multifunctional nanoassemblies of block copolymers for future cancer therapy. Sci Technol Adv Mat. Feb 2010;11(1).
162. Nasongkla N, Bey E, Ren JM, et al. Multifunctional polymeric micelles as cancer-targeted, MRI-ultrasensitive drug delivery systems. Nano letters. Nov 8 2006;6(11):2427-2430.
163. Khemtong C, Kessinger CW, Ren JM, et al. In vivo Off-Resonance Saturation Magnetic Resonance Imaging of alpha(v)beta(3)-Targeted Superparamagnetic Nanoparticles. Cancer Res. Feb 15 2009;69(4):1651-1658.
164. Kim D, Gao ZG, Lee ES, Bae YH. In Vivo Evaluation of Doxorubicin-Loaded Polymeric Micelles Targeting Folate Receptors and Early Endosomal pH in Drug-Resistant Ovarian Cancer. Mol Pharmaceut. Sep-Oct 2009;6(5):1353-1362.
165. Tsai HC, Chang WH, Lo CL, et al. Graft and diblock copolymer multifunctional micelles for cancer chemotherapy and imaging. Biomaterials. Mar 2010;31(8):2293-2301.
80
Multicellular Tumor Spheroids for Evaluation of Cytotoxicity and Tumor Growth Inhibitory Effects of Nanomedicines In Vitro: A Comparison of Docetaxel- Loaded Block Copolymer Micelles and Taxotere®
Andrew S. Mikhail, Sina Eetezadi and Christine Allen
Reprint from PLoS ONE 8(4): e62630. (April 2013).
DOI: 10.1371/journal.pone.0062630
Experiments by S. Eetezadi (Fig. 3b, 5, 6, 7, S1, S2, S3) and A.S. Mikhail
(Fig. 3a, 4, 8, 9, 10). Written by A.S. Mikhail with assistance of S. Eetezadi.
Illustrations by S. Eetezadi (Fig. 1, 2). Edited by C. Allen.
This is an open-access article distributed under the terms of the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any
medium, provided the original author and source are credited.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 81
3.1. Abstract
While 3-D tissue models have received increasing attention over the past several decades
in the development of traditional anti-cancer therapies, their potential application for the
evaluation of advanced drug delivery systems such as nanomedicines has been largely
overlooked. In particular, new insight into drug resistance associated with the 3-D tumor
microenvironment has called into question the validity of 2-D models for prediction of in vivo
anti-tumor activity. In this work, a series of complementary assays was established for
evaluating the in vitro efficacy of docetaxel (DTX) -loaded block copolymer micelles
(BCM+DTX) and Taxotere® in 3-D multicellular tumor spheroid (MCTS) cultures. Spheroids
were found to be significantly more resistant to treatment than monolayer cultures in a cell
line dependent manner. Limitations in treatment efficacy were attributed to mechanisms of
resistance associated with properties of the spheroid microenvironment. DTX-loaded
micelles demonstrated greater therapeutic effect in both monolayer and spheroid cultures in
comparison to Taxotere®. Overall, this work demonstrates the use of spheroids as a viable
platform for the evaluation of nanomedicines in conditions which more closely reflect the in
vivo tumor microenvironment relative to traditional monolayer cultures. By adaptation of
traditional cell-based assays, spheroids have the potential to serve as intermediaries between
traditional in vitro and in vivo models for high-throughput assessment of therapeutic
candidates.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 82
3.2. Introduction
It has become increasingly clear that resistance to chemotherapy is not only facilitated by
processes at the cellular level, but also by mechanisms associated with the tumor
microenvironment.1,2 In growing tumors, the heterogeneous architecture of the vasculature,
irregular blood flow, large intervascular distances and nature of the extracellular matrix limit
the access of cells to oxygen, nutrients, and systemically administered therapies.3,4 Within the
tumor interstitium, gradients in the rate of cell proliferation are established wherein rapidly
dividing cells reside close to the tumor vasculature and quiescent cells are situated deep
within the extravascular space. However, many anti-neoplastic agents exert limited toxicity
against slowly- or non-proliferating cells and are less effective in the hypoxic and acidic
microenvironments of poorly perfused tissues.5,6 These therapeutic limitations are
exacerbated by high interstitial fluid pressure which inhibits the penetration of
chemotherapeutic agents through the tumor interstitium by limiting convective transport.7 As
a result cells located distant from blood vessels may be less sensitive to treatment and also
be exposed to sub-therapeutic drug concentrations.
The use of in vitro cell culture is critical in drug discovery and formulation development for
rapid identification of lead candidates and for investigating mechanisms of drug efficacy at
the cellular and molecular levels. In contrast to in vivo tumor models, in vitro cultures are
better suited for systematic studies of formulation parameters in a highly controlled
environment. However, cytotoxic effects observed in conventional monolayer cultures often
fail to translate into similar effects in vivo. 8,9 This is due to the inherent inability of 2-D cultures
to account for mechanisms of drug resistance and transport restrictions associated with the
3-D tumor microenvironment. As such, there is increasing interest in applying 3-D in vitro
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 83
models that enable rapid, high throughput screening of drug formulations for selection of lead
candidates to move forward to in vivo evaluation.10–12
As depicted in Error! Reference source not found., 3-D tissue cultures such as MCTS serve
as an intermediary between the oversimplified structure of monolayer cultures and the highly
complex nature of in vivo tumors. Spheroid cultures possess a complex network of cell-cell
contacts and advanced extracellular matrix development, as well as pH, oxygen, metabolic
and proliferative gradients analogous to the conditions in poorly vascularized and avascular
regions of solid tumors.13–15 In general, a spheroid is comprised of an outer region of
proliferating cells which surrounds intermediate layers of quiescent cells and, if the spheroid
is large enough, a necrotic core. This arrangement parallels the radial organization of tissues
surrounding tumor blood vessels. To date, a variety of 3-D in vitro tissue models have been
applied for the study of anticancer therapies including natural and synthetic tissue
scaffolds,16,17 multicellular layers,18–22 and multicellular tumor spheroids.16,23,24 MCTS are
particularly relevant in the development of nanomedicines since the penetration of the
encapsulated drug in tumor tissues may be significantly altered by properties of the delivery
vehicle. To date, however, there remain limited examples of the use of MCTS for the
evaluation of nanomedicines.25–29 DTX is a potent chemotherapeutic agent that is
administered as Taxotere® (Sanofi-Aventis) and used for treatment of cancers of the breast,
prostate, lung, head and neck, and stomach.30 DTX is also being investigated in a phase II
clinical trial for treatment of metastatic colorectal adenocarcinoma in combination with
gemcitabine and has been investigated as a single agent for treatment of cervical cancer.31,32
However, Taxotere® is known to be associated with significant side effects that can require
reduction of the administered dose.33 Encapsulation of chemotherapeutic agents within
biocompatible nanosystems such as block copolymer micelles (BCMs) has proven to be a
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 84
promising approach for mitigating the burden of toxicity on normal tissues and increasing
tumor-specific drug accumulation.34 The primary objective of this study was to adapt and
apply traditional cell-based assays in a systematic and complementary manner for the
evaluation of Taxotere® and a DTX-containing nanomedicine in both monolayer and MCTS
cultures (Figure 3-2).
Figure 3-1: 3-D cultures as intermediary between 2-D cultures and animal models. Intermediate
in complexity, 3-D cultures permit the systematic, high-throughput assessment of formulation
properties in a controlled environment that approximates important properties of in vivo tumors
in the absence of complex parameters which may confound data interpretation.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 85
3.3. Materials and Methods
3.3.1. Materials
Methoxy poly(ethylene glycol) (CH3O-PEG-OH; Mn = 5000, Mw/Mn = 1.06) was obtained
from Sigma-Aldrich (Oakville, ON, Canada). e-Caprolactone and dichloromethane (Sigma-
Aldrich) were dried using calcium hydride prior to use. Hydrogen chloride (HCl) (1.0 M in
diethyl ether), N,N-dimethylformamide (DMF), diethyl ether, hexane and acetonitrile (Sigma-
Aldrich) were used without further purification. Alexa Fluor 488 (AF488) carboxylic acid
succinimidyl ester was purchased from Molecular Probes (Eugene, OR). The hypoxia marker,
EF5, and Cy5-conjugated anti-EF5 antibody were purchased from the Department of
Radiation Oncology, University of Pennsylvania, (Philadelphia, PA). DTX was purchased from
Jari Pharmaceutical Co. (Jiangsu, China).
3.3.2. Synthesis of CH3O-PEG-b-PCL (PEG-b-PCL) Copolymers
PEG-b-PCL copolymer was prepared as previously described.35 Briefly, CH3O-PEG-OH was
used to initiate the ring- opening polymerization of e-CL in the presence of HCl. The reaction
was carried out for 24 h at room temperature prior to termination by addition of triethylamine
(TEA) and precipitation in diethyl ether and hexane (50:50, v/v%). The product was dried under
vacuum at room temperature.
3.3.3. Preparation and Characterization of BCM+DTX
PEG-b-PCL copolymers and DTX were dissolved at a copolymer:drug weight ratio of 20:1
in DMF and stirred for 30 min. DMF was evaporated under N2 at 30 C and residual solvent
was removed under vacuum. Dry copolymer-drug films were then heated to 60 C in a water
bath prior to the addition of PBS buffer (pH 7.4) at the same temperature. Resultant micelle
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 86
solutions were vortexed, stirred for 24 h at room temperature and finally sonicated
(Laboratory Supplies Co., NY) for 1 h. Undissolved drug crystals were removed by
centrifugation at 4400 g for 12 min (Eppendorf 5804R). The final copolymer concentration
was 10 mg/mL. The amount of physically entrapped DTX in BCM samples was determined
by HPLC analysis (Agilent series 1200) with UV detection (Waters 2487) at a wavelength of
227 nm. An XTerra C18 reverse phase column was employed with ACN/ water (60/40, v/v%)
as the mobile phase. Drug loading was quantified using a calibration curve generated from a
series of DTX standards.
3.3.4. Sizing of BCM+DTX
The average hydrodynamic diameter of the BCMs was determined by dynamic light
scattering (DLS) using a 90Plus Particle Size Analyzer (Brookhaven Instruments Corp.,
Holtsville, NY) at an angle of 90u and temperature of 37 C. The samples were diluted to a
copolymer concentration of 0.5 mg/mL prior to measurement. Analysis was performed using
the 90Plus Particle Sizing Software.
3.3.5. Transmission Electron Microscopy (TEM)
BCMs were observed by TEM using a Hitachi 7000 microscope operating at an
acceleration voltage of 75 kV (Schaumburg, IL). Samples were diluted in double distilled
water immediately prior to analysis and negatively stained with a 1% uranyl acetate (UA)
solution. The final copolymer concentration was 0.5 mg/mL. The samples were then
deposited on copper grids that had been pre coated with carbon and negatively charged
(Ted Pella Inc., Redding, CA) and briefly air-dried prior to analysis.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 87
3.3.6. Drug Release
The release of DTX from BCMs and Taxotere® was analyzed using a dialysis method.
Aliquots (1 mL) of BCM+DTX, DTX in DMSO, and Taxotere® were placed in individual dialysis
bags (MWCO 2 kDa, Spectra/Por, Rancho Dominguez, CA) and dialyzed separately against
100 mL of PBS at pH 7.4 in an incubator at 37 C ensuring that sink conditions were
maintained. At selected timepoints, 50 mL samples were withdrawn from the dialysis bags
and DTX content was measured by HPLC as described above.
3.3.7. Tissue Culture and Growth of MCTS
Human cervical (HeLa) and colon (HT29) (ATCC, Manassas, VA) cancer cells were
incubated at 37 C and 5% CO2 in DMEM containing 1% penicillin-streptomycin and
supplemented with 10% FBS. For growth of MCTS, cells were suspended using trypsin-EDTA
and 2000 and 5000 HT29 and HeLa cells were seeded onto non-adherent 96-well round-
bottomed Sumilon PrimeSurfaceTM plates (Sumitomo Bakelite, Tokyo, Japan), respectively, in
200 µL of media per well. During growth, 50% of the media was exchanged every other day.
MCTS were grown for 7 days until they reached 500 µm in diameter before use. For
determination of cell packing density 12 MCTS per seeding density were disaggregated at
day 7 using trypsin-EDTA. The cell density in MCTS was determined by dividing the total
number of cells per spheroid by the volume of the spheroid as determined by image analysis
(slope of the linear regression).
3.3.8. Immunohistochemical Analysis of MCTS
MCTS were washed in PBS and transferred onto a vinyl specimen mold (Cryomold®,
Tissue-Tek, Sakura Finetek, CA) prior to addition of Tissue-Tek® O.C.T. compound (Sakura
Finetek, Torrance, CA). MCTS were then submersed in an isopentane bath cooled by liquid
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 88
nitrogen, cut into 5 µm thick sections using a microtome and mounted on glass slides.
Histological staining was conducted for the identification of cellular proliferation (Ki67) and
stained with hematoxylin and eosin (H&E). For identification of hypoxic regions, MCTS were
incubated with 0.5 mM EF5 and soaked in PBS prior to cryosectioning. EF5 in the MCTS
sections was identified by binding with cyanine-5-conjugated mouse anti-EF5 (1/50) anti-
body. The positive signal distribution for Ki67 was analyzed using a customized MATLAB®
algorithm, as described previously.36 Briefly, images containing Ki67-stained MCTS sections
were thresholded for positive color intensity. Using a distance map, signal intensities were
summed within three concentric regions of equidistant thickness (periphery, intermediate and
core), each equivalent to 1/3 of the MCTS radius. The distribution of Ki67 positive signal is
expressed as a percentage of total positive signal in the MCTS section.
Figure 3-2: In vitro assays used in this study for analysis of formulation efficacy in spheroids.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 89
3.3.9. Measurement of MCTS Growth
Spheroids were imaged using a light microscope with a 106 objective lens (VWR
VistaVisionTM) connected to a digital camera (VWR DV-2B). The diameter and volume of
MCTS were determined by measuring their cross-sectional area using an automated image
analysis macro developed for use with the ImageJ software package (NIH, Bethesda, MD,
Version 1.44 m). The automated method was validated by comparison to manual
determination of spheroid diameter and volume (Figure 3-12). For the automated method,
images were converted into 8-bit greyscale and the perimeter of an individual MCTS was
recognized by an automated threshold function and the image converted to a 2-D mask. The
area of the spheroid mask was recorded, applying an image of known scale as calibration.
Finally, the volume of the MCTS was calculated by assuming a spherical shape as follows: V
= 4/3*π*(d/2)3. Data was fit using the Gompertz equation for tumor growth as follows: V(t) =
V(0)exp(α/β(1-exp(-β *t))) where V(t) is volume at time t, V(0) the initial volume and α and β
are constants.37
3.3.10. Cytotoxicity in Monolayer and Spheroid Tissue Cultures
The cytotoxicity of BCM+DTX and Taxotere® in monolayer and spheroid cell cultures was
determined using the established acid phosphatase (APH) assay which is based on
quantification of cytosolic acid phosphatase activity.38 For this assay, p-nitrophenyl
phosphate is added in cell culture and hydrolyzed in viable cells to p-nitrophenol via
intracellular acid phosphatase. Briefly, MCTS (one spheroid per well) and monolayer cultures
(4000 cells per well) were treated with Taxotere® or BCM+DTX for 24 h over a range of drug
concentrations. Following treatment, monolayers and spheroids were washed three times
with fresh media and cultured for an additional 48 h. Monolayers and spheroids were then
washed with PBS buffer prior to the addition of 100 µL of freshly prepared reaction buffer (2
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 90
mg/ml p-nitrophenyl phosphate (Sigma) and 0.1% v/v Triton-X-100 in 0.1 M sodium acetate
buffer at pH 5.5). Following incubation for 2 h in the cell incubator, 10 µL 1 M sodium
hydroxide was added to each well and cell viability was determined by measuring the UV
absorbance at 405 nm using an automated 96-well plate reader (SpectraMax Plus 384,
Molecular Devices, Sunnyvale, CA). Results were normalized to controls as follows: %viability
= (Atreatment – Amedia)/(Acontrol – Amedia), where A = mean absorbance. All experiments
were performed in triplicate.
3.3.11. Growth Inhibition of MCTS
BCM+DTX or Taxotere® was administered to spheroids for 24 h at a DTX equivalent
concentration of 2, 20 or 200 ng/ mL. The culture media was replaced following the incubation
period. Subsequently, half of the culture media was replaced by pipette every other day.
Images of spheroids were captured using a light microscope with a 106 objective lens (VWR
VistaVisionTM) connected to a digital camera (VWR DV-2B). Spheroid size was determined by
measuring their 2-D cross-sectional area using the automated image analysis method
described previously. The data are reported as the mean volume of six spheroids 6 SD.
3.3.12. Clonogenic Survival Assay
The clonogenic assay was used to determine the ability of single cells to replicate and form
colonies (50 cells) following exposure to BCM+DTX and Taxotere®. Single cell suspensions
derived from monolayer and disaggregated spheroids were diluted in culture media and cells
were plated in 6-well plates in desired numbers. MCTS were disaggregated by incubation in
trypsin-EDTA for 10 min, followed by gentle agitation. Drug formulations were added
immediately at a DTX equivalent concentration of 20 ng/mL. After treatment for 24 h, cells
were washed with PBS and 2 mL of fresh media was added to each well. For treatment of
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 91
intact spheroids, drug formulations were added directly into wells containing individual MCTS.
After 24 h, MCTS were collected and rinsed in PBS, suspended as single-cell suspensions in
fresh media following trypsinization, and seeded onto 6-well plates. Cells were incubated for
14–16 days prior to fixation with methanol and staining with 1% crystal violet solution.
Colonies consisting of at least 50 cells were counted. The surviving fraction (SF) was
expressed as the number of colonies divided by the product of the number of cells plated and
the plating efficiency. The plating efficiency was determined by dividing the number of
colonies formed by the number of cells plated for untreated controls.
Figure 3-3: Characterization of micelle morphology and size. a) Transmission electron
micrograph (Scale bar in represents 100 nm) and b) size distribution of BCM+DTX as
determined by dynamic light scattering at 37 °C.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 92
3.4. Results
3.4.1. Characterization of BCM+DTX
PEG-b-PCL copolymer micelles containing physically encapsulated DTX were formulated
with a spherical morphology (Figure 3-3, a). The size distribution of the micelles was
monomodal with an average hydrodynamic diameter of 49.2 ± 2.3 nm (Figure 3-3, b). Drug
loading resulted in a final DTX equivalent concentration of 258.7 ± 35.5 µg/mL at a loading
efficiency of 52.7 ± 7.1%. Release of DTX from BCMs occurred over the course of 24 h wherein
74 % of the drug was released by 12 h. In contrast, the release of docetaxel from Taxotere®
was complete by 12 h (Figure 3-4).
Figure 3-4: Drug release. Release of docetaxel from dialysis bags containing
BCM+DTX, Taxotere®, and DTX in DMSO, n = 3
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 93
3.4.2. Growth of MCTS
Spheroids were grown using a modified liquid overlay technique by seeding HT29 or HeLa
cells onto non-adherent U-bottom tissue culture wells without the use of an agarose surface
coating. MCTS were spherical, followed a sigmoidal growth profile, and were grown until a
diameter of ~500 µm was reached prior to use (Figure 3-5).
3.4.3. Cytotoxicity in Monolayer and MCTS Culture
Cell viability following exposure to BCM+DTX or Taxotere® was assessed using the APH
assay (Figure 3-6). This assay was validated by assessment of the relationship between UV
absorbance and cell number in both monolayer and spheroid cultures. As shown in Figure
3-12, a linear relationship was obtained. A well-established tetrazolium salt-based assay
(WST-8) was also evaluated and did not yield a similar correlation (Figure 3-13). Spheroid
cultures were substantially less sensitive to BCM+DTX and Taxotere® relative to their
monolayer counterparts. HeLa cells were less responsive to treatment with either BCM+DTX
or Taxotere® than HT29 cells in monolayer culture. However, in spheroid culture, HT29 cells
were less sensitive to treatment. The IC50 of HeLa and HT29 monolayer cultures treated with
BCM+DTX were 0.37 ± 0.01 and 0.01 ± 0.004 ng/mL, respectively. When treated with
Taxotere®, the IC50 of HeLa and HT29 monolayer cultures were 2.2 ± 0.5 and 0.09 ± 0.01 ng/
mL, respectively. The IC50 of HeLa cells cultured as MCTS was 139 ± 198 ng/mL for
BCM+DTX and 1558 ± 103 ng/mL for Taxotere® whereas HT29 MCTS maintained a viability
above 80% at all drug concentrations.
3.4.4. Inhibition of MCTS Growth
MCTS volume was plotted over a 30 day period following a 24 h incubation with 2, 20, and
200 ng/mL BCM+DTX or Taxotere® (Figure 3-7). The growth of HeLa MCTS was completely
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 94
impeded following incubation with DTX concentrations of 20 and 200 ng/mL. No significant
difference in growth was observed following exposure to 2 ng/mL of DTX relative to untreated
controls. In the case of HT29 MCTS, incubation with 20 ng/mL of BCM+DTX and Taxotere®
only resulted in a partial reduction in MCTS volume. Similarly to HeLa MCTS, complete
inhibition of growth was observed following incubation with 200 ng/mL of drug. Unlike HeLa
MCTS, however, a slight growth delay was also observed at 2 ng/mL. Interestingly, following
re-treatment on day 14 at a DTX concentration of 20 ng/mL, BCM+DTX demonstrated greater
inhibition of spheroid growth in HT29 cultures than Taxotere®.
Figure 3-5: Spheroid packing density and growth. a) Cells per HeLa and HT29 spheroid of given volume,
n = 12. b) Growth of HeLa and HT29 spheroids, n = 6. Data was fit using the Gompertz equation for
tumor growth. The dashed lines indicate spheroid properties used in the studies.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 95
3.4.5. Immunohistochemistry
Immunohistochemical analysis of MCTS cross-sections was performed in order to identify
regions of necrosis, cellular proliferation and hypoxia (Figure 3-8). Staining with the pro-
liferation marker Ki67 revealed a greater proportion of proliferative cells in HeLa MCTS relative
to HT29. Quantitative image analysis revealed that 88.6% of proliferating cells were located
within the periphery of HT29 MCTS (Figure 3-9). In contrast, only 51% of the total proliferating
cells were located in the periphery of HeLa MCTS and 25% and 24% were located in the
intermediate region and core, respectively. Signs of necrosis were visible following staining
with H&E in HT29 MCTS. Incubation of MCTS with EF5 allowed for identification of regions of
hypoxia following exposure to Cy5-conjugated anti-EF5 antibody. Hypoxic conditions were
observed primarily in the core and intermediate regions of HT29 MCTS. In contrast, HeLa
MCTS did not demonstrate any regional hypoxia. The relative distributions of cellular
proliferation, hypoxia and necrosis in the MCTS are summarized in Figure 3-10.
3.4.6. Clonogenic Survival
The surviving fractions (SF) of HeLa and HT29 cells were determined following treatment
with BCM+DTX or Taxotere® as monolayer and MCTS cultures (Figure 3-10). The SF was
significantly higher for all intact MCTS cultures relative to monolayers. HeLa cells were less
sensitive to treatment than HT29 when cultured as monolayers, but more sensitive than HT29
cells when the cells were exposed to treatment as MCTS. In all cases, the SF was significantly
lower when treated with BCM+DTX compared to Taxotere®. Furthermore, cells exposed to
treatment immediately following MCTS disaggregation demonstrated residual resistance to
both BCM+DTX and Taxotere®.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 96
3.5. Discussion
In recent years, the tumor microenvironment has been implicated in the coordination of
tumor growth, metastasis and resistance to anti-cancer therapies.39,40 As such, effective
evaluation of novel therapeutic agents requires the use of tissue models which closely mimic
native conditions within the intratumoral space. Yet, the vast majority of chemotherapeutic
agents are screened for cytotoxic effects in monolayer cultures which do not account for
critical mechanisms of drug resistance associated with the tumor microenvironment.
Consequently, these models poorly predict a drug’s therapeutic efficacy in vivo.8 In contrast,
3-D MCTS better approximate the state of cancer cells in their native environment and thus
can be used to more accurately estimate a drug’s therapeutic potential. A variety of methods
have been used to grow MCTS for use in cancer research including spinning culture flasks,41
hanging drops,42 liquid overlay on agarose,43 micropatterned plates,44 and recently, using inter-
cellular linkers.45 However, many of these techniques are impractical, time-consuming, and
involve delicate handling procedures, limiting the use of the MCTS model in drug screening
and development. In addition, practical application of traditional cell-based assays in MCTS
cultures remains poorly established. In the current study, the performance of BCM+DTX and
Taxotere® was evaluated by adaptation of conventional cytotoxicity and survival assays in
monolayer and MCTS cultures using a robust MCTS culture technique.
MCTS grew according to sigmoidal growth patterns reflective of tumor growth in vivo
(Figure 3-5) and possessed histological features similar to those of the native tumor
microenvironment including gradients in cell proliferation and regions of hypoxia and necrosis
(Figure 3-8, Figure 3-14). Cells grown in spheroid cultures demonstrated considerably greater
resistance to treatment with BCM+DTX or Taxotere® relative to cells grown in monolayer
cultures. This may be a result of the limited exposure of cells within MCTS to treatment due
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 97
to poor penetration of DTX or BCMs, the limited sensitivity of cells within MCTS to DTX due
to a reduction in cellular proliferation and/or resistance associated with 3-D cell adhesion (i.e.
contact effect). In a study by Kyle et al., the penetration half-depth (the depth from the surface
at which the amount of drug falls to half of its maximum concentration) of DTX in multicellular
layers was found to be <25 µm following a 2 h incubation at a concentration of 0.3 µM.46 Peak
tissue levels did not increase proportionally following a 10-fold increase in drug concentration
although the depth of penetration was improved indicating partial saturation of tissue binding.
Therefore, it is likely that high intracellular binding and consumption of DTX by peripheral cells
in the MCTS limits the toxicity to cells distant from the surface. For drugs which are rapidly
consumed by cells, encapsulation in BCMs which minimize interactions and uptake by cells
may improve drug penetration.47 For example, Pun et al. reported ameliorated penetration of
doxorubicin into MCTS when encapsulated in triblock copolymer micelles.25 However, BCMs
which penetrate poorly through tissues may limit the penetration of the encapsulated drug.
Overall, the extent to which the BCMs influence drug penetration will depend on the relative
rates of drug release and BCM penetration in the MCTS. We have previously found that PEG-
b-PCL BCMs of 55 nm diameter can achieve a homogeneous distribution in MCTS of HeLa
and HT29 cell lines following a 24 h incubation (unpublished data).
In addition to potential limitations in MCTS penetration associated with the drug and
BCMs, the discrepancy between MCTS and monolayer cytotoxicity may also be a result of
drug resistance imparted by the MCTS microenvironment. A marked decrease in the
proportion of proliferating cells was observed in MCTS with increasing depth from the surface
(Figure 3-9). Since DTX exerts its therapeutic effect on cycling cells, cells located near the
MCTS surface will respond to treatment similarly to cells cultured as monolayers. By contrast,
quiescent cells that are located in the intermediate and core regions of the MCTS will be less
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 98
sensitive to treatment. This notion is supported by the observation that cells exposed to
treatment immediately following disaggregation of MCTS demonstrated greater clonogenic
survival than monolayer cells, but less than cells treated as intact MCTS. Therefore, there
exists a population of cells within the MCTS that is more resistant to treatment than cells
cultured as monolayers even in the absence of any physical barrier to drug penetration. As
such, the limited sensitivity of MCTS to treatment is likely a result of both restricted transport
and mechanisms of drug resistance associated with the MCTS microenvironment.
Figure 3-6: .Cytotoxicity of Taxotere® and BCM+DTX in spheroid and monolayer cultures. Viability of a)
HeLa and b) HT29 cells cultured as monolayers and spheroids as measured using the APH assay. Data
is expressed as the percent viability relative to untreated controls and fit to the Hill equation. c)
Cytotoxicity of blank PEG-b-PCL micelles as a function of copolymer concentration. Each plot represents
the mean of three independent experiments 6 SD (n = 3).
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 99
The extent to which culturing cells as MCTS influenced the therapeutic effect of BCM+DTX
and Taxotere® relative to monolayers was found to be cell-line specific. In monolayer cultures,
BCM+DTX and Taxotere® demonstrated greater cytotoxicity against HT29 cells relative to
HeLa cells. In contrast, culturing cells as MCTS imparted a greater enhancement in
therapeutic resistance (i.e. greater increase in IC50) to HT29 cells than to HeLa cells. We have
previously shown significantly greater penetration of BCMs into HeLa MCTS than HT29 MCTS
due to the former’s lower cell packing density and large intercellular channels (unpublished
data). In the current study, significant cell line-dependent differences in MCTS
microenvironment were observed. Limited permeability of HT29 MCTS and/or high
consumption of oxygen by peripheral cells was reflected by the presence of central hypoxia
and necrosis. Importantly, HT29 MCTS contained a greater proportion of non-proliferating
cells relative to HeLa MCTS. It is likely that some quiescent cells within the MCTS retained
their clonogenic potential following exposure to sub-therapeutic amounts of DTX and were
capable of recommencing proliferative activity when replated as monolayers. The greater
clonogenic potential of HeLa cells following disaggregation of MCTS relative to HT29 cells
likely reflects the greater sensitivity of HT29 monolayer cells to DTX rather than greater
residual resistance of MCTS-derived HeLa cells.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 100
Figure 3-7: Inhibition of spheroid growth. a) Sequential images of the same HeLa and HT29 spheroids
following treatment with BCM+DTX at a concentration of 20 ng/mL. Bars represent 100 µm. Growth
inhibition of HeLa (b,c) and HT29 (d,e) MCTS by BCM+DTX and Taxotere® at concentrations of 2, 20 and
200 ng/mL. Cells were re-treated after two weeks (arrow). Box represents expanded region of plots b)
and d). Data is expressed as the mean volume of six spheroids (n = 6) ± SD.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 101
One of the important advantages of the MCTS model is that it allows for treatment efficacy
to be observed over an extended period of time. In order to evaluate the potential of surviving
cells to repopulate MCTS, the growth of MCTS following treatment with BCM+DTX and
Taxotere® was evaluated for 28 days with treatment re-applied after 14 days. The results of
this study demonstrate both dose- and time-dependent changes in MCTS growth following
incubation with the drug formulations. Near complete elimination of HeLa MCTS was
observed following treatment at 20 ng/mL or greater with either BCM+DTX or Taxotere®. In
contrast, only partial growth inhibition was observed in HT29 MCTS when exposed to the
same concentration. This observation is consistent with the results obtained from the
cytotoxicity and clonogenic assays in which HT29 MCTS demonstrated greater resistance to
treatment relative to HeLa MCTS. A slight inhibitory effect in HT29 MCTS following
administration of DTX formulations at 2 ng/mL was likely due to the cytotoxicity and shedding
of surface cells, consistent with the response of HT29 cells to treatment in monolayer
cultures. In addition, the apparent discrepancy between the limited cytotoxicity in HT29
spheroids revealed using the AP® assay (measured 2 days post drug incubation) and the
marked growth inhibition at 20 ng/mL is consistent with the observed 4 day delay in growth
inhibitory effect. Interestingly, little difference in spheroid growth inhibition was observed
between BCM+DTX and Taxotere® following initial treatment. It should be noted, however,
that following retreatment after 14 days of culture, BCM+DTX demonstrated a greater growth
inhibitory effect relative to Taxotere®.
Several factors may have contributed to the greater cytotoxicity of BCM+DTX relative to
Taxotere® in monolayer and MCTS cultures. It has been hypothesized that DTX is taken up
more rapidly by cells following release from BCMs in close proximity to the cell membrane
due to an increase in the local transmembrane concentration gradient.48–50 Slower efflux of
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 102
BCM-encapsulated DTX relative to free DTX, by avoidance of membrane efflux pumps, may
also contribute to the greater therapeutic effect of the DTX-loaded BCMs.51–53 While these
results are promising, further investigation is required to fully elucidate the mechanism of
cytotoxicity that lead to enhanced therapeutic effects of BCM+DTX relative to Taxotere® in
vitro.
Figure 3-8: Histological assessment of spheroid microenvironment. HeLa (a–c) and HT29 (d–f)
MCTS cross-sections stained with H&E (a, d), Ki67 proliferation marker (b, e) and EF5 (c, f), a marker
of hypoxia. Scale bars represent 100 µm. g) Properties of the spheroid microenvironment and their
spatial distribution. ‘‘++’’, ‘‘+’’, and ‘‘–’’, indicate high, intermediate and low levels of the corresponding
feature, respectively.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 103
Overall, as outlined in Figure 3-2, each of the three assays employed in this study is unique
and together they provide complementary information on the therapeutic potential of drug
formulations. Importantly, comparison of results obtained in monolayer and spheroid cultures
demonstrated the important influence of the microenvironment and 3-D tissue structure on
formulation efficacy. Therefore, 3-D cultures such as MCTS may serve as important tools for
investigating the performance of nanomedicines in environments that more closely mimic
intratumoral conditions in vivo. However, while spheroids share several important structural
and microenvironmental properties with native tumors, there are important differences which
may limit the extent to which this in vitro model can be used to predict drug efficacy in vivo.
Notably, the MCTS model does not account for the potential influence of convective flow or
presence of stromal cells on drug and nanoparticle transport. Despite these limitations,
evaluation of formulation efficacy in spheroids rather than monolayer cultures is expected to
more accurately reflect therapeutic performance in vivo.
Figure 3-9: Spatial distribution of proliferating cells in spheroids. Ki67 positive signal distribution
relative to radial position in a) HeLa and b) HT29 MCTS as a percent of total positive stain, n = 6
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 104
Figure 3-10: Clonogenic potential of cells following treatment. Clonogenic survival of HeLa
and HT29 cells following 24 h treatment with 20 ng/mL of BCM+DTX or Taxotere® as a)
monolayers, b) disaggregated spheroids and c) intact spheroids.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 105
3.6. Supporting Information
Figure 3-11: Measurement of spheroid volume. a) Schematic representation of the analysis
process using a macro developed for ImageJ (version 1.44 m). b) Correlation between manual
and automated volume measurements of HeLa MCTS. MCTS were imaged at selected
intervals of growth. Manual measurement of MCTS volume was performed by determining
the average of the largest and smallest diameters using the captured images and assuming
a spherical MCTS morphology. Automated volume measurement was achieved using an
image recognition technique in ImageJ. Firstly, MCTS images were converted into 8-bit
greyscale and the perimeter of the MCTS was recognized by an automated threshold function.
The area of the 2-D MCTS mask was recorded and converted to µm2 by calibration using an
image of known scale and subsequently used to calculate the volume.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 106
Figure 3-12: Validation of the acid phosphatase (APH) assay. Results from the APH assay
using HeLa (left column) and HT29 cells (right column) grown as spheroids (top row) and
monolayers (bottom row) demonstrate a linear relationship between cell number and UV
absorption at 405 nm. Each data point represents the mean of three independent
experiments 6 SD (n=3).
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 107
Figure 3-13: Failure of WST-8 assay. Results from the WST-8 assay demonstrate a
non-linear correlation between the number of cells and OD450 in spheroid culture.
Figure 3-14: Fluorescence images of HT29 (a) and HeLa (b) tumor xenografts displaying
markers of hypoxia (EF5 - blue) and blood vessels (CD31 - red). Scale bars represent 100
µm.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 108
3.7. References
1. Tre´dan O, Galmarini CM, Patel K, Tannock IF (2007) Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst 99: 1441–1454.
2. Swartz MA, Iida N, Roberts EW (2012) Tumor Microenvironment Complexity: Emerging Roles in Cancer Therapy. Cancer Res 72: 2473–2480.
3. Jain RK, Stylianopoulos T (2010) Delivering nanomedicine to solid tumors. Nat Rev Clin Oncol 7: 653–664.
4. Minchinton AI, Tannock IF (2006) Drug penetration in solid tumours. Nat Rev Cancer 6: 583–592.
5. Vaupel P, Mayer A (2007) Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev 26: 225–239.
6. Kim JW, Ho WJ, Wu BM (2011) The role of the 3D environment in hypoxia- induced drug and apoptosis resistance. Anticancer Res 31: 3237–3245.
7. Chauhan VP, Stylianopoulos T, Boucher Y, Jain RK (2011) Delivery of Molecular and Nanoscale Medicine to Tumors: Transport Barriers and Strategies. Annu Rev Chem Biomol Eng 2: 281–298.
8. Johnson JI, Decker S, Zaharevitz D, Rubinstein L V, Venditti JM, et al. (2001) Relationships between drug activity in NCI preclinical in vitro and in vivo models and early clinical trials. Br J Cancer 84: 1424–1431.
9. Phillips R, Bibby M, Double J (1990) A Critical Appraisal of the Predictive Value of in Vitro Chemosensitivity Assays. J Natl Cancer Inst 82: 1457–1463.
10. Elliott N (2010) A review of three-dimensional in vitro tissue models for drug discovery and transport studies. J Pharm Sci 100: 2–8.
11. Nyga A, Cheema U, Loizidou M (2011) 3D tumour models: novel in vitro approaches to cancer studies. J Cell Commun Signal 5: 239–248.
12. Vinci M, Gowan S, Boxall F, Patterson L, Zimmermann M, et al. (2012) Advances in establishment and analysis of 3D tumour spheroid-based functional assays for target validation and drug evaluation. BMC Biol 10: 1–20.
13. Rotin D, Robinson B, Tannock IF (1986) Influence of hypoxia and an acidic environment on the metabolism and viability of cultured cells: potential implications for cell death in tumors. Cancer Res 46: 2821–2826.
14. Nederman T, Norling B, Glimelius B, Carlsson J, Brunk U (1984) Demonstration of an Extracellular Matrix in Multicellular Tumor Spheroids. Cancer Res 44: 3090–3097.
15. Acker H, Carlsson J, Mueller-Klieser W, Sutherland RM (1987) Comparative pO2 measurements in cell spheroids cultured with different techniques. Br J Cancer 56: 325–327.
16. Fischbach C, Chen R, Matsumoto T, Schmelzle T, Brugge JS, et al. (2007) Engineering tumors with 3D scaffolds. Nat Methods 4: 6–11.
17. Horning JL, Sahoo SK, Vijayaraghavalu S, Dimitrijevic S, Vasir JK, et al. (2008) 3-D tumor model for in vitro evaluation of anticancer drugs. Mol Pharm 5: 849– 862.
18. Hicks KO, Ohms SJ, Van Zijl PL, Denny WA, Hunter PJ, et al. (1997) An experimental and mathematical model for the extravascular transport of a DNA intercalator in tumours. Br J Cancer 76: 894–903.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 109
19. Tannock IF, Lee CM, Tunggal JK, Cowan DSM, Egorin MJ (2002) Limited penetration of anticancer drugs through tumor tissue: a potential cause of resistance of solid tumors to chemotherapy. Clinical cancer research: an official journal of the American Association for Cancer Res 8: 878–884.
20. Cowan DS, Tannock IF (2001) Factors that influence the penetration of methotrexate through solid tissue. Int J Cancer 91: 120–125.
21. Chetprayoon P, Kadowaki K, Matsusaki M, Akashi M (2012) Survival and structural evaluations of three-dimensional tissues fabricated by the hierarchical cell manipulation technique. Acta Biomater in press.
22. Hosoya H, Kadowaki K, Matsusaki M, Cabral H, Nishihara H, et al. (2012) Engineering fibrotic tissue in pancreatic cancer: A novel three-dimensional model to investigate nanoparticle delivery. Biochem Biophys Res Commun 419: 32–37.
23. Grantab R, Sivananthan S, Tannock IF (2006) The penetration of anticancer drugs through tumor tissue as a function of cellular adhesion and packing density of tumor cells. Cancer Res 66: 1033–1039.
24. Hirschhaeuser F, Menne H, Dittfeld C, West J, Mueller-Klieser W, et al. (2010) Multicellular tumor spheroids: an underestimated tool is catching up again. J Biotechnol 148: 3–15.
25. Kim T-H, Mount CW, Gombotz WR, Pun SH (2010) The delivery of doxorubicin to 3-D multicellular spheroids and tumors in a murine xenograft model using tumor-penetrating triblock polymeric micelles. Biomaterials 31: 7386–7397.
26. Perche F, Patel NR, Torchilin VP (2012) Accumulation and toxicity of antibody- targeted doxorubicin-loaded PEG-PE micelles in ovarian cancer cell spheroid model. J Control Release in press.
27. Oishi M, Nagasaki Y, Nishiyama N, Itaka K, Takagi M, et al. (2007) Enhanced growth inhibition of hepatic multicellular tumor spheroids by lactosylated poly(ethylene glycol)-siRNA conjugate formulated in PEGylated polyplexes. ChemMedChem 2: 1290–1297.
28. Tsukioka Y, Matsumura Y, Hamaguchi T, Koike H, Moriyasu F, et al. (2002) Pharmaceutical and biomedical differences between micellar doxorubicin (NK911) and liposomal doxorubicin (Doxil). Jpn J Cancer Res 93: 1145–1153.
29. Bae Y, Nishiyama N, Fukushima S, Koyama H, Yasuhiro M, et al. (2005) Preparation and biological characterization of polymeric micelle drug carriers with intracellular pH-triggered drug release property: tumor permeability, controlled subcellular drug distribution, and enhanced in vivo antitumor efficacy. Bioconjug Chem 16: 122–130.
30. Prescribing Information, Taxotere (2010). Sanofi-Aventis. Available: http:// products.sanofi.us/Taxotere/taxotere.html.
31. ClinicalTrials.gov. Available: http://clinicaltrials.gov/ct2/show/ NCT01639131?term = colon+and+docetaxel&rank = 15. Accessed 2013 March 18.
32. ClinicalTrials.gov. Available: http://clinicaltrials.gov/ct2/show/ NCT00003445?term = cervix+and+docetaxel&rank = 3. Accessed 2013 March 18.
33. Baur M, Van Oosterom AT, Die´ras V, Tubiana-Hulin M, Coombes RC, et al. (2008) A phase II trial of docetaxel (Taxotere) as second-line chemotherapy in patients with metastatic breast cancer. J Cancer Res Clin Oncol 134: 125–135.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 110
34. Mikhail A, Allen C (2009) Block copolymer micelles for delivery of cancer therapy: transport at the whole body, tissue and cellular levels. J Control Release 138: 214–223.
35. Mikhail A, Allen C (2010) Poly(ethylene glycol)-b-poly(epsilon-caprolactone) micelles containing chemically conjugated and physically entrapped docetaxel: synthesis, characterization, and the influence of the drug on micelle morphology. Biomacromolecules 11: 1273–1280.
36. Zahedi P, Stewart J, De Souza R, Piquette-Miller M, Allen C (2012) An injectable depot system for sustained intraperitoneal chemotherapy of ovarian cancer results in favorable drug distribution at the whole body, peritoneal and intratumoral levels. J Control Release 158: 379–385.
37. Marusic´ M, Bajzer Z, Freyer JP, Vuk-Pavlovic´ S (1994) Analysis of growth of multicellular tumour spheroids by mathematical models. Cell Prolif 27: 73–94.
38. Friedrich J, Eder W, Castaneda J, Doss M, Huber E, et al. (2007) A reliable tool to determine cell viability in complex 3-d culture: the acid phosphatase assay. J Biomol Screen 12: 925–937.
39. Joyce J a, Pollard JW (2009) Microenvironmental regulation of metastasis. Nat Rev Cancer 9: 239–252.
40. Hanahan D, Weinberg R a (2011) Hallmarks of cancer: the next generation. Cell 144: 646–674.
41. Okubo H, Matsushita M, Kamachi H, Kawai T, Takahashi M, et al. (2002) A novel method for faster formation of rat liver cell spheroids. Artif Organs 26: 497–505.
42. Del Duca D, Werbowetski T, Del Maestro RF (2004) Spheroid preparation from hanging drops: characterization of a model of brain tumor invasion. J Neurooncol 67: 295–303.
43. Friedrich J, Seidel C, Ebner R, Kunz-Schughart L a (2009) Spheroid-based drug screen: considerations and practical approach. Nat Protoc 4: 309–324.
44. Hardelauf H, Frimat J-P, Stewart JD, Schormann W, Chiang Y-Y, et al. (2011) Microarrays for the scalable production of metabolically relevant tumour spheroids: a tool for modulating chemosensitivity traits. Lab Chip 11: 419–428.
45. Ong S-M, Zhao Z, Arooz T, Zhao D, Zhang S, et al. (2010) Engineering a scaffold-free 3D tumor model for in vitro drug penetration studies. Biomaterials 31: 1180–1190.
46. Kyle AH, Huxham L a, Yeoman DM, Minchinton AI (2007) Limited tissue penetration of taxanes: a mechanism for resistance in solid tumors. Clin Cancer Res 13: 2804–2810.
47. Han M, Oba M, Nishiyama N, Kano MR, Kizaka-Kondoh S, et al. (2009) Enhanced percolation and gene expression in tumor hypoxia by PEGylated polyplex micelles. Mol Ther 17: 1404–1410.
48. Fonseca C, Simo˜ es S, Gaspar R (2002) Paclitaxel-loaded PLGA nanoparticles: preparation, physicochemical characterization and in vitro anti-tumoral activity. J Control Release 83: 273–286.
49. Nemati F, Dubernet C, Poupon M, Couvreur P (1996) Reversion of multidrug resistance using nanoparticles in vitro: influence of the nature of the polymer. Int J Pharm: 237–246.
50. Cavallaro G, Fresta M, Giammona G, Puglisi G, Villari A (1994) Entrapment of B-lactams antibiotics in polyethylcyanoacrylate nanoparticles: Studies on the possible in vivo application of this colloidal delivery system. Int J Pharm 111: 31–41.
Chapter 3: Multicellular Tumor Spheroids for of Nanomedicines 111
51. Panyam J, Labhasetwar V (2003) Dynamics of endocytosis and exocytosis of poly(D,L-lactide-co-glycolide) nanoparticles in vascular smooth muscle cells. Pharm Res 20: 212–220.
52. Yoo HS, Park TG (2004) Folate-receptor-targeted delivery of doxorubicin nanoaggregates stabilized by doxorubicin-PEG-folate conjugate. J Control Release 100: 247–256.
53. Yoo HS, Park TG (2004) Folate receptor targeted biodegradable polymeric doxorubicin micelles. J Control Release 96: 273–283.
112
Effects of Doxorubicin Delivery Systems and Mild Hyperthermia on Tissue Penetration in 3D Cell Culture Models of Ovarian Cancer Residual Disease
Sina Eetezadi, Raquel De Souza, Mirugashini Vythilingam, Rodrigo Lessa Cataldi and Christine Allen
Reprint from Molecular Pharmaceutics (September 2015)
DOI: 10.1021/acs.molpharmaceut.5b00426
Experiments by S. Eetezadi, M. Vythilingam and R.L. Cataldi under
supervision of S. Eetezadi. Written by S. Eetezadi and Raquel De Souza.
Illustrations by S. Eetezadi. Edited by C. Allen.
The copyright of this article belongs to the American Chemical Society (ACS), the publisher of
Molecular Pharmaceutics. Permission for publishing the article as part of this dissertation, is
granted by ACS.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 113
4.1. Abstract
Current chemotherapy strategies for second-line treatment of relapsed ovarian cancer are
unable to effectively treat residual disease post-cytoreduction. The findings presented herein
suggest that tissue penetration of drug is not only an issue for large, unresectable tumors, but
also for invisible, microscopic lesions. The present study sought to investigate the potential
of a block copolymer micelle (BCM) formulation, which may reduce toxicities of doxorubicin
(DOX) in a similar way to pegylated liposomal doxorubicin (PLD, Doxil® / Caelyx®), while
enhancing penetration into tumor tissue and improving intratumoral availability of drug. To
achieve this goal, 50 nm-sized BCMs capable of high DOX encapsulation (BCM-DOX) at drug
levels ranging from 2 to 7.6 mg/mL were formulated using an ultrafiltration technique. BCM-
DOX was evaluated in 2D and 3D cell culture of the human ovarian cancer cell lines HEYA8,
OV-90 and SKOV3. Additionally, the current study examines the impact of mild hyperthermia
(MHT) on the cytotoxicity of DOX. The BCM-DOX formulation fulfilled the goal of controlling
drug release while providing up to 9-fold greater cell monolayer cytotoxicity in comparison to
PLD. In 3D cell culture, using multicellular tumor spheroids (MCTS) as a model of residual
disease post-surgery, BCM-DOX achieved the benefits of an extended release formulation of
DOX and resulted in improvements in free drug accumulation over PLD, while yielding drug
levels approaching that achievable by exposure to DOX alone. In comparison to PLD, this
translated into superior MCTS growth inhibition in the short-term and comparable inhibition
in the long-term. Overall, although MHT appeared to enhance drug accumulation in HEYA8
MCTS treated with BCM-DOX and DOX alone in the short term, improved growth inhibition of
MCTS by MHT was not observed after 48 h of drug treatment. Evaluation of BCM-DOX in
comparison to PLD as well as the effects of MHT is warranted in vivo.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 114
4.2. Introduction
Innovative approaches are urgently needed for ovarian cancer treatment, as no significant
improvements have been achieved since the introduction of platinum-based therapy over 30
years ago.1, 2 Although treatment of localized disease yields a 5-year survival rate of up to 92
%,3 most patients are not diagnosed prior to presentation of advanced and metastatic
disease.4 Standard epithelial ovarian cancer treatment consists of cytoreductive surgery
followed by adjuvant chemotherapy with carboplatin and paclitaxel.3, 5 The extent of residual
disease is a main prognostic factor, but challenges with optimal cytoreduction still exist.6
Further cytoreduction beyond the optimal standard (<1 cm residual disease) is preferable,
whereby only microscopic residual disease remains.7, 8 Current chemotherapy strategies are
unable to effectively treat residual disease post-cytoreduction. Patients respond well to
adjuvant chemotherapy, with a response rate (RR) of over 80 %, but as many as 75 %
eventually relapse, often within 18 months, with progressively resistant disease.9, 10 The
current adjuvant regimen for residual disease has been in place since the early 2000’s when
carboplatin replaced cisplatin due to better tolerability. The 5-year overall survival (OS) rate
associated with this regimen is less than 35 %.5
Doxorubicin (DOX) has a prominent role in the management of recurrent disease. A
liposomal formulation of DOX (PLD, Doxil/Caelyx®), which consists of the drug loaded in 90
nm pegylated liposomes, is approved as second-line therapy for patients with recurrent,
platinum-refractory disease.11-13 However, despite a greatly improved circulation half-life and
increased tumor accumulation, enhancements in therapeutic efficacy by PLD have been
modest relative to free DOX administered systemically.14 This is largely attributed to
persistent retention of drug within the liposomes, with only 50 % of DOX available to its
intracellular target at tumor sites.15 The large size of the liposomes also results in
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 115
heterogeneous distribution of drug, owing to poor tumor penetration of this
nanotechnology.16 These features reduce the cardiotoxic character of DOX, as the size of
liposomes hinders uptake in the heart due to the tight junctions present in this tissue and also
reduces hematologic and gastrointestinal toxicities.17, 18 Yet, no significant superiority in
response is achieved with PLD, and as currently available agents lead to a marginal response,
no consensus exists on the choice of agent for the management of platinum resistant
disease.2, 19, 20 In platinum-resistant recurrence, single-agent DOX is attractive since it does
not show cross-resistance with platinum.21, 22
The addition of MHT to chemotherapy regimens in ovarian cancer, referred to as
hyperthermic intraperitoneal chemotherapy (HIPEC), is defined as heating target areas to 41
- 43 °C.23 This procedure is founded on the enhanced tumor penetration and intracellular
uptake of drugs that ensues in response to hyperthermia, as well as the reported synergism
between hyperthermia and chemotherapeutics.24, 25 Hyperthermia has been shown to hinder
the repair of sublethal damage induced by DNA damaging agents.26 A host of additional
mechanisms are reportedly activated in response to MHT that render the cells more sensitive
to chemotherapeutics.23 DOX is one of the agents commonly used for HIPEC, due to its heat
stability and high molecular weight.27 As a result, the current study examines the impact of
MHT on the cytotoxicity of DOX.
The present study sought to investigate the potential of a micellar formulation for second-
line ovarian cancer treatment. It was hypothesized that this drug delivery platform may reduce
DOX-associated toxicities in a similar way to PLD while accelerating drug release from the
carrier and enhancing penetration into tumor tissue.15, 28 To achieve this goal, small-sized
BCMs were employed for formulation of DOX. BCMs consist of amphiphilic copolymers that
spontaneously form a core-shell structure wherein a hydrophilic shell surrounds a
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 116
hydrophobic core.29 This core offers a suitable environment for encapsulation of hydrophobic
drugs, thereby enhancing the drug’s solubility in aqueous media and shielding the drug from
the biological milieu.29, 30 A number of DOX formulations relying on BCMs have entered clinical
development.31 The BCM-DOX formulation presented herein met the following selection
criteria: 1) small sized delivery system with a hydrodynamic diameter of < 50 nm, potentially
enabling greater penetration into residual lesions than achievable with PLD; 2) efficient drug
loading of at least 2 mg/mL DOX comparable to the clinically available DOX formulations; and
3) adequate stability and prolonged drug release under physiologically relevant conditions.
The focus of this study was to evaluate the BCM-DOX formulation at the tumor-tissue
interface. Specifically, we compared BCM-DOX to DOX formulated in PLD and to DOX alone
as a small molecule control, under normothermia (NT) and mild hyperthermia (MHT).
Multicellular tumor spheroids (MCTS) with an approximate diameter of 500 µm were used as
3D in vitro model systems of microscopic residual disease after optimal cytoreduction.8
MCTS comprised of an outer region of proliferative cancer cells surrounding intermediate
regions of quiescent cells and, if the MCTS is large enough, a necrotic core, are considered
suitable representations of clinical disease.32, 33 The use of 3D cell culture and MCTS is
particularly relevant when assessing nanomedicines, since in many cases the delivery vehicle
determines the extent of penetration of the encapsulated drug into tumor tissues.34, 35 The
development and physicochemical characterization of BCM-DOX are presented herein,
followed by assessment of monolayer cytotoxicity and MCTS penetration and growth
inhibition with or without MHT.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 117
4.3. Materials and Methods
4.3.1. Materials
α-Methoxy-ω-hydroxypoly(ethylene glycol) (CH3O-PEG-OH) was obtained from Sigma-
Aldrich (Oakville, ON, Canada) and dried by azeodistillation of toluene. ε-caprolactone was
purchased from Sigma-Aldrich and was dried using calcium hydride. Acetonitrile (ACN),
triethyl amine (TEA), acetic acid, P-nitrophenyl phosphate, TritonTM-X-100, sodium acetate,
Tween® 80 and bovine serum albumin (BSA) were also purchased from Sigma-Aldrich and
used without further purification. DOX was purchased from Polymed Therapeutics, Inc.
(Houston, TX, USA) and PLD was purchased as Caelix® from the hospital pharmacy at the
Princess Margaret Cancer Centre in Toronto. MCDB cell media was obtained from Sigma-
Aldrich, while all other cell media, phosphate buffered saline (PBS) and fetal bovine serum
(FBS) were obtained from Life Technologies (Carlsbad, CA, USA). Ovarian carcinoma cell lines
were purchased from the American Type Culture Collection (ATCC Manassas, VA, USA;
SKOV3 and OV-90 cells) or from the M. D. Anderson Cancer Center (Houston, TX, USA; HEYA8
cells).
4.3.2. Preparation of BCM-DOX
MePEG-b-PCL copolymers were prepared by metal-free cationic ring opening
polymerization of ε-caprolactone with CH3O-PEG-OH as macroinitiator using an established
method previously reported.36 For micelle formation, 100 mg of copolymer was dissolved in
1.5 mL ACN at 40 °C for 24 h. DOX was dissolved in 1 mL ACN pretreated with 3.0 molar
equivalents of TEA and stirred for 2 h. Both solutions were mixed together at varying feed
ratios of drug to copolymer and added dropwise to 17 mL of purified water prior to dialysis
against 500 mL of 0.9 % sodium chloride injection USP (B. Braun Medical Inc., USA) using a
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 118
Spectra/Por 2 dialysis membrane (Spectrum Labs, Rancho Dominguez, CA, USA) with a
molecular weight (MW) cut off of 1 kDa. After 24 h the micelle solutions were centrifuged at
5000 rpm for 10 min to remove residual drug crystals prior to concentration. The supernatant
was concentrated to a final volume of 2 mL using an Amicon Stirred Cell system with a MW
cut-off of 5 kDa (EMD Milipore, Billerica, MA USA).
4.3.3. Physico-Chemical Characterization of BCM-DOX
The hydrodynamic diameter of the micelles was determined by dynamic light scattering
(DLS) using a 90Plus Particle Size Analyzer (Brookhaven Instruments Corp., Holtsville, NY) at
an angle of 90 ° and temperature of 25 °C. The samples were diluted to 0.5 mg/mL of
copolymer prior to DLS analysis (n = 3). Transmission electron microscopy (TEM) on a Hitachi
7000 microscope was performed on micelle solutions diluted to 0.1 mg/mL and pre-stained
with a 1 % uranyl acetate solution immediately prior to analysis. For sizing, three images were
taken of each sample and 25 micelles were manually measured per image. DOX loading was
determined using an Agilent series 1200 HPLC (Mississauga, ON, Canada) with a Waters
XTerra C18 reverse phase column (Waters corporation, Milford, MA, USA) connected to a UV
detector (Waters 2487) at a wavelength of 485 nm (n = 3). A mixture of 60 % ACN and 40 %
water with 1 % acetic acid was utilized as the mobile phase. DOX encapsulation efficiency
and loading content were calculated using the following equations:
𝐸𝑛𝑐𝑎𝑝𝑠𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 (𝐿𝐸𝑛)[%] =𝐹𝑖𝑛𝑎𝑙 𝐷𝑜𝑥𝑜𝑟𝑢𝑏𝑖𝑐𝑖𝑛 [
𝑚𝑔𝑚𝐿]
𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝐷𝑜𝑥𝑜𝑟𝑢𝑏𝑖𝑐𝑖𝑛 [𝑚𝑔𝑚𝐿]
× 100
𝐿𝑜𝑎𝑑𝑖𝑛𝑔 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (𝐿𝐶𝑛) [%] =𝐹𝑖𝑛𝑎𝑙 𝐷𝑜𝑥𝑜𝑟𝑢𝑏𝑖𝑐𝑖𝑛 [
𝑚𝑔𝑚𝐿]
𝐹𝑖𝑛𝑎𝑙 𝐶𝑜𝑝𝑜𝑙𝑦𝑚𝑒𝑟 [𝑚𝑔𝑚𝐿]
× 100%
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 119
4.3.4. Evaluation of Drug Release
For drug release studies, 1 mL aliquots of micelle samples in dialysis bags (MW cut-off of
2 kDa) were placed into 250 mL of PBS containing 0.1 % of Tween® 80 (i.e. the acceptor
medium), with gentle agitation (n = 3 batches per formulation). At selected time intervals, 1
mL samples (n = 3 samples per time point) were withdrawn from the acceptor medium and
replaced with an equivalent volume of fresh acceptor medium. Release behavior of BCM-DOX
at a drug to copolymer ratio of 0.2 as well as PLD was further investigated with addition of
BSA at a 50 mg/mL concentration inside the dialysis bag to reflect a physiologically relevant
concentration of the protein that is typically found in blood.37 DOX concentrations were
determined using a SPECTRA maxPLUS384 plate reader (Molecular Devices Technologies,
Sunnyvale, CA, USA) set at 485 nm using a DOX standard curve.
4.3.5. Cell Lines
The three ovarian cancer cell lines were maintained as monolayers at 37 °C in a 5 % CO2
atmosphere and 90 % relative humidity. SKOV3 cells were cultured in McCoy's 5a medium
supplemented with 10 % FBS, OV-90 in a 1:1 mixture of MCDB 105 and Media 199
supplemented with 15 % FBS, and HEYA8 in RPMI 1640 supplemented with 10 % FBS. 1 %
penicillin/streptomycin was added to all media.
4.3.6. Monolayer Cytotoxicity
A method routinely employed in our laboratory was used to assess monolayer cytotoxicity
of PLD, BCM-DOX and free DOX, using ten 1:4 serial dilutions, starting with 125 ug DOX/mL.
34 Briefly, subconfluent cells were harvested and seeded onto 96-well plates at a seeding
density of 4,000 cells per well and allowed to adhere overnight. Following a 48 h incubation
period with varying drug concentrations, cells were washed with PBS prior to assessment of
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 120
viability via the alkaline phosphatase (APH) assay. In brief, cells were incubated with 100 μL
of freshly prepared reaction buffer (sodium acetate buffer at pH 5.5 containing 1 % Triton-X
supplemented with 2 mg/mL p-nitrophenyl phosphate) for 2 h at 37 °C.34 Cell viability was
determined by measuring the UV absorbance at 485 nm using an automated 96-well plate
reader (SpectraMax Plus 384, Molecular Devices) following addition of 10 μL of 1 M sodium
hydroxide to each well. The average absorbance (A) of three independent experiments was
normalized to controls as follows:
𝑣𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦 [%] =(𝐴𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 – 𝐴𝑚𝑒𝑑𝑖𝑎)
(𝐴𝑐𝑜𝑛𝑡𝑟𝑜𝑙 – 𝐴𝑚𝑒𝑑𝑖𝑎)
4.3.7. MCTS Growth Studies
MCTS of HEYA8 and OV-90 cells were grown as previously established by our group.34 As
reported in the literature and confirmed in the present study (data not shown), SKOV3 cells
did not form stable MCTS.38 Subconfluent cells were seeded onto non-adherent 96-well
round-bottomed Sumilon PrimeSurfaceTM spheroid plates (MS-9096U; Sumitomo Bakelite,
Tokyo, Japan) containing conventional growth media and incubated for 7 days at 37 °C.
Following cellular aggregation, each well contained a single MCTS. During the 7-day growth
period, 50 % of media was exchanged every other day, so as to not disturb the structure of
the MCTS, prior to imaging using a light microscope with a 10x objective lens (VWR
VistaVisionTM) connected to a digital camera (VWR DV-2B). The volume of each MCTS was
determined according to a published method.34 Briefly, the 2D cross-sectional area was
measured using an automated image analysis macro developed for use with the ImageJ
software package (Version 1.48V) and volumes were calculated assuming a spherical shape.
Growth curves were fit to the Gompertz growth equation of tumor growth.34
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 121
4.3.8. MCTS Cell Packing Density
To determine the number of cells per MCTS volume, cells were seeded at different initial
cell numbers and grown for 7 days as described above. Twelve MCTS per seeding density
were washed by addition and removal of PBS and gentle disaggregation using trypsin-EDTA.
Cell suspensions were placed in vials for centrifugation and resuspended in fresh media. The
number of cells was determined using a Bio-Rad TC10 automated cell counter (n = 3 samples
counted per 12 MCTS). Cell density was calculated by dividing the mean number of cells per
spheroid by the volume of each spheroid. For imaging studies, MCTS were incubated with 1x
CellMaskTM Green plasma membrane stain (Life Technologies) and imaged using a Zeiss
LSM700 confocal microscope (Carl Zeiss AG, Oberkochen, Germany) with a FITC filter (Ex
522 nm, Em 535 nm).
4.3.9. Drug Penetration into MCTS
MCTS were grown as described above until they reached a diameter of about 500 µm, at
which point they were transferred onto Nalge Nunc Lab-Tek II #1.5 German Coverglass
microscopy slides for live cell imaging (Figure 4-1). Three MCTS per group were treated with
40 µg DOX/mL as DOX alone, BCM-DOX or PLD and incubated in a cell culture incubator (37
°C, 95 % humidity, 5 % CO2) for 2 hours prior to imaging using a Zeiss LSM700 confocal
microscope with a 20x objective and DOX filter (Ex 488 nm, Em 580 nm). The experiment only
evaluated free DOX accumulation within, and not surrounding, the MCTS over time, which
results in highly increased fluorescence signal within the MCTS. To select the microscope
offset settings employed here, MCTS were incubated with DOX alone, as this resulted in the
brightest fluorescence signal. The signal of free drug detected from the media surrounding
the MCTS was considered baseline for all treatment groups. As such, only signal above this
baseline was considered true DOX accumulation. Fluorescence laser gain was established so
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 122
as to minimize bleed-through artifacts, which were most prominent following incubation with
DOX alone. Microscopy settings were kept constant for imaging of all MCTS to enable
standardized comparisons. At selected timepoints, 3 optical slices per MCTS were imaged at
30, 60 and 90 µm from the MCTS surface (slices 1, 2 and 3) to obtain spatiotemporal
information on drug distribution. A recognized limitation of this methodology is that excitation
and fluorescence emission have to travel through more cell layers to and from the center of
the MCTS, respectively, than at the borders, due to the spherical nature of the MCTS. This
causes attenuation of the signal, a limitation which applies in the same way to all treatment
groups and conditions tested. To determine the extent of DOX penetration into each MCTS,
DOX signal intensity was determined based on a previously published method with
modifications.35 Briefly, fluorescence intensity per unit area was determined for periphery
(red), intermediate (blue) and core (green) regions of each optical slice, using a custom
MATLAB algorithm (MathWorks Inc., MA, USA). The 3 regions were equally spaced, concentric
and mirrored the MCTS perimeter in shape. Fluorescence signal from each region was
averaged for each MCTS, and these values were then averaged for the 3 MCTS within each
treatment group. Fluorescence intensity values for each region are reported as mean ±
standard deviation (SD) for n = 3 MCTS (Figure 4-1). Additional images were taken at a 50-µm
depth from the MCTS surface to provide a qualitative overview of drug penetration.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 123
Figure 4-1: Workflow for evaluating drug penetration in MCTS. Three MCTS per treatment
group were imaged using fluorescent live cell imaging on a confocal microscope. Images
were taken at 30, 60 and 90 µm (slices 1, 2 and 3) from the MCTS surface. For each of these
optical slices, a custom MATLAB algorithm was used to determine the fluorescence intensity
per unit area for three equally spaced concentric regions (periphery/P/red,
intermediate/I/blue and core/C/green) that mirror the MCTS perimeter. For each of the three
defined regions (red, blue and green), the average signal intensity per unit area within the three
optical slices per MCTS was calculated. This value was finally reported as mean ± SD for n =
3 MCTS.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 124
4.3.10. MCTS Growth Inhibition
MCTS were grown as described above until they reached a diameter of about 500 µm. Six
MCTS per treatment (free DOX, BCM-DOX, PLD) were incubated for 48 h with 500 ng DOX/mL,
or an equivalent volume of cell media as untreated controls. Following incubation, MCTS were
washed with media. Every other day thereafter, 50 % of media was replaced prior to imaging
for MCTS volume determination as described above.
4.3.11. MHT Experiments
The impact of MHT on free DOX, BCM-DOX and/or PLD performance in terms of monolayer
cytotoxicity, MCTS drug penetration and growth inhibition was assessed at a temperature
relevant to HIPEC (i.e. 42 °C).25 For all experiments involving MHT, media temperature was
monitored using a fiber optic infrared temperature probe carefully placed so as not to disturb
cells or MCTS (Luxtron Model 790, LumaSense Technologies Inc.). For cytotoxicity studies,
cells were plated in monolayers and incubated overnight as described above in “monolayer
cytotoxicity.” For MCTS growth inhibition studies, MCTS were grown as described above in
“MCTS growth inhibition.” To assess monolayer cytotoxicity and MCTS growth inhibition
under MHT, plates were first allowed to reach 37 °C through incubation at 37 °C for 30 min.
MHT was then induced by placing the plates for a short (3 - 5 min) time in an incubator set at
45 °C; when media temperature reached 42 °C as assessed using the fiber optic infrared
temperature probe mentioned above, plates were immediately transferred to a dedicated 42
°C cell culture incubator for 1 h. This constituted the MHT period. Plates were incubated at 37
°C for the remainder of the 48 h treatment period. Thereafter, monolayer viability and MCTS
volume changes were assessed as described above in “monolayer cytotoxicity” and “MCTS
growth inhibition,” respectively. To assess MCTS penetration under MHT, drug-containing
media was added to MCTS and transferred onto microscopy slides as described above in
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 125
“MCTS drug penetration.” Slides were directly transferred into the 42°C cell incubator. The
thermal conductivity of the glass bottom slides precluded the need for pre-incubation in a
45°C incubator in order for the media to quickly arrive at a temperature of 42°C. MCTS were
incubated at 42 °C for 1 h. Slides were then transferred to 37 °C for 40 min, such that the total
duration of drug exposure was the same as for MCTS in the NT group (i.e. 2 h) prior to
imaging. Following incubation, MCTS were imaged as described above in “MCTS drug
penetration.” Using the same method, optical slides were imaged at 30 and 50 µm from the
MCTS surface to provide a qualitative overview of drug penetration.
4.3.12. Statistical Analysis
Data are expressed as the mean of at least 3 independently performed experiments ± SD.
For comparisons between multiple groups, one-way ANOVA was employed with subsequent
post-hoc analyses with Bonferroni correction. For comparisons between two groups,
Student’s t-tests (2-tailed, independent samples) were performed. All statistical tests were
performed with IBM SPSS statistical software (SPSS Inc.). Significance was assigned at p <
0.05.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 126
4.4. Results
4.4.1. Physicochemical Characterization of BCM-DOX
DOX was physically entrapped within the hydrophobic core of the MePEG-b-PCL
copolymer micelles. The low solvent:water ratio used in the micelle formation process prior
to concentration via ultrafiltration enabled encapsulation of the drug at a high encapsulation
efficiency.39 Progressively higher drug loading was achieved by increasing the drug feed (drug
to copolymer wt/wt) ratio, to a maximum of 7.6 ± 0.5 mg DOX/mL, achieved at a ratio of 0.3
(Table 4-1). At ratios beyond 0.3, drug precipitation was observed during the
ultracentrifugation step. Increases in loading content (LC) up to 14.3 ± 0.8 % (wt/wt) were
observed for drug to copolymer ratios of 0.1 to 0.3 (Table 4-1). An increase in the drug feed
ratio also increased the encapsulation efficiency (EE), suggesting that interactions between
DOX molecules further drive stabilization and encapsulation within the micelle core. The size
of BCMs containing DOX increased slightly as the drug to copolymer ratio was increased and
the size distribution of micelles was found to be monomodal at all ratios. TEM images
revealed a homogenous population of spherical micelles with core sizes of 20 - 29 nm (Figure
4-2).
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 127
Table 4-1: Characteristics of BCM-DOX at different drug to copolymer ratios and a constant
copolymer concentration of 50 mg/mL (n = 3 independently produced batches). DLS =
Dynamic Light Scattering, TEM = Transmission Electron Microscopy, LC = Loading Content,
EE = Encapsulation Efficiency.
Feed ratioa [wt/wt]
Drug Loading [mg / mL]
Size [nm] LCb [%] EEc [%]
DLS TEM
0.1 2.0 ± 0.2 38.5 ± 5.9 22.2 ± 1.1 3.9 ± 0.3 39.1 ± 3.1
0.2 4.4 ± 0.5 47.7 ± 8.3 20.6 ± 2.0 8.6 ± 0.5 47.4 ± 2.9
0.3 7.6 ± 0.5 51.0 ± 2.6 29.3 ± 1.6 14.3 ± 0.8 63.7 ± 3.4
(a) Ratio of initial drug to copolymer used to prepare the formulation. (b) Weight percent of drug relative to copolymer in the final product. (c) Weight percent of drug content in the final formulation relative to amount added at the beginning of the process.
Figure 4-2: Representative TEM images of final BCM-DOX formulations at 0.1, 0.2 and 0.3 drug
to copolymer feed ratios [wt/wt]. BCM-DOX formulations were imaged using a Hitachi 7000
microscope at 40000x magnification.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 128
4.4.2. DOX Release from BCM-DOX
The DOX release from BCM-DOX prepared from drug to copolymer ratios of 0.1 to 0.3
(wt:wt) was assessed in vitro in comparison to PLD and with DOX alone as a control. The
extent of drug release over 7 days from the BCMs in PBS buffer at pH 7.4 decreased with an
increase in the drug-to-copolymer ratio, likely as a result of increased self-aggregation of DOX
molecules (Figure 4-3). Assessment of drug release from the BCM formulation with a 0.2 drug
to copolymer ratio was also performed in PBS containing physiologically relevant
concentrations of BSA (50 mg/mL). As expected, the addition of BSA resulted in increased
release of drug from the BCMs, with 57 ± 2.5 % of drug released in 7 days. This may be a
result of destabilizing interactions between BSA and the micelles, such as polymer extraction
or formation of mixed micelles with polymer-protein complexes as reported previously.40
However, no detectable drug was released from PLD throughout the 7-day incubation period
in the presence or absence (data not shown) of BSA. The use of drug alone as a control
validated the experimental setup, with 100 % of drug detected in the release compartment
within a few hours. For further in vitro evaluation in ovarian cancer cell lines, BCM-DOX
formulated with a 0.2 feed ratio was used.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 129
Figure 4-3: In vitro release of DOX from BCMs prepared from drug-to-copolymer ratios of 0.1
to 0.3 (wt:wt). Release studies were conducted in PBS (pH 7.4) and/or PBS containing BSA
(50 mg/mL).
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 130
4.4.3. Monolayer Cytotoxicity Studies
Cytotoxicity of PLD, BCM-DOX and DOX alone was evaluated in three distinct epithelial
ovarian cancer cell lines that represent a variety of histological subtypes of the disease (Figure
4-4): (1) HEYA8, papillary cystadenocarcinoma of undifferentiated histology;41, 42 (2) OV-90,
serous adenocarcinoma,41-43 the most common histological subtype of epithelial ovarian
cancer;2 and (3) SKOV3, clear cell adenocarcinoma.42 As shown in Figure 4-4, DOX alone was
the most cytotoxic treatment in all cell lines and PLD was the least effective, associated with
the highest IC50 values in cell monolayer experiments. Under these conditions, BCM-DOX
provided up to a 9.6-fold enhancement in cytotoxicity when compared to PLD.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 131
Figure 4-4: Assessment of cytotoxicity of DOX, BCM-DOX and PLD in cell monolayers using
the APH assay in HEYA8 (A), OV-90 (B) and SKOV3 (C) ovarian cancer cell lines after 48 h
treatment (n = 3 per treatment). Under these conditions, BCM-DOX cytotoxicity was superior
in all cell lines when compared to PLD (p < 0.05). A (*) denotes statistically significant
differences between IC50 values resulting from the three treatments.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 132
4.4.4. MCTS 3D Cell Model
Evaluation of the formulations was further continued using 3D cell culture as a means to
evaluate efficacy in vitro. MCTS of the ovarian cancer cell lines HEYA8 and OV-90 were
established as a model of microscopic residual disease after optimal cytoreduction using a
modified liquid overlay technique as published previously.35 Adequate MCTS formation was
not achieved with SKOV3 cells. MCTS formed from both HEYA8 and OV-90 cell lines showed
an exponential growth pattern over the 15-day period, in accordance with the Gompertz
equation for tumor growth (Figure 4-5).45 The initial cell number was chosen such that MCTS
reached a mean size of 500 µm after 7 days of growth. The one-week growth period is an
important strategy used in this study, as sufficient time is required for cells to form coherent
MCTS structures. This differs fundamentally from cell aggregates that form after short
incubation times of 1 - 2 days.44, 45
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 133
Figure 4-5: MCTS growth studies for HEYA8 and OV-90 ovarian cancer cell lines. Different
numbers of cells were seeded, and growth was monitored over a 15-day period. The
horizontal line indicates a diameter of 500 µm. Subsequent studies were conducted with initial
cell numbers of 1000 cells/well and 2000 cells/well for HEYA8 and OV-90, respectively. This
yielded a size of about 500 µm after 7 days of growth. Growth curves were fit to the Gompertz
growth equation of tumor growth on a log plot of MCTS volume as a function of time.35
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 134
4.4.5. MCTS Cell Organization and Density
To assess the cellular organization and cell density, the cellular membranes within MCTS
were stained and optical slices analyzed using confocal microscopy. The analysis revealed
significant differences in cellular organization and cell density between the HEYA8 and OV-90
MCTS (Figure 4-6). HEYA8 MCTS were homogenously spherical in shape, whereas OV-90
MCTS displayed a more irregular but coherent cell mass. Both remained intact during transfer
and washing steps. As shown in Figure 4-6 A, HEYA8 MCTS included densely packed cells
throughout the spheroid. OV-90 MCTS, on the other hand, were composed of a heterogeneous
arrangement of cells with areas of very tight packing adjacent to large gaps (Figure 4-6 A).
Overall, the number of cells per volume was 7.6-fold greater in the HEYA8 MCTS in
comparison to the OV-90 MCTS (Figure 4-6 B).
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 135
Figure 4-6: Confocal images of MCTS with CellMaskTM Green Plasma Membrane Stain (A) and
MCTS cellular density, expressed as cells per volume for both cell lines (B). Scale bars
represent 100 µm.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 136
4.4.6. MCTS DOX Penetration Studies
To provide a representative illustration of free DOX distribution under NT and MHT, HEYA8
and OV-90 MCTS were imaged at a 50-µm depth from the MCTS surface after a 2 h exposure
to DOX alone, BCM-DOX or PLD using confocal microscopy (Figure 4-7). In accordance with
the cellular organization discussed above (Figure 4-6), it can be observed that free DOX
distribution is very homogenous in HEYA8 MCTS, while in OV-90 MCTS some areas show
high drug accumulation, whereas others appear to contain very little drug, even at the rim of
the MCTS.
The inherent fluorescence of DOX is commonly used to quantify the concentration of this
drug.28, 46 This approach is limited in that DOX’s fluorescence properties can change as a
function of its environment. In the current experimental set-up, penetration of free drug into
the MCTS provides an opportunity for DOX to interact with various cellular components.47 It
has been shown that DOX is highly quenched when interacting with DNA, while its
fluorescence intensity is amplified when interacting with histones.47, 48 To minimize
confounding effects due to alterations in the fluorescence signal of free DOX, drug penetration
was assessed at the earliest time point allowed by the experimental setup (2 h). DOX
interactions with intracellular components are likely amplified at later times.
To select the microscope offset settings employed here, MCTS were incubated with DOX
alone, which showed the brightest fluorescence signal. Following incubation with DOX alone,
the signal of free drug detected from the media surrounding the MCTS was considered
baseline fluorescence. As such, for all treatment groups, fluorescence intensity above this
baseline as a result of free drug accumulation in MCTS was considered to be true DOX
accumulation. Moreover, fluorescence laser gain was chosen to minimize bleed-through
artifacts of fluorescence signal, which was most prominent upon incubation with DOX alone.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 137
With this setup, the levels of free drug released from the carrier were observed to be very low
with PLD.
Free DOX penetration into the MCTS upon treatment was quantitatively assessed by
imaging MCTS at optical slices captured at 30, 60 and 90 µm from the surface of the MCTS
after incubation with DOX alone, BCM-DOX and PLD. The penetration of free DOX due to the
different delivery strategies (DOX alone, BCM-DOX, PLD) was analyzed (1) separately for each
of the three regions (periphery, intermediate, core) and (2) comparing the regions, both under
NT and MHT.
1) Penetration of free DOX into each MCTS region: As shown in the upper panel of Figure
4-8, the distribution of free DOX resulting from incubation with DOX alone, BCM-DOX or PLD
differed significantly in the peripheral region of the HEYA8 MCTS. PLD resulted in the lowest
levels of free drug in the periphery, while incubation with DOX alone resulted in the highest
levels. In the intermediate region, incubation with DOX alone and BCM-DOX resulted in
equivalent DOX content, which were both significantly higher than that achieved with PLD.
Finally, in the core region, the three delivery strategies resulted in statistically equivalent drug
concentrations. The same pattern was observed under NT and MHT, with the exception of
the core layer, where MHT yielded higher drug concentrations upon BCM-DOX exposure in
comparison to PLD. Overall, similar observations were made with OV-90 MCTS as shown in
the lower panel of Figure 4-8. At the periphery, the three DOX delivery strategies resulted in
significant differences in DOX levels under both temperature conditions examined, with free
DOX yielding the highest level of free drug concentration and PLD yielding the lowest. At the
intermediate and core regions, BCM-DOX resulted in higher DOX levels than PLD under MHT
but not NT.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 138
2) Differences in free DOX penetration between the three MCTS regions: For the three DOX
delivery strategies assessed in this study, distinct penetration patterns of free DOX into the
MCTS were observed. For DOX alone, a progressive decrease in drug penetration from the
periphery to intermediate and core regions of HEYA8 MCTS was observed (Figure 4-8, top
panel), with significant differences in free drug concentrations between the three regions both
under NT and MHT. A similar pattern was observed with BCM-DOX, with the exception of
intermediate and core regions, which were not statistically different from each other. In
contrast, free DOX penetration resulting from PLD exposure was homogeneously low in all
three regions of HEYA8 MCTS, with no significant differences in DOX content under both
temperature conditions tested. Penetration into the irregularly shaped, leaky OV-90 MCTS
was found to be generally similar to what was observed in the more spherical and robust
HEYA8 MCTS (Figure 4-8, bottom panel), however some differences emerged. Under NT,
differences in drug content in OV-90 MCTS resulting from treatment with DOX alone were only
detected between the periphery and core. In contrast, under MHT, DOX alone yielded
significantly higher free drug levels in the periphery than in intermediate and core regions,
while these two inner regions showed no differences in DOX levels. For both BCM-DOX and
PLD, differences in free DOX content were found between all regions except intermediate and
core, which showed equivalent levels under NT. In contrast, under MHT, only periphery and
core regions showed differences in DOX content for both BCM-DOX and PLD.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 139
Figure 4-8 (next page): Assessment of free DOX penetration into MCTS of HEYA8 and OV-90
ovarian carcinoma cell lines following incubation with DOX alone, BCM-DOX or PLD at 37 °C
(NT) or 42 °C (MHT) with free drug penetration quantified using fluorescence imaging. The
total fluorescence intensity per unit area was separately calculated for periphery (P),
intermediate (I) and core (C) regions of each MCTS as outlined in the methods section. Data
points represent the mean fluorescence intensity of each area within 3 MCTS ± SD.
Statistically significant differences between each region within the same treatment are
denoted by (*); between treatments for each specific region are denoted by (#); and
between MHT and NT for each treatment and region are denoted by ().
Figure 4-7: Qualitative assessment of DOX penetration into MCTS formed from HEYA8 or OV-
90 ovarian carcinoma cells. Representative images show drug penetration at a 50-µm depth
from the MCTS surface, imaged using a Zeiss LSM700 confocal microscope with a 20x
objective after 2 h incubation with each treatment. PLD visualization is limited due to low
levels of free drug released, yielding low fluorescence intensity. Individual images represent
635 µm fields of view. Scale bars: 100 µm.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 140
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 141
Figure 4-9: Representative images of free DOX when delivered as DOX alone or as BCM-DOX
after 2 h incubation at a depth of 30 µm under NT and MHT. In general, DOX alone quickly
accumulated within cell nuclei, whereas drug delivered via BCM-DOX accumulated
substantially in the cell membrane (white arrows). Mild hyperthermia appeared to create a
more diffuse drug distribution pattern for both delivery strategies and cell lines. PLD could not
be visualized due to undetectably low free drug levels that yielded low fluorescence intensity.
Scale bars: 10 µm.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 142
4.4.7. MCTS Drug Distribution
In addition to drug penetration, free DOX distribution within the cells of MCTS tissue
following 2 h exposure to DOX alone or BCM-DOX was examined qualitatively at 30-µm depth
to allow for sufficient fluorescence intensity for display purposes (Figure 4-9). In both HEYA8
and OV-90 MCTS, BCM-DOX yielded substantial free DOX accumulation within the cell
membranes whereas drug accumulated quickly within cell nuclei upon treatment with DOX
alone. In all cases, MHT yielded a more diffuse DOX accumulation pattern. With the chosen
microscope settings and laser gain described in the methods section, fluorescence intensity
from MCTS treated with PLD was too low for visualization of free drug.
4.4.8. MCTS Growth Inhibition Studies
Evaluation of the growth inhibitory effect of free DOX, BCM-DOX and PLD in OV-90 and
HEYA8 MCTS was performed after 48 h of drug exposure under NT or 1 h of MHT with the
remainder of the incubation period under NT (Figure 4-10). Overall, the growth inhibitory
effects of BCM-DOX were superior to that of PLD. BCM-DOX not only inhibited MCTS growth
to a significantly greater extent than PLD at days 6 and 8, but also displayed a level of growth
inhibition comparable to that of free DOX. At certain instances (HEYA8 37˚C day 2; OV-90 37˚C
day 4; HEYA8 42˚C day 2; OV-90 42˚C day 4), the volumes of untreated MCTS were
comparable to those of PLD-treated MCTS, but not to those of BCM-DOX-treated MCTS.
Although BCM-DOX was superior to PLD in terms of growth inhibition, PLD displayed
significant growth inhibition at later timepoints compared to untreated controls. In general,
the addition of MHT to the three DOX delivery strategies did not enhance MCTS growth
inhibition. MHT itself had a statistically significant effect on OV-90 MCTS growth observed in
non-treated controls at the last timepoint (p=0.034). Marginal significance was also
found in non-treated controls of HEYA8 MCTS on days 2 and 8 (p=0.047).
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 143
Figure 4-10: HEYA8 and OV-90 MCTS growth inhibition of free DOX, BCM-DOX and PLD after
48 h drug exposure under NT or 1 h MHT followed by 47 h NT. Data points represent the mean
volume of 3 - 6 MCTS ± SD. A () denotes instances when BCM-DOX performance is superior
to PLD and comparable to that of free DOX (p < 0.05).
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 144
4.4.9. MHT Effects on Monolayer Cytotoxicity
To further explore the apparent ineffectiveness of MHT in enhancing DOX-induced MCTS
growth delay under our experimental conditions, the combination was assessed in HEYA8
and OV-90 monolayers. As shown in Figure 4-11, at low DOX concentrations, an apparent
increase in cytotoxicity can be observed in HEYA8 cells when normalized to control cells at
NT. However, when the cytotoxic effects of MHT itself are controlled for, it becomes evident
that MHT has no effect on DOX cytotoxicity under these experimental conditions. For OV-90
no effect of MHT could be observed.
Figure 4-11: Cytotoxicity of DOX under NT and MHT in HEYA8 and OV-90 cell monolayers,
assessed after 48 h treatment using the APH assay. Data for MHT is reported relative to NT
control and relative to MHT control. Although MHT does impact HEYA8 cell viability, it does
not affect DOX-induced cytotoxicity in either cell line.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 145
4.5. Discussion
At the tumor site, properties of the delivery system and the drug, as well as properties of
the tumor itself, determine overall treatment success. A tumor’s properties are subject to
tremendous heterogeneity, given the numerous cell types and complex microenvironment.49
Although tumor heterogeneity influences all aspects of nanomedicine efficacy, it is by and
large beyond our control.31, 50 Per contra, formulation scientists have the ability to develop
delivery systems that can be modified in terms of performance-dependent parameters,
thereby altering the intratumoral distribution of free drug. These capabilities may enable
scientists to adapt delivery strategies to tumor properties in order to improve the therapeutic
indices of drugs.31 The micelle system developed in the present work was formulated with
the goal of building upon the positive attributes of PLD while enhancing poor tumor drug
availability and penetration.15, 28 In previous studies by Mikhail et al., MCTS were shown to
provide a useful platform for assessment of tumor penetration of fluorescently-labeled
polymeric micelles of different sizes.34, 35 In that study, the penetration data obtained using
MCTS in vitro correlated with penetration of BCMs in vivo in solid tumors.35 MCTS mimic the
microscopic residual tumor nodules that either remain following optimal cytoreductive
surgery or develop due to metastasis.32-35 The addition of MHT as a strategy to improve drug
effect was explored due to the promising results achieved with HIPEC in the intraoperative,
adjuvant setting of ovarian cancer treatment which is currently under investigation in 18
randomized clinical trials.24, 51, 52
The high drug loading and sustained drug release provided by our BCM-DOX system,
concurrently with its improved level of cytotoxicity over PLD, contributes to this system’s
potential as an improved nanosystem for DOX delivery. BCM-DOX has a diameter of 50 nm,
compared to 90 nm for PLD. Assuming spherical shapes, this translates to a reduction of 1/6
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 146
in terms of volume, yet significantly higher drug loading was achieved. This was
accomplished using ultrafiltration, which has historically been used to concentrate
macromolecules.53 Indeed, the BCM-DOX formulation was capable of DOX encapsulation at
drug levels ranging from 2 to 7.6 mg/mL contingent upon the drug to copolymer ratio used.
This is a significant improvement not only over PLD (2 mg/mL drug loading), but also over
other published DOX micelle systems, with the highest drug loading level reported to date
being 2 mg/mL for physically entrapped drug.54 In vitro, DOX release from BCM-DOX was
observed over time, yet release from PLD was undetectable within the timeframe and
experimental conditions employed (Figure 4-3). Accordingly, limited drug release is a
recognized drawback of PLD and results in limited availability of drug at target sites.15, 55 This
corresponded well with our cell monolayer experimental results, in which PLD-induced
cytotoxicity was up to 52-fold lower than that induced by DOX alone in the three cell lines
investigated (Figure 4-4). The BCM-DOX formulation, however, fulfilled the goal of controlling
drug release while providing superior ovarian cancer cell cytotoxicity (up to 9-fold greater) in
comparison to PLD within the experimental timeframe used here. Accordingly, free DOX
accumulation in MCTS was found to be very limited following incubation with PLD, with free
drug levels up to 480-fold lower than that achieved with BCM-DOX. Drug accumulation
resulting from BCM-DOX approached levels achieved following incubation with DOX alone
(Figure 4-8). Differences in the cellular distribution pattern of DOX, resulting following
incubation with BCM-DOX in comparison to DOX alone, suggest that not only increased
release of free drug from BCM-DOX, but also penetration of micelles may account for the
increased signal in comparison to PLD (Figure 4-9).
In general, enhanced drug accumulation due to MHT was observed in MCTS incubated
with BCM-DOX and DOX alone (Figure 4-8). MHT further amplified the free drug accumulation
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 147
of BCM-DOX over PLD in HEYA8 MCTS. Free DOX accumulation was approximately 410,
4,450 and 14,900-fold greater with BCM-DOX than PLD at the periphery, intermediate and core
regions, respectively, compared to 90, 480 and 310-fold observed under NT. MHT is known to
increase the rate of diffusion of drugs, possibly leading to the enhanced penetration of free
drug observed in HEYA8 MCTS.56 This phenomenon was not observed in OV-90 MCTS, where
the difference in free drug accumulation between BCM-DOX and PLD within all three regions
was not notably impacted by MHT. This may be attributed to the size differences between
BCM-DOX and PLD, as has been previously reported,35, 57 as well as the morphology of the two
MCTS types. Since the BCM-DOX formulation is about 1/6 the volume of PLD, differences in
penetration are potentially more evident in the more densely packed structure with
homogenous cell packing of the HEYA8 MCTS in comparison to the OV-90 MCTS, that are
less densely packed and have a heterogeneous structure with areas that include large gaps
between cells (Figure 4-6). Overall, although MHT seems to enhance drug accumulation in
HEYA8 MCTS treated with BCM-DOX and DOX alone in the short term, improved growth
inhibition of MCTS by MHT was not observed after 48 h of drug treatment (Figure 4-10). This
was further investigated in HEYA8 and OV-90 monolayers, revealing that DOX cytotoxicity is
not enhanced by MHT within our experimental setup (Figure 4-11). Previous studies have
reported similar findings in cell monolayers for DOX alone with a short incubation time of 1 h
under NT and MHT (41 C) followed by 72 h post-incubation prior to assessment of viability,
58 although in another study by the same group employing a comparable experimental setup,
a slight reduction in IC50 was observed with MHT in two cell lines, but not in a third.59 The
effect of MHT on the cytotoxicity of DOX remains to be fully elucidated. For instance, effects
may be cell-line dependent. To our knowledge, no other studies published to date have
investigated the effects of MHT on HEYA8 or OV-90 ovarian cancer cell lines.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 148
Overall, in terms of drug accumulation into the MCTS model of residual disease, BCM-DOX
achieves the benefits of an extended release formulation of DOX and results in improvements
in free drug accumulation over PLD, while yielding drug levels approaching that achievable by
exposure to DOX alone. Up to day 4 post-treatment, the extent of MCTS growth inhibition for
each of the formulations corresponded to what was expected based on free drug
accumulation in MCTS and in vitro drug release or drug availability associated with each of
the three DOX delivery strategies. Beyond that point, PLD treatment approaches the level of
growth inhibition induced by BCM-DOX while resulting in a similar rate of volume decline
(Figure 4-10). Hence, it is still to be determined whether the advantages of BCM-DOX over
PLD observed in vitro at early timepoints will result in enhanced tumor growth delay in vivo.
As shown by the in vitro release experiment, serum albumin appears to destabilize BCM-DOX
(Figure 4-3). In vivo, this could potentially lead to premature release of free drug into the
circulation and result in accelerated clearance and lower tumor accumulation of drug.60 As
such, BCM formulations of DOX such as NK911 or SP1049C have reported clinical plasma
AUCs of 1.8 and 3.3 µg h/mL in comparison to 1.6 µg h/mL of DOX alone and 902 µg h/mL
of PLD.61 Furthermore, liposomes have been shown to accumulate at the rim of the tumor
over several days and may act as a depot that releases drug over time,62 whereas BCMs of
15 - 55 nm in diameter are normally cleared after 72h.35, 57 The main question to be addressed
in vivo is whether the initial advantage of BCM-DOX in terms of free drug availability and tumor
penetration would be outweighed by the longer blood circulation half-life and enhanced tumor
residence time of PLD.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 149
4.6. Conclusion
Collectively, the findings presented herein suggest that tissue penetration of drug is not
only an issue for large, unresectable tumors, but also for invisible, microscopic tumor lesions.
Presumably, insufficient drug penetration into these microscopic residual lesions may be a
significant contributing factor that results in relapse in ovarian cancer patients following an
initial positive response to adjuvant chemotherapy. The goal of improving upon PLD in terms
of target tissue availability and penetration of free drug was achieved with BCM-DOX, which
led to superior MCTS growth inhibition in the short term and comparable inhibition in the long
term. Furthermore, while MHT did not enhance DOX cytotoxicity in our in vitro studies, it is
plausible that the effect may be different in vivo, given that MHT has been shown to result in
enhanced blood perfusion and blood vessel permeability in tumors. Clinically relevant animal
xenograft models will be used in future studies to evaluate whether the faster release provided
by BCM-DOX or slow release associated with PLD will ultimately provide superior antitumor
efficacy in vivo.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 150
4.7. Acknowledgements
S.E. is funded by the NSERC CREATE Biointerfaces Training Program and holds an Ontario
Trillium Scholarship. C.A. acknowledges GlaxoSmithKline for an endowed chair in
Pharmaceutics and Drug Delivery. The authors thank Dr. Changhai Lu for his assistance with
synthesis of the block copolymers.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 151
4.8. References
1. Siegel, R.; Naishadham, D.; Jemal, A. Cancer statistics, 2013. Ca-Cancer J Clin 2013, 63, (1), 11-30.
2. Vaughan, S.; Coward, J. I.; Bast, R. C., Jr.; Berchuck, A.; Berek, J. S.; Brenton, J. D.; Coukos, G.; Crum, C. C.; Drapkin, R.; Etemadmoghadam, D.; Friedlander, M.; Gabra, H.; Kaye, S. B.; Lord, C. J.; Lengyel, E.; Levine, D. A.; McNeish, I. A.; Menon, U.; Mills, G. B.; Nephew, K. P.; Oza, A. M.; Sood, A. K.; Stronach, E. A.; Walczak, H.; Bowtell, D. D.; Balkwill, F. R. Rethinking ovarian cancer: recommendations for improving outcomes. Nat Rev Cancer 2011, 11, (10), 719-25.
3. Syrios, J.; Banerjee, S.; Kaye, S. B. Advanced Epithelial Ovarian Cancer: From Standard Chemotherapy to Promising Molecular Pathway Targets - Where Are we Now? Anticancer Res 2014, 34, (5), 2069-2077.
4. Vargas, A. N. Natural history of ovarian cancer. Ecancermedicalscience 2014, 8, 465.
5. Gonzalez-Martin, A.; Sanchez-Lorenzo, L.; Bratos, R.; Marquez, R.; Chiva, L. First-line and maintenance therapy for ovarian cancer: current status and future directions. Drugs 2014, 74, (8), 879-89.
6. Lee, H.; Kim, J.; Kim, H.; Kim, Y.; Choi, Y. A folate receptor-specific activatable probe for near-infrared fluorescence imaging of ovarian cancer. Chem Commun 2014, 50, (56), 7507-7510.
7. Bristow, R. E.; Tomacruz, R. S.; Armstrong, D. K.; Trimble, E. L.; Montz, F. J. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: A meta-analysis. Journal of Clinical Oncology 2002, 20, (5), 1248-1259.
8. Elattar, A.; Bryant, A.; Winter-Roach, B. A.; Hatem, M.; Naik, R. Optimal primary surgical treatment for advanced epithelial ovarian cancer. Cochrane Db Syst Rev 2011, (8).
9. Schmid, B. C.; Oehler, M. K. New perspectives in ovarian cancer treatment. Maturitas 2014, 77, (2), 128-36.
10. Colombo, P. E.; Fabbro, M.; Theillet, C.; Bibeau, F.; Rouanet, P.; Ray-Coquard, I. Sensitivity and resistance to treatment in the primary management of epithelial ovarian cancer. Critical reviews in oncology/hematology 2014, 89, (2), 207-16.
11. Barenholz, Y. Doxil(R)--the first FDA-approved nano-drug: lessons learned. Journal of controlled release : official journal of the Controlled Release Society 2012, 160, (2), 117-34.
12. Markman, M. Pegylated liposomal doxorubicin: appraisal of its current role in the management of epithelial ovarian cancer. Cancer management and research 2011, 3, 219-25.
13. Lawrie, T. A.; Bryant, A.; Cameron, A.; Gray, E.; Morrison, J. Pegylated liposomal doxorubicin for relapsed epithelial ovarian cancer. Cochrane Database Syst Rev 2013, 7, CD006910.
14. O'Brien, M. E.; Wigler, N.; Inbar, M.; Rosso, R.; Grischke, E.; Santoro, A.; Catane, R.; Kieback, D. G.; Tomczak, P.; Ackland, S. P.; Orlandi, F.; Mellars, L.; Alland, L.; Tendler, C.; Group, C. B. C. S. Reduced cardiotoxicity and comparable efficacy in a phase III trial of pegylated liposomal doxorubicin HCl (CAELYX/Doxil) versus conventional doxorubicin for first-line treatment of metastatic breast cancer. Ann Oncol 2004, 15, (3), 440-9.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 152
15. Laginha, K. M.; Verwoert, S.; Charrois, G. J. R.; Allen, T. M. Determination of doxorubicin levels in whole tumor and tumor nuclei in murine breast cancer tumors. Clinical Cancer Research 2005, 11, (19), 6944-6949.
16. Yuan, F.; Leunig, M.; Huang, S. K.; Berk, D. A.; Papahadjopoulos, D.; Jain, R. K. Microvascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograft. Cancer Res 1994, 54, (13), 3352-6.
17. Cronin, B.; Robison, K.; Raker, C.; Moore, R.; Granai, C.; Dizon, D. Pegylated Liposomal Doxorubicin in Recurrent Ovarian Cancer: Is There a Role for Maintenance Therapy? Clinical Ovarian and Other Gynecologic Cancer 2014, 6, (1/2), 17-20.
18. Gibson, J. M.; Alzghari, S.; Ahn, C.; Trantham, H.; La-Beck, N. M. The role of pegylated liposomal doxorubicin in ovarian cancer: a meta-analysis of randomized clinical trials. The oncologist 2013, 18, (9), 1022-31.
19. Coleman, R. L.; Monk, B. J.; Sood, A. K.; Herzog, T. J. Latest research and treatment of advanced-stage epithelial ovarian cancer. Nature Reviews Clinical Oncology 2013, 10, (4), 211-224.
20. Banerjee, S.; Kaye, S. B. New Strategies in the Treatment of Ovarian Cancer: Current Clinical Perspectives and Future Potential. Clinical Cancer Research 2013, 19, (5), 961-968.
21. Safra, T.; Borgato, L.; Nicoletto, M. O.; Rolnitzky, L.; Pelles-Avraham, S.; Geva, R.; Donach, M. E.; Curtin, J.; Novetsky, A.; Grenader, T.; Lai, W. C. V.; Gabizon, A.; Boyd, L.; Muggia, F. BRCA Mutation Status and Determinant of Outcome in Women with Recurrent Epithelial Ovarian Cancer Treated with Pegylated Liposomal Doxorubicin. Molecular cancer therapeutics 2011, 10, (10), 2000-2007.
22. Adams, S. F.; Marsh, E. B.; Elmasri, W.; Halberstadt, S.; VanDecker, S.; Sammel, M. D.; Bradbury, A. R.; Daly, M.; Karlan, B.; Rubin, S. C. A high response rate to liposomal doxorubicin is seen among women with BRCA mutations treated for recurrent epithelial ovarian cancer. Gynecologic oncology 2011, 123, (3), 486-491.
23. McDaniel, J. R.; Dewhirst, M. W.; Chilkoti, A. Actively targeting solid tumours with thermoresponsive drug delivery systems that respond to mild hyperthermia. International journal of hyperthermia : the official journal of European Society for Hyperthermic Oncology, North American Hyperthermia Group 2013, 29, (6), 501-10.
24. van Driel, W. J.; Lok, C. A.; Verwaal, V.; Sonke, G. S. The role of hyperthermic intraperitoneal intraoperative chemotherapy in ovarian cancer. Current treatment options in oncology 2015, 16, (4), 329.
25. Eskander, R. N.; Cripe, J.; Bristow, R. E. Intraperitoneal chemotherapy from Armstrong to HIPEC: challenges and promise. Current treatment options in oncology 2014, 15, (1), 27-40.
26. van de Vaart, P. J.; van der Vange, N.; Zoetmulder, F. A.; van Goethem, A. R.; van Tellingen, O.; ten Bokkel Huinink, W. W.; Beijnen, J. H.; Bartelink, H.; Begg, A. C. Intraperitoneal cisplatin with regional hyperthermia in advanced ovarian cancer: pharmacokinetics and cisplatin-DNA adduct formation in patients and ovarian cancer cell lines. European journal of cancer 1998, 34, (1), 148-54.
27. Chiva, L. M.; Gonzalez-Martin, A. A critical appraisal of hyperthermic intraperitoneal chemotherapy (HIPEC) in the treatment of advanced and recurrent ovarian cancer. Gynecologic oncology 2015, 136, (1), 130-135.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 153
28. Seynhaeve, A. L. B.; Dicheva, B. M.; Hoving, S.; Koning, G. A.; ten Hagen, T. L. M. Intact Doxil is taken up intracellularly and released doxorubicin sequesters in the lysosome: Evaluated by in vitro/in vivo live cell imaging. Journal of Controlled Release 2013, 172, (1), 330-340.
29. Torchilin, V. P. Micellar nanocarriers: Pharmaceutical perspectives. Pharmaceutical research 2007, 24, (1), 1-16.
30. Mikhail, A. S.; Allen, C. Block copolymer micelles for delivery of cancer therapy: transport at the whole body, tissue and cellular levels. Journal of controlled release : official journal of the Controlled Release Society 2009, 138, (3), 214-23.
31. Eetezadi, S.; Ekdawi, S. N.; Allen, C. The challenges facing block copolymer micelles for cancer therapy: In vivo barriers and clinical translation. Advanced drug delivery reviews 2014.
32. Shield, K.; Ackland, M. L.; Ahmed, N.; Rice, G. E. Multicellular spheroids in ovarian cancer metastases: Biology and pathology. Gynecologic oncology 2009, 113, (1), 143-148.
33. Zietarska, M.; Maugard, C. M.; Filali-Mouhim, A.; Alam-Fahmy, M.; Tonin, P. N.; Provencher, D. M.; Mes-Masson, A. M. Molecular description of a 3D in vitro model for the study of epithelial ovarian cancer (EOC). Molecular carcinogenesis 2007, 46, (10), 872-885.
34. Mikhail, A. S.; Eetezadi, S.; Allen, C. Multicellular Tumor Spheroids for Evaluation of Cytotoxicity and Tumor Growth Inhibitory Effects of Nanomedicines In Vitro: A Comparison of Docetaxel-Loaded Block Copolymer Micelles and Taxotere (R). PloS one 2013, 8, (4).
35. Mikhail, A. S.; Eetezadi, S.; Ekdawi, S. N.; Stewart, J.; Allen, C. Image-based analysis of the size- and time-dependent penetration of polymeric micelles in multicellular tumor spheroids and tumor xenografts. International journal of pharmaceutics 2014, 464, (1-2), 168-77.
36. Mikhail, A. S.; Allen, C. Poly(ethylene glycol)-b-poly(epsilon-caprolactone) Micelles Containing Chemically Conjugated and Physically Entrapped Docetaxel: Synthesis, Characterization, and the Influence of the Drug on Micelle Morphology. Biomacromolecules 2010, 11, (5), 1273-1280.
37. Doumas, B. T.; Watson, W. A.; Biggs, H. G. Albumin standards and the measurement of serum albumin with bromcresol green. Clinica Chimica Acta 1997, 258, (1), 21-30.
38. Sodek, K. L.; Ringuette, M. J.; Brown, T. J. Compact spheroid formation by ovarian cancer cells is associated with contractile behavior and an invasive phenotype. Int J Cancer 2009, 124, (9), 2060-70.
39. Aliabadi, H. M.; Elhasi, S.; Mahmud, A.; Gulamhusein, R.; Mahdipoor, P.; Lavasanifar, A. Encapsulation of hydrophobic drugs in polymeric micelles through co-solvent evaporation: the effect of solvent composition on micellar properties and drug loading. International journal of pharmaceutics 2007, 329, (1-2), 158-65.
40. Savic, R.; Azzam, T.; Eisenberg, A.; Maysinger, D. Assessment of the integrity of poly(caprolactone)-b-poly(ethylene oxide) micelles under biological conditions: A fluorogenic-based approach. Langmuir : the ACS journal of surfaces and colloids 2006, 22, (8), 3570-3578.
41. Ahluwalia, A.; Hurteau, J. A.; Bigsby, R. M.; Nephew, K. P. DNA methylation in ovarian cancer - II. Expression of DNA methyltransferases in ovarian cancer cell lines and normal ovarian epithelial cells. Gynecologic oncology 2001, 82, (2), 299-304.
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 154
42. Shaw, T. J.; Senterman, M. K.; Dawson, K.; Crane, C. A.; Vanderhyden, B. C. Characterization of intraperitoneal, orthotopic, and metastatic xenograft models of human ovarian cancer. Molecular Therapy 2004, 10, (6), 1032-1042.
43. Provencher, D. M.; Lounis, H.; Champoux, L.; Tetrault, M.; Manderson, E. N.; Wang, J. C.; Eydoux, P.; Savoie, R.; Tonin, P. N.; Mes-Masson, A. M. Characterization of four novel epithelial ovarian cancer cell lines. In Vitro Cell Dev-An 2000, 36, (6), 357-361.
44. Lee, J. M.; Mhawech-Fauceglia, P.; Lee, N.; Parsanian, L. C.; Lin, Y. G.; Gayther, S. A.; Lawrenson, K. A three-dimensional microenvironment alters protein expression and chemosensitivity of epithelial ovarian cancer cells in vitro. Laboratory investigation; a journal of technical methods and pathology 2013, 93, (5), 528-42.
45. Hirschhaeuser, F.; Menne, H.; Dittfeld, C.; West, J.; Mueller-Klieser, W.; Kunz-Schughart, L. A. Multicellular tumor spheroids: An underestimated tool is catching up again. Journal of biotechnology 2010, 148, (1), 3-15.
46. Sagnella, S. M.; Duong, H.; MacMillan, A.; Boyer, C.; Whan, R.; McCarroll, J. A.; Davis, T. P.; Kavallaris, M. Dextran-based doxorubicin nanocarriers with improved tumor penetration. Biomacromolecules 2014, 15, (1), 262-75.
47. Mohan, P.; Rapoport, N. Doxorubicin as a Molecular Nanotheranostic Agent: Effect of Doxorubicin Encapsulation in Micelles or Nanoemulsions on the Ultrasound-Mediated Intracellular Delivery and Nuclear Trafficking. Mol Pharmaceut 2010, 7, (6), 1959-1973.
48. Karukstis, K. K.; Thompson, E. H. Z.; Whiles, J. A.; Rosenfeld, R. J. Deciphering the fluorescence signature of daunomycin and doxorubicin. Biophys Chem 1998, 73, (3), 249-263.
49. Hanahan, D.; Weinberg, R. A. Hallmarks of Cancer: The Next Generation. Cell 2011, 144, (5), 646-674.
50. Mikhail, A. S.; Allen, C. Block copolymer micelles for delivery of cancer therapy: Transport at the whole body, tissue and cellular levels. Journal of Controlled Release 2009, 138, (3), 214-223.
51. Spiliotis, J.; Halkia, E.; Lianos, E.; Kalantzi, N.; Grivas, A.; Efstathiou, E.; Giassas, S. Cytoreductive Surgery and HIPEC in Recurrent Epithelial Ovarian Cancer: A Prospective Randomized Phase III Study. Annals of surgical oncology 2015, 22, (5), 1570-5.
52. Roviello, F.; Roviello, G.; Petrioli, R.; Marrelli, D. Hyperthermic intraperitoneal chemotherapy for the treatment of ovarian cancer: A brief overview of recent results. Critical reviews in oncology/hematology 2015.
53. Kumar, M.; Lawler, J. Preparation and characterization of negatively charged organic-inorganic hybrid ultrafiltration membranes for protein separation. Separation and Purification Technology 2014, 130, 112-123.
54. Inoue, T.; Yamashita, Y.; Nishihara, M.; Sugiyama, S.; Sonoda, Y.; Kumabe, T.; Yokoyama, M.; Tominaga, T. Therapeutic efficacy of a polymeric micellar doxorubicin infused by convection-enhanced delivery against intracranial 9L brain tumor models. Neuro-oncology 2009, 11, (2), 151-7.
55. Natarajan, J. V.; Nugraha, C.; Ng, X. W.; Venkatraman, S. Sustained-release from nanocarriers: a review. Journal of controlled release : official journal of the Controlled Release Society 2014, 193, 122-38.
56. DeWitt, M. R.; Pekkanen, A. M.; Robertson, J.; Rylander, C. G.; Rylander, M. N. Influence of Hyperthermia on Efficacy and Uptake of Carbon Nanohorn-Cisplatin Conjugates. J Biomech Eng-T Asme 2014, 136, (2).
Chapter 4: Effect of Doxorubicin Delivery Systems on Tissue Penetration 155
57. Lee, H.; Hoang, B.; Fonge, H.; Reilly, R. M.; Allen, C. In Vivo Distribution of Polymeric Nanoparticles at the Whole-Body, Tumor, and Cellular Levels. Pharmaceutical research 2010, 27, (11), 2343-2355.
58. Li, L.; ten Hagen, T. L.; Haeri, A.; Soullie, T.; Scholten, C.; Seynhaeve, A. L.; Eggermont, A. M.; Koning, G. A. A novel two-step mild hyperthermia for advanced liposomal chemotherapy. Journal of controlled release : official journal of the Controlled Release Society 2014, 174, 202-8.
59. Li, L.; ten Hagen, T. L.; Hossann, M.; Suss, R.; van Rhoon, G. C.; Eggermont, A. M.; Haemmerich, D.; Koning, G. A. Mild hyperthermia triggered doxorubicin release from optimized stealth thermosensitive liposomes improves intratumoral drug delivery and efficacy. Journal of controlled release : official journal of the Controlled Release Society 2013, 168, (2), 142-50.
60. Owen, S. C.; Chan, D. P. Y.; Shoichet, M. S. Polymeric micelle stability. Nano Today 2012, 7, (1), 53-65.
61. Sutton, D.; Nasongkla, N.; Blanco, E.; Gao, J. M. Functionalized micellar systems for cancer targeted drug delivery. Pharmaceutical research 2007, 24, (6), 1029-1046.
62. Ekdawi, S. N.; Stewart, J. M.; Dunne, M.; Stapleton, S.; Mitsakakis, N.; Dou, Y. N.; Jaffray, D. A.; Allen, C. Spatial and temporal mapping of heterogeneity in liposome uptake and microvascular distribution in an orthotopic tumor xenograft model. Journal of controlled release : official journal of the Controlled Release Society 2015, 207, 101-111.
156
Combination Index Studies of Olaparib and Doxorubicin in 2D and 3D Spheroid Cell Culture Models of BRCA Deficient and Proficient Serous Ovarian Cancer
Sina Eetezadi, Raquel De Souza and Christine Allen
Experiments by S. Eetezadi. Written by S. Eetezadi and Raquel De Souza.
Illustrations by S. Eetezadi. Edited by C. Allen.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 157
5.1. Abstract
The present study explores the combination of a DNA damaging agent, doxorubicin (DOX),
with a poly(ADP-ribose) polymerase enzyme inhibitor, olaparib (OLP), in order to achieve
optimal synergy of both drugs in serous ovarian cancer. This approach was evaluated and
optimized in 2D monolayers and 3D multicellular tumor spheroids (MCTS) using a genetically
and histologically characterized panel of nine ovarian cancer cell lines with or without HRR
deficiencies, such as BRCA1 or BRCA2 mutations. MCTS are clinically representative of
metastatic spheroids that form during late-stage ovarian cancer, as well as microscopic
disease that remains following cytoreductive surgery. Combination index values of DOX and
OLP were determined using the Chou and Talalay method. The potency of antitumor efficacy
resulting from drug combinations was found to rely heavily on the molar ratios at which the
two drugs are combined, with ratio choice ultimately determining synergy or antagonism. In
general, MCTS growth inhibition was reflective of the patterns predicted by the combination
index values obtained in monolayers. At present, clinical trials are frequently designed to
administer both drugs at their respective maximum tolerated doses, often resulting in trial
termination due to amplified toxicities. Promising combination ratios identified herein warrant
further preclinical and clinical investigation.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 158
5.2. Introduction
Ovarian cancer (OC) is the fourth leading cause of death in women of developed countries,
with dismal survival improvements achieved in the past three decades.1,2 While the median
survival rate for OC as a whole is 40 to 50 % at 10 years post-diagnosis, it falls significantly to
21 % and less than 6 % when the disease is diagnosed at stages III and IV, respectively, as
occurs in the majority of cases.1 About 90 % of OCs are epithelial ovarian cancer (EOC), and
between 60 and 70 % of EOCs are high-grade serous ovarian cancer (HGSOC), the most
common and deadliest form of the disease that is widely associated with p53 gene
mutations.3-5 As a consequence of the initially asymptomatic or ambiguously symptomatic
nature of EOC, most patients present with advanced, metastatic disease at diagnosis, which
is currently treated by tumor cytoreductive surgery prior to carboplatin and paclitaxel adjuvant
chemotherapy.3,6,7 HGSOC patients initially respond well to this regimen (response rate (RR)
>80 %), but as many as 75 % relapse, often within 18 months, with progressively resistant
disease.4,8 Once platinum resistance occurs, selection of the most appropriate treatment for
further management of the disease remains a challenge.1,9 Pegylated liposomal doxorubicin
(PLD, Doxil® / Caelyx®) is predominant in the treatment of OC recurrence, yet it is associated
with low RRs of 9 - 16 %.10,11
Inhibitors of the poly(ADP-ribose) polymerase (PARP) enzymes are a novel class of
molecularly targeted small molecule drugs that shows promise in the setting of resistant
recurrence. The first PARP inhibitor to be clinically assessed is olaparib (OLP, Lynparza®),
which has recently been approved by the FDA for refractory, advanced OC associated with
germline BRCA mutations.12 PARP inhibitors capitalize on deficiencies or aberrations in DNA
damage repair pathways that are present in about half of all HGSOC.1,6,13,14 About 51 % of
HGSOC display deficiencies in homologous recombination repair (HRR), the high-fidelity DNA
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 159
double-strand break (DSB) repair pathway, 15 – 20 % of which display germline or somatic
mutations in the BRCA1 or BRCA2 genes.13,15,16 Through active recruitment of repair proteins,
the PARP1 enzyme plays a critical role in the repair of DNA single-strand breaks (SSBs).17-19
Upon catalytic inhibition by a PARP inhibitor, PARP1 becomes trapped at SSB sites, which
leads to stalling of the replication fork and subsequent formation of SSBs.20 Additionally,
PARP inhibitors hinder the role of PARP1 in DSB repair given that PARP1 is reported to aid in
the recruitment of HRR proteins.17 On its own, PARP inhibition is relatively non-toxic to cells
with intact HRR capabilities. However, synthetic lethality occurs upon PARP inhibition in cells
with deficiencies in HRR due to BRCA mutations18,19 or the “BRCAness” phenotype, which
renders cells more susceptible to DNA damage.13,17 PARP inhibitors have shown promise
clinically in both BRCA-deficient and proficient HGSOCs.21-23 In a recent phase II trial involving
recurrent, platinum-sensitive HGSOC patients, it was found that the inclusion of the PARP
inhibitor olaparib into the paclitaxel-carboplatin standard regimen significantly improved
progression-free survival, especially in the patient subpopulation with BRCA mutations.24
Due to heightened susceptibility to SSB and compromised repair capabilities, the
combination of PARP inhibitors with DNA-damaging agents has the potential to
synergistically affect HGSOC cells. The combination of a PARP inhibitor with doxorubicin
(DOX) is particularly promising in HGSOC. DOX inhibits DNA polymerase activity and
topoisomerase II, and induces free radical formation, causing DSBs and SSBs.11 Further, DOX
shows enhanced efficacy in HRR-deficient OC, as BRCA mutation carriers treated with PLD
show greater RR and longer overall survival (OS) compared to sporadic cases.10,11 Moreover,
PARP inhibitors have been shown to enhance DOX antitumor activity in p53-deficient
cancers,25,26 which is promising for HGSOC, as 90-100% of cases show p53 mutations.13,14,27
Further, PARP inhibitors increase topoisomerase II expression, thereby specifically
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 160
potentiating the action of DOX.28 Yet, in clinical practice, dose limiting toxicities have been a
major hurdle for such combinations, and have resulted in the need to halt a number of clinical
trials and/or to reduce drug doses mid-trial.6,29 Generally, combination therapies have been
designed with the assumption that maximal therapeutic effect is achieved when the
maximum tolerated dose (MTD) of each drug is employed. However, it has been shown that
the effect of drug combinations does not only depend on the nature of each drug’s distinct
mechanism of action, but also on the molar ratios at which they have been combined.30,31 For
example, a given drug may induce a strong cell defense response that in turn reduces the
therapeutic effect of a second drug when given in combination. Hence, the choice of ratio is
important; while certain ratios are synergistic, others can be antagonistic.32
Based on the promising potential of combining PARP inhibitors with DNA damaging
agents, the present work investigated optimized combinations of the PARP inhibitor OLP with
the DNA-damaging agent DOX by identifying and evaluating the synergistic potential of
specific molar ratios.32 To assess this strategy in a setting representative of clinical disease
presentation, we sought to employ in vitro models encompassing the heterogeneity of EOC.
About 100 OC cell lines are currently available through public cell banks;33 however, the
histopathological origin and genomic characteristics are unknown for the majority of these
lines, limiting their utility in accurately representing aspects of interest of the disease. 33,34 For
this study, multiple cell lines were carefully selected, based on data from three recent large-
scale cell line characterization studies, to histologically and molecularly represent an array of
serous and HGSOCs, including HRR-deficient variations.33-35 The potential of each cell line to
form multicellular tumor spheroids (MCTS) using an established liquid-overlay technique was
assessed.36 MCTS are spherical, 500 µm structures comprised of an outer region of
proliferative cells surrounding intermediate regions of quiescent cells.37,38 MCTS are clinically
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 161
representative 3D models of metastatic cell spheroids that form in advanced disease states
as well as microscopic residual disease that remains after cytoreductive surgery in
HGSOC.39,40 In particular, to the best of our knowledge, MCTS of BRCA mutated cell lines have
not been reported in the literature thus far. In this study, combination effects of DOX and OLP
at ten different molar ratios were evaluated in 2D and 3D cell models of the selected ovarian
cancer cell lines using the well-established Chou and Talalay method to evaluate treatment
synergy.41
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 162
5.3. Materials and Methods
5.3.1. Materials
DOX was purchased from Polymed (Houston, TX, USA). OLP was obtained from
Tongchuang Pharma, Suzhou Co. Ltd (China). Acetic acid, P-nitrophenyl phosphate, TritonTM-
X-100, G-418 disulfate salt, sodium pyruvate and sodium acetate were purchased from
Sigma-Aldrich (Oakville, ON, CA). MCDB 105 and DMEM cell media were obtained from Sigma-
Aldrich, while all other cell media, sterile phosphate buffered saline (pH = 7.4), CellMaskTM
Green plasma membrane stain and fetal bovine serum (FBS) were obtained from Life
Technologies (Burlington, ON, CA). MEGM™ Mammary Epithelial Cell Growth Medium Kit was
purchased from Lonza (Mississauga, ON, CA). UWB1.289, UWB1.289+BRCA1, OV-90 and
SKOV3 cell lines were purchased from the American Type Culture Collection (ATCC;
Manassas, VA, USA). PEO1, PEO4 and COV362, originally from the European Collection of Cell
Cultures (ECACC, Public Health England; Salisbury, UK), were purchased through Sigma-
Aldrich. HEYA8 was obtained from the M. D. Anderson Cancer Center (Houston, TX, USA).
OVCAR8 was obtained from the Biological Testing Branch of the National Cancer Institute
(NCI; Frederick, MD, USA).
5.3.2. Cell Culture
All cell lines were cultured as monolayers in the media recommended by the supplier, and
maintained in a dedicated cell culture incubator at 37 °C, 5 % CO2 atmosphere and 90 %
relative humidity. UWB1.289 and UWB1.289+BRCA1 were maintained in a 1:1 mixture of
RPMI-1640 and MEGMTM including the five components of the MEGM™ Mammary Epithelial
Cell Growth Medium Kit and 3 % FBS. UWB1.289+BRCA1 growth media was additionally
supplemented with 200 µl/ml G-418 to maintain BRCA1 expression. HEYA8 and OVCAR8
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 163
were grown in RPMI-1640 media with 10 % FBS. PEO1 and PEO4 were maintained in RPMI-
1640 media with 10 % FBS and additionally supplemented with 2 mM sodium pyruvate.
COV362 was cultured in DMEM and SKOV3 in McCoy's 5a medium, both supplemented with
10 % FBS, and OV-90 was grown in a 1:1 mix of MCDB 105 and Media 199 supplemented with
15 % FBS. A 1 % penicillin/streptomycin solution was added to all media. Cultures were
regularly refreshed from frozen stocks to keep passage numbers below 25.
5.3.3. Cell Line Authentication
Short tandem repeat (STR) analysis was performed by The Centre for Applied Genomics
at The Hospital for Sick Children (Toronto, ON, CA) to verify the authenticity of cell lines
employed here. To this end, the GenePrint®10 System (Promega Corporation; Madison, WI,
USA) was used, following manufacturer’s instructions. Briefly, the kit employs an STR
multiplex assay that amplifies nine loci (the ANSI Standard (ASN-0002) recommends eight)
and the Amelogenin gender-determining marker in a single PCR amplification. To prepare
each reaction, 10 ng of genomic DNA was incubated with a master mix (5 L GenePrint® 10
5X Master Mix, 5 L GenePrint® 10 5X Primer Pair Mix, water for a 25 L total volume). The
cycling conditions were 96 °C for 1 min; 30 cycles of (94 °C for 10 s, 59 °C for 1 min, 72 °C for
30 s); 1 cycle of 60 °C for 10 min and 4 °C maintenance. Samples were run on an AB 3130
Genetic Analyzer using POP7, Promega’s internal lane standard 600, dye set F and analyzed
in GeneMapper v3.7. Since allelic nomenclature for the 10 loci is standardized worldwide, the
obtained STR profiles were compared to those made available by each cell line’s
corresponding repository or previous publications (Table 5-5).
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 164
5.3.4. BRCA1 and BRCA2 Mutation Sequencing
Regions pertaining to BRCA1 and BRCA2 gene variants of interest were amplified for
Sanger sequencing by The Centre for Applied Genomics at The Hospital for Sick Children
(Toronto, ON, CA) to confirm BRCA mutation status of relevant cell lines. Primer sequences
are outlined in Table 5-1: Primers used in sequencing BRCA1 and BRCA2 regions of interest..
Briefly, each 25 L PCR reaction was prepared with 2 mM dNTPs (2.5 L), 10X PCR buffer II
(2.5 L; Life Technologies), 25 mM MgCl2 (1.3 L), AmpliTaq DNA polymerase (0.25 L; Life
Technologies), 5M betaine (5 L), water (10.45 L), 10 µM primers (1 L each) and genomic
DNA (1 L; normalized to 50-100 ng/L). Samples were amplified by PCR using the following
cycling conditions: 30 cycles of 94 ºC for 10 s, 60 ºC for 30 s and 72 ºC for 1 min. Agarose gel
was used to visualize the PCR products, which were purified using the AxyPrep PCR Clean-up
Kit (Axygen Biosciences). A total of 50 ng of purified product was Sanger sequenced from the
forward primer using BigDye3.1 chemistry on a ABI3730XL capillary DNA analyzer (Life
Technologies).
Table 5-1: Primers used in sequencing BRCA1 and BRCA2 regions of interest.
Line Gene Variant (effect; hg19 coordinate)
Forward primer Reverse primer
UWB1.289 UWB1.289+BRCA1
BRCA1 2594delC; chr17:41,245,073 AATGACTGGCGCTTTGAAAC
AATGCTGAAGACCCCAAAGA
PEO1 PEO4
BRCA2 Y1655Y/X/L; chr13:32,913,457 TTAGCCATCAATGGGCAAAG
TCTGGTTGACCATCAAATATTCC
COV362 (mutation 1)
BRCA1 c.G4095+1T; chr17:41,243,451 GCTCATTCAGTCAAAGATGACG
CACAGTGCAGTGAATTGGAAG
COV362 (mutation 2)
BRCA1 c.2611_2612ins1; chr17:41,244,935
CTGACCAACCACAGGAAAGC
GGAAGGCAAAAACAGAACCA
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 165
5.3.5. Doubling Time
For evaluation of cell proliferation rates, subconfluent cells were harvested, seeded onto
6-well plates at a seeding density of 100,000 cells per well and allowed to adhere overnight.
Over a 6-day period, cells were detached at selected timepoints with TrypLETM Express cell
dissociation solution and counted using a Countless II FL automated cell counter (Life
Technologies). Cell doubling time was determined by nonlinear regression based on an
exponential growth equation that was fit to the cell counts at each timepoint using Graphpad®
Prism software.
5.3.6. Monolayer IC50 Evaluation
Monolayer cell cytotoxicity for DOX, OLP and all drug combinations was determined using
the acid phosphatase (APH) assay based on a method previously published by our group.36
Briefly, subconfluent cells were harvested, seeded onto 96-well plates and allowed to adhere
overnight prior to incubation for 72 h using ten 1:4 serial dilutions of single drug or drug
combinations (n = 6 wells per dilution). Seeding density was determined separately for each
cell line so that the control was maintained subconfluent through to the end of the experiment
(5000 cells/well for OV-90 and COV362; 4000 cells/well for PEO1 and PEO4; 3000 cells/well
for UWB1.289, UWB1.289+BRCA1 and SKOV3; 1000 cells/well for OVCAR8 and HEYA8 cell
lines). Following treatment, cells were washed with PBS prior to assessment of viability using
the APH assay. In brief, cells were incubated with 100 μL of freshly prepared reaction buffer
(sodium acetate buffer at pH 5.5 containing 1 % Triton-X supplemented with 2 mg/ml p-
nitrophenyl phosphate) for 2 h at 37 °C. Cell viability was determined by measuring the UV
absorbance at 485 nm using an automated 96-well plate reader (Synergy 2 from BioTek;
Winooski, VT, USA) following addition of 10 μL 1 M sodium hydroxide to each well. Average
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 166
absorbance (A) for each drug concentration of three independent experiments performed on
different days was expressed relative to controls as follows:
𝑣𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦 [%] =(𝐴𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 – 𝐴𝑚𝑒𝑑𝑖𝑎)
(𝐴𝑐𝑜𝑛𝑡𝑟𝑜𝑙 – 𝐴𝑚𝑒𝑑𝑖𝑎)
The concentration that kills 50 % (IC50, or fraction affected (Fa) = 0.5) and 75 % (Fa = 0.75)
of cells was determined by fitting the data to the Hill equation using Graphpad® Prism
software.
5.3.7. Determination of Combination Index (CI) Values
CI values were determined according to a widely used method established by Chou and
Talalay.42,43 Briefly, to determine each CI value the following monolayer cytotoxicity studies
were conducted: 1) DOX as a single drug; 2) OLP as a single drug; and 3) DOX-OLP
combinations at a series of specific molar ratios. For all experiments, Fa = 0.5 and Fa = 0.75
molar drug concentrations were determined, as outlined above. CI values were then
calculated for each combination and for both effect levels based on the formula below for
mutually exclusive drugs as established by Chou and Talalay, where DSD is the concentration
of DOX that kills 50% of cells (for Fa = 0.5) or 75% of cells (for Fa = 0.75), DSO is the
concentration of OLP at Fa = 0.5 or Fa = 0.75, DCD is the concentration of DOX in combination
with OLP at Fa = 0.5 or Fa= 0.75, and DCO is the concentration of OLP in combination with DOX
at Fa = 0.5 or Fa = 0.75:
𝐶𝐼 = 𝐷𝐶𝐷
𝐷𝑆𝐷+
𝐷𝐶𝑂
𝐷𝑆𝑂+
𝐷𝐶𝐷𝐷𝐶𝑂
𝐷𝑆𝐷𝐷𝑆𝑂
Based on this method, CI values are indicative of strong synergism (< 0.7), synergism (0.7
– 0.9), additive effect (0.9 – 1.1), antagonism (1.1 – 3.3) or strong antagonism (> 3.3).42
Microsoft Excel was used to generate a tricolor system based on these values, where strong
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 167
synergism is represented by green, additive effect by yellow and strong antagonism by red.
The software interpolates the color of each value in between these constraints accordingly.
5.3.8. MCTS Growth Studies
MCTS establishment was attempted for all cell lines using a method previously reported
by our group.44 Subconfluent cells were seeded onto non-adherent 96-well round-bottomed
Sumilon PrimeSurfaceTM spheroid plates (MS-9096U; Sumitomo Bakelite, Tokyo, Japan)
containing the recommended complete growth media for each cell line and incubated for 10
days at 37°C, such that each well contained a single MCTS. Images were taken using a light
microscope with a 10x objective lens (VistaVisionTM; VWR, Mississauga, ON, CA) connected
to a digital camera (DV-2B; VWR). The volume of each MCTS was determined according to a
published method.44 Briefly, the 2D cross-sectional area was measured using an automated
image analysis macro developed for use with the ImageJ software package (Version 1.48V)
and volumes were calculated assuming spherical shape. Growth curves were fit to the
Gompertz growth equation of tumor growth.44 For morphology studies, MCTS were incubated
for 1 h with 1X CellMaskTM Green plasma membrane stain (Life Technologies) in PBS and
imaged using a Zeiss LSM700 confocal microscope (Carl Zeiss AG; Oberkochen, Germany)
with a FITC filter (Ex 522 nm, Em 535 nm).
5.3.9. MCTS Growth Inhibition Studies
MCTS of OVCAR8, OV-90 and HEYA8 cell lines were grown as described above. Initial cell
seeding density was chosen such that MCTS reached a diameter of about 500 µm after 7
days. Prior to combination studies, MCTS were treated with DOX (2 µM, 1 µM and 0.2 µM) for
72 h to determine a drug concentration sufficient to inhibit MCTS growth while still allowing
MCTS to remain structurally intact, such that assessment of different molar combinations of
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 168
DOX with OLP would be possible. MCTS were incubated for 72 h with single drug or drug
combinations at selected molar ratios (n = 6 per treatment group for all studies). Following
the 72 h treatment period, MCTS were washed with media. Every other day thereafter, 50 %
of media was replaced prior to imaging for MCTS volume determination, as described above.
5.3.10. Statistical Analysis
Results are reported as mean of at least 3 independently conducted experiments
standard deviation (SD). Differences between treatments were compared using one-way
ANOVA’s F-test followed by post-hoc analysis using Bonferroni correction, with significance
assigned at p 0.05.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 169
5.4. Results
5.4.1. Properties of the HGSOC Cell Panel
Based on three large-scale cell line characterization studies,33,34,45 nine ovarian cancer cell
lines were selected to represent EOC, particularly the HGSOC subtype (Table 5-2). Cell line
authentication was performed on the basis of STR analysis (Table 5-5). UWB1.289 is derived
from a tumor of papillary serous histology, the most common form of OC.46 This cell line’s
BRCA1 wild-type allele is absent as the gene is mutated within exon 11. UWB1.289 was
transfected with a BRCA1 construct to restore BRCA1 function, creating the
UWB1.289+BRCA1 cell line.46 PEO1 and PEO4 were derived from the ascites of the same
patient before and after development of resistance to platinum-based chemotherapy,
respectively.47 Subsequent analysis revealed that PEO1 had a BRCA2 loss-of-function
mutation, while PEO4 showed a secondary mutation that caused functional restoration of
BRCA2 protein, causing PEO4 to be more resistant to platinum or PARP inhibitors than
PEO1.48 COV362 is another cell line with BRCA1 loss-of-function which was originally
described as being derived from a tumor of endometrioid subtype,49 with later analysis
qualifying it as HGSOC.33,34 OVCAR8 was derived from a platinum-refractory patient who
showed disease progression even after receiving high doses of platinum.50 BRCA1 promoter
methylation was later identified in this cell line.51 This epigenetic modification results in lack
of BRCA1 mRNA expression, and has been clinically reported in 8.1 % 52 and 13.3 % 53,54 of
EOCs and in 14.8 % of HGSOC cases.55 OV-90, derived from previously untreated ovarian
malignant ascites, forms tumors of serous histology and expresses functional BRCA1 and
BRCA2.56 HEYA8 was derived from a mouse xenograft tumor after three passages of the
patient-derived cells in immunocompromised mice.57,58 The original xenografted tumor was
described under the name HX-62 as a serous ovarian cystadenocarcinoma derived from
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 170
peritoneal metastasis.51,59 SKOV3 was established in 1973 and is one of the most widely used
ovarian cancer cell lines based on publication records.34,60 SKOV3 has been characterized as
being of either serous or clear cell histotype.33
Table 5-2: Characteristics of the HGSOC cell panel selected for these studies.
Line Histology BRCA Status MCTS T2 Supplier
UWB1.289 Serous BRCA1 deficient No 54 h ATCC UWB1.289+BRCA1 Serous BRCA1 restored No 38 h ATCC PEO1 HGSOC BRCA2 deficient No 40 h ECACC PEO4 HGSOC BRCA2 restored No 58 h ECACC COV362 HGSOC BRCA1 deficient No 99 h ECACC OVCAR8 Serous BRCA1 p.m. Yes 29 h NIH OV-90 HGSOC BRCA1 and 2 wt Yes 33 h ATCC HEYA8 Serous BRCA1 and 2 wt Yes 16 h MDA SKOV3 Clear Cell BRCA1 and 2 wt Yes 41 h ATCC
* Abbreviations: HGSOC = high-grade serous ovarian cancer; MCTS = multicellular tumor spheroid formation; NIH = National Institutes of Health; p.m. = promoter methylation; wt = wild type; T2 = cell doubling time; MDA = M. D. Anderson Cancer Centre.
While HEYA8 expresses wild-type p53,61 all other cell lines employed here have been
reported to have a p53 mutation, a common trait of HGSOC 33,34,62. BRCA status was
confirmed by sequencing regions of BRCA1 and/or BRCA2 gene variants of interest for
UWB1.289, UWB1.289+BRCA1, PEO1, PEO4 and COV362 cells (Table 5-3).
The doubling time (T2) of all cell lines was determined by serial cell counts and curve fitting
to an exponential growth equation. The cell line with the highest proliferation rate was HEYA8
(16 h), while COV362 had the lowest rate (99 h). T2 appeared to be unrelated to BRCA
mutation status or other factors assessed. The BRCA1-restored UWB1.289+BRCA1 cell line
displayed a T2 that was 30 % shorter than that of UWB1.289. Conversely, T2 for the BRCA2-
restored cell line PEO4 was 31 % faster than for PEO1, which are cells derived from the same
patient at different stages of the disease.47
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 171
Table 5-3: Confirmation of BRCA1 or 2 status for PEO1, PEO4, UWB1.289, UWB1.289+BRCA1
and COV362 ovarian cancer cell lines. Gene sequences were confirmed as described in the
Materials and Methods section.
Cell Line Expected BRCA
Mutation Consequence of Mutation
Sequencing Results
Consequence
PEO1 BRCA2 Y1655X 48 No detectable BRCA2 protein
BRCA2 Y1655X BRCA2 non-functional
PEO4 BRCA2 Y1655Y 48 Restored BRCA2
functional protein
BRCA2 Y1655Y BRCA2 restored
COV362 BRCA1 G4095+1T
and 2611_2612ins1 34 No detectable BRCA1 protein
BRCA1 G4095+1T and 2611_2612ins1
BRCA1 non-functional
UWB1.289 BRCA1 2594delC homozygous 63
No detectable BRCA1 protein
BRCA1 2594delC homozygous
BRCA1 non-functional
UWB1.289+BRCA1 BRCA1 2594delC heterozygous 63
Restored BRCA1 functional
protein
BRCA1 2594delC heterozygous
BRCA1 restored
5.4.2. Cell Monolayer Viability Studies of DOX and OLP treatment
All cell lines were treated with DOX or OLP over 72 h for determination of IC50 values (Figure
5-1). In general, the DNA damaging agent DOX was observed to be approximately three orders
of magnitude more cytotoxic than OLP. UWB1.289+BRCA1 was the most resistant cell line to
DOX and the second most resistant cell line to OLP. On the other hand, the BRCA2-proficient
cell line PEO4 was relatively sensitive to DOX when compared to the other cell lines but
displayed the lowest sensitivity to OLP. OV-90 cells were relatively sensitive to OLP but this
cell line was found to be one of the least sensitive to DOX. However, certain cell lines showed
similar relative sensitivity to both drugs. For instance, PEO1 and HEYA8 were the most
sensitive cell lines in general, and showed similar levels of susceptibility to both DOX and OLP.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 172
Consideration of the IC50 values with respect to BRCA status (Table 5-2) revealed that
BRCA1 or 2 deficiencies are not sufficient to predict sensitivity to OLP, or DOX (Figure 5-2).
For this analysis, cell lines were categorized as “BRCA proficient” and “BRCA deficient”
according to Table 5-2, and for the purpose of this analysis, OVCAR8 (light orange) was
classified as “BRCA deficient”. As shown in Figure 5-2, cell lines that are BRCA deficient
showed a very narrow distribution of IC50 values for both drugs, whereas BRCA-proficient cells
showed a more broad distribution in IC50 values, especially for OLP. When the IC50 values for
each group were considered together, comparison between the two groups found no
statistically significant difference. However, the BRCA proficient PEO4 (purple) and
UWB1.289+BRCA1 (dark blue) were indeed two to three times more resistant to OLP than
their BRCA-deficient counterparts PEO1 and UWB1.289, respectively.
Figure 5-1: IC50 values for cell lines after 72 h single-drug treatment with DOX (left) or OLP
(right) alone.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 173
5.4.3. DOX-OLP Combination Studies
Following the assessment of single drug cytotoxicity, DOX and OLP combination
treatments at 10 different molar ratios were evaluated over a 72 h period. For each ratio, three
independent experiments were performed on different days, and the molar drug
concentrations required to kill 50 % (Fa = 0.5) and 75 % (Fa = 0.75) of cells were determined.
The median-effect algorithm based on the widely used method established by Chou and
Talalay was employed to calculate the CI for each ratio at both effect levels (i.e. Fa = 0.5 and
Fa = 0.75) as outlined in the methods section.64 The CI equation was used to generate CI
values, which categorize the effect of the drug-drug combination at specific ratios as
synergistic, additive or antagonistic, as described in the methods section. A synergistic effect
implies that the two drugs are more effective together than what would be expected from
adding the effects of each drug when used separately.41,42
Figure 5-2: IC50 values for cell lines, grouped as BRCA deficient or proficient, following
treatment with DOX (left) or OLP (right). When the IC50 values for BRCA deficient (i.e. mutated
or methylated) cell lines are considered together and compared to values for the BRCA
proficient cell lines, no significant difference is identified.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 174
Figure 5-3: Summary of CI values for the combinations of DOX and OLP following 72h
incubation in nine ovarian cancer cells lines at the indicated molar ratios required for Fa = 0.5
(A) and Fa = 0.75 (B). Each CI value was calculated and a heat map was generated as outlined
in Materials and Methods on the basis of three independent IC50 experiments performed on
different days for DOX, OLP and each combination ratio.
(A)
Fa =
0.5
DO
X:O
LP1:
100
1:50
1:10
1:5
1:2
2:1
5:1
10:1
50:1
100:
1
UW
B1.
289
0.21
0.24
2.10
1.89
1.72
0.59
0.62
0.44
0.32
0.15
UW
B1.
289+
BR
CA
10.
570.
780.
970.
700.
700.
570.
430.
600.
360.
34
PEO
10.
951.
150.
901.
021.
171.
461.
331.
190.
550.
96
PEO
41.
161.
151.
421.
491.
811.
501.
351.
571.
380.
92
CO
V36
20.
940.
691.
201.
341.
530.
851.
061.
220.
890.
81
OV
CA
R8
0.90
1.07
0.94
1.24
1.24
1.34
1.33
0.61
0.69
1.00
OV
-90
0.75
0.81
0.63
0.84
0.74
0.77
0.86
0.95
0.68
0.79
HEY
A8
0.71
0.84
0.87
0.86
1.18
1.02
0.64
0.37
0.17
0.27
SKO
V3
1.22
1.00
0.91
0.96
0.99
1.24
1.03
0.91
0.95
0.61
Ave
rage
0.81
0.83
1.04
1.24
1.32
1.09
1.05
0.98
0.74
0.78
(B)
Fa =
0.7
5
DO
X:O
LP1:
100
1:50
1:10
1:5
1:2
2:1
5:1
10:1
50:1
100:
1
UW
B1.
289
0.41
0.37
1.61
1.41
1.28
0.56
0.66
0.65
0.59
0.59
UW
B1.
289+
BR
CA
10.
480.
640.
670.
700.
620.
500.
480.
610.
500.
51
PEO
10.
821.
070.
710.
700.
781.
131.
091.
120.
720.
91
PEO
40.
770.
911.
091.
111.
030.
721.
161.
391.
401.
05
CO
V36
20.
821.
001.
221.
341.
331.
071.
621.
661.
351.
30
OV
CA
R8
0.91
0.93
0.96
1.22
1.19
1.23
1.24
0.80
0.82
1.05
OV
-90
0.61
0.63
0.53
0.60
1.21
0.55
0.76
0.85
0.81
0.80
HEY
A8
0.48
0.54
0.61
0.63
0.73
0.75
0.51
0.49
0.35
0.41
SKO
V3
0.69
0.81
0.60
0.65
0.60
0.96
0.89
0.99
1.13
1.21
Ave
rage
0.64
0.72
0.96
1.00
1.05
0.77
0.92
0.92
0.83
0.82
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 175
In the case of OLP+DOX combinations, it is important to evaluate combination effects at
different effect levels (i.e. Fa) since dose-response curves for both DOX and OLP are not linear.
Consequently, the cytotoxicity of these drugs changes incrementally with dose, and the
relative cytotoxicity contribution of each drug will differ with varying Fa.41,65 For cancer
chemotherapy, higher effect levels of Fa = 0.8, for instance, are commonly used and more
relevant for treatment, where complete cancer cell eradication is the goal.42,66,67 The CI results
are presented in a “heat map” for each ratio of DOX:OLP at both Fa levels (Figure 5-3, A, B).
Overall, the combination of OLP and DOX resulted in synergistic effects when the molar
concentrations of each drug differed from each other to a greater extent, such as DOX:OLP at
1:100, 1:50, 50:1 and 100:1. More equimolar ratios of 1:2, 1:5 and 1:10 as well as the inverse
resulted in an additive effect. For Fa = 0.75, a trend towards more synergy was generally
observed in comparison to Fa= 0.5, with the exception of the 50:1 and 100:1 ratios (Figure 5-3,
B). Notably, OLP concentrations required when in combination with DOX were 10 to 1000-fold
below the IC50 values of OLP as a single agent. Neither BRCA status nor sensitivity to DOX or
OLP single-drug treatment appeared to be predictive of DOX-OLP synergism. For example,
although the BRCA1-deficient cell line UWB1.289 was more sensitive to single drug
treatments by a factor of three to four in comparison to UWB1.289+BRCA1, the CI values for
UWB1.289 were lower than for UWB1.289+BRCA1 at certain ratios (1:100, 1:50, 10:1, 50:1 and
100:1) but not others (1:10, 1:5, 1:2, 2:1, 5:1). CI values for the BRCA2-deficient cell line PEO1
were generally lower than for its BRCA2-proficient counterpart PEO4, except for the 1:50 and
100:1 ratios. Furthermore, while HEYA8 and PEO1 are both quite sensitive to DOX, most
DOX:OLP ratios resulted in synergism in the former cell line and an additive effect or
antagonism in the latter. Conversely, DOX-OLP combinations for UWB1.289+BRCA1, the most
DOX-resistant cell line, were found to be very effective, especially at Fa = 0.75. As was the
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 176
case with DOX, single-drug OLP sensitivity did not correlate with combination outcomes.
While PEO4 showed the strongest resistance to single-drug OLP as well as overall
antagonistic effects in response to DOX-OLP combinations, OVCAR8 was highly sensitive to
single-drug OLP but the DOX-OLP combination did not yield particularly synergistic effects. In
summary, the data suggests that synergism of the DOX:OLP combination is drug-ratio-
dependent and more pronounced with higher Fa, but cannot be predicted from BRCA status
or sensitivity to single-drug treatments.
5.4.4. MCTS Model Development and Morphology Studies
For further assessment of the DOX-OLP combination effects in 3D cell culture, growth of
each cell line as MCTS was attempted using a previously established liquid-overlay method.36
3D cell cultures based on MCTS with diameters of around 500 µm can be used as in vitro
models of microscopic ovarian cancer residual disease, which occurs in patients after optimal
cytoreductive surgery.40 For establishment of MCTS, cells were seeded onto non-adherent 96-
well plates and growth was assessed over 10 days using a light microscope (Table 5-4). In
order to qualify as a suitable 3D cell culture model, cell lines were required to form a coherent
spheroid structure that grows in volume over the 10-day timeframe. These criteria ensured
that the MCTS differed fundamentally from simple cell aggregates that form shortly after
incubation (1-2 days). All cell lines were evaluated for their ability to form MCTS, finding that
only OVCAR8, OV-90 and HEYA8 were capable of forming MCTS that complied with the
required criteria outlined above. SKOV3 and PEO4 initially appeared to form MCTS, but these
were not sufficiently stable to withstand the required experimental manipulation. Additionally,
SKOV3 aggregates did not increase in volume over time. The remaining cell lines formed a
layer of cells at the bottom of each well, failing to form MCTS. Interestingly, unexpected
differences were observed within each of the two cell line pairs. Despite the fact that the PEO1
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 177
and PEO4 cell lines were derived from the same patient, they resulted in very distinct
morphologies in vitro. PEO1 cells appeared not to aggregate and simply accumulated at the
bottom of the well, whereas PEO4 cells aggregated and grew in volume, but were not stable
enough to be used as MCTS models. In contrast, the UWB1.289 and UWB1.289+BRCA1 cell
line pair showed very similar structure to each other, with monolayers outgrowing the
perimeter of the microscope objective.
Table 5-4: MCTS growth studies for each cell line. Images show MCTS grown using the liquid-
overlay technique 36 after initial seeding of 3000 cells per well. All scale bars represent 100
µm.
Name Day 3 Day 7 Day 10
UWB1.289
UWB1.289+BRCA1
PEO1
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 178
Name Day 3 Day 7 Day 10
PEO4
COV362
OVCAR8
OV-90
HEYA8
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 179
Name Day 3 Day 7 Day 10
SKOV3
Cellular organization was assessed for OVCAR8, OV-90 and HEYA8 MCTS using confocal
microscopy following staining of cell membranes with CellMaskTM Green fluorescent stain
(Figure 5-4). The three cell lines displayed distinct patterns of cellular organization upon MCTS
formation. HEYA8 MCTS showed a densely packed, homogenous pattern of similarly sized
cells throughout the entire structure. OVCAR8 MCTS were also densely packed, but cells were
of different sizes and shapes and organized in an irregular fashion. Finally, OV-90 MCTS
showed the most heterogeneous structure, with areas of very tightly packed cells
interspersed with voids in other areas. Single OV-90 cells appeared to be more uniform than
OVCAR8 cells in MCTS, but were of various sizes, in contrast to HEYA8.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 180
5.4.5. MCTS Growth and DOX Sensitivity Studies
In accordance with the criteria outlined above, MCTS for OVCAR8, OV-90 and HEYA8
displayed an exponential growth pattern that fit well with the Gompertz equation for tumor
growth (Figure 5-5, A, C, E).36 Following seeding of 1000, 3000 and 5000 cells per well, MCTS
volume was monitored over 13 days. OVCAR8 and HEYA8 MCTS initially grew very quickly
until day 9, when growth rates began to decline (Figure 5-5, A, E). By day 13, all OVCAR8 and
HEYA8 MCTS reached the same size regardless of initial cell seeding density. OV-90 MCTS
showed the same size convergence pattern, but at a slower overall growth rate (Figure 5-5,
C). Furthermore, in comparison to OVCAR8 and HEYA8 MCTS, greater size variations were
observed in OV-90 MCTS. Subsequent studies were conducted with initial cell seeding
numbers of 3000 cells/well for OVCAR8 and OV-90, and 1000 cells/well for HEYA8. This
yielded a diameter of about 500 µm after seven days of growth for all MCTS.
OVCAR8 OV-90 HEYA8
Figure 5-4: Representative confocal images of cellular organization of OVCAR8, OV-90 and
HEYA8 MCTS. MCTS were incubated for 1 h with 1X CellMaskTM Green plasma membrane
stain in PBS, and the fluorescent stain was detected using confocal microscopy. Scale bars
represent 100 µm.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 181
Figure 5-5: MCTS volume (left panels) and growth inhibition upon DOX treatment (right panels) for OVCAR8 (A, B), OV-90 (C, D) and HEYA8 (E, F) ovarian cancer cell lines. MCTS volume as a function of time is shown on a log plot fit to the Gompertz equation of tumor growth following seeding of 1000, 3000 and 5000 cells per well (A, C, E). The dotted line indicates a diameter of 500 µm, which was chosen as baseline for further studies. MCTS growth inhibition was assessed following treatment with DOX for a 72 h period (B, D, F). Every other day, 50 % of the media was replaced prior to imaging and determination of MCTS volume.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 182
MCTS were used to determine the DOX concentration needed for subsequent DOX:OLP
combination studies (Figure 5-5, B, D, F). Specifically, the goal was to identify a DOX
concentration that resulted in MCTS growth inhibition while preserving the structural integrity
of the spheroids. For this purpose, MCTS were treated with DOX concentrations of 2 µM, 1
µM and 0.2 µM for 72 h prior to washing and image-based volume determination. Given the
heightened resistance of MCTS to drug treatments, these concentrations are 2 to 200-fold
greater than the IC50 for DOX in cell monolayers (Figure 5-2).36 OVCAR8 MCTS proved to be
the most resistant to treatment, where only concentrations of 1 and 2 µM significantly
inhibited MCTS growth starting on day 5 after treatment initiation (Figure 5-5, B). At the
concentrations evaluated, all OVCAR8 MCTS remained structurally intact. OV-90 MCTS, on
the other hand, lost structural integrity immediately following treatment with the higher doses,
but remained intact with a 0.2 µM dose, which induced significant growth inhibition (Figure
5-5, D). For HEYA8 MCTS, drug effect was immediately observed at all doses (Figure 5-5, F).
At the highest DOX dose of 2 µM, structural integrity of the MCTS was lost five days after
initiation of treatment. With 1 µM DOX, spheroid integrity was maintained, but the MCTS
decreased to small nodules. With a 0.2 µM dose, the remaining nodules were larger and
structurally sound, which was more desirable for the purposes of the combination studies.
5.4.6. MCTS DOX and OLP Combination Studies
Combination studies were conducted using a fixed dose of DOX that resulted in a
significant reduction in MCTS volume while maintaining their structural integrity. For OVCAR8,
1 µM DOX was used, while 0.2 µM DOX was selected for OV-90 and HEYA8 MCTS.
Corresponding concentrations of OLP that resulted in molar ratios of 1:2 and 1:100 as well as
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 183
2:1 and 100:1 DOX:OLP were employed as treatments for 72 h prior to washing and imaging
of the MCTS over a 10 day period.
When compared to the IC50 values for DOX as a single agent in cell monolayers, the DOX
concentrations required to induce growth inhibition in the MCTS combination studies were
20-fold higher for OVCAR8 MCTS (0.05 µM vs 1 µM) and HEYA8 MCTS (0.01 µM vs 0.2 µM),
but were similar for OV-90 MCTS (0.15 µM vs 0.2 µM). The OLP concentration at the highest
1:100 DOX:OLP ratio was about 7-fold higher than the IC50 value of single-drug OLP in OVCAR8
monolayers (14 µM vs 100 µM) and significantly lower for HEYA8 (12.7 µM vs 2 µM) and OV-
90 (54 µM vs 2 µM). Therefore, OLP likely plays a role as a sensitizer to DOX treatment rather
than a particularly cytotoxic agent, an argument supported by the fact that MCTS growth was
not influenced at these concentrations when using OLP alone (Figure 5-7).
In the MCTS, DOX:OLP combination treatment appeared to be more effective in
comparison to DOX alone for OV-90 (1:100 on days 7 and 9, 100:1 on day 9) and HEYA8 (1:100
and 100:1 on day 7, and all ratios on day 9), but only in one instance for OVCAR8 (100:1, day
7) (Figure 5-6). Analogous to experiments in cell monolayers, higher ratios (1:100, 100:1)
appeared to be more effective than lower ones (1:2, 2:1) in HEYA8 MCTS (Figure 5-6, B). For
OV-90 MCTS, a very heterogeneous pattern was observed, where some MCTS remained
intact while others lost integrity within the same treatment group. This yielded a high level of
variation in the resulting data, which perhaps prevented the detection of statistically
significant differences between ratios. However, a trend towards greater efficacy in response
to treatment with higher ratios (1:100 and 100:1) was still observed when compared to the
lower ratios (1:2 and 2:1) (Figure 5-6, C). Collectively, the cytotoxic effects observed for
DOX:OLP in comparison to DOX alone were comparable to what was observed in monolayer
CI studies, yet appeared less pronounced in the more resistant MCTS model systems.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 184
Importantly, higher ratios (1:100 and 100:1) yielded greater efficacy than lower ratios (1:2 and
2:1) in HEYA8 and OV-90 MCTS, corroborating the trend observed in the monolayer CI studies.
Figure 5-6 (Next page): Growth inhibition studies for OVCAR8 (A), HEYA8 (B) and OV-90 (C)
MCTS in response to treatment with DOX:OLP at select molar ratios in comparison to DOX
alone. The initial DOX concentration was chosen as described above to be 1 µM for OVCAR8
MCTS and 0.2 µM for OV-90 and HEYA8 MCTS. OLP concentrations were calculated for each
molar ratio based on each DOX concentration. Following the 72 h treatment period, 50 % of
media was replaced every other day prior to imaging and determination of MCTS volume. A
(*) indicates significant differences between the indicated treatment group pairs, while a (#)
represents significant differences between DOX and each ratio (p 0.05).
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 185
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 186
5.5. Discussion
The dismal survival rates for OC patients that have acquired resistance to platinum-based
therapy present a demand for innovative treatment options. Recurrent OC is particularly
challenging to treat, since the tumor genome seems to lack characteristic, exploitable drug
targets. A very recently published whole-genome characterization of 92 OC patients with
chemoresistant disease reports higher mutational burden and structural variants in recurrent
disease than matched pre-treatment tumors as a result of adaptation of the tumor genome
in response to previous cycles of chemotherapy.68 In patients with originally BRCA-deficient
disease, reversion mutations that restore BRCA function have been observed upon
recurrence, although mutations do remain in a larger subset of patients with recurrent
disease.47,68 For this highly variable patient population, combination therapy using a DNA
damaging agent and a molecularly targeted agent that can exploit HRR deficiencies appears
to be a reasonable approach given the current state of knowledge.69 The combination of DOX
and the recently approved PARP inhibitor, OLP (Lynparza®), is particularly promising in this
setting, since DOX lacks cross-resistance with platinum, and OLP is the most clinically
advanced PARP inhibitor.10-12 This report aimed to conduct a thorough preclinical evaluation
of the conditions under which DOX and OLP can achieve synergistically beneficial outcomes
in 2D and 3D in vitro cell model systems representative of recurrent and/or residual HGSOC.
The nine ovarian cancer cell lines employed in this study reflect some of the challenges
associated with the high heterogeneity in HGSOC in terms of BRCA status, proliferation rate
and cellular organization when in 3D cell culture. Consequently, a wide range of sensitivities
to DOX and OLP administered individually or in combination were observed. Interestingly,
cytotoxicity was not reliably predictable based on known cell line characteristics, such as
BRCA status. Others have made similar observations.33,68 As a general observation across all
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 187
cell lines, it was found that the combination effect of DOX:OLP in cell monolayers was
predominantly additive for more equimolar drug ratios and synergistic when there were large
differences in molar concentrations, such as 1:100, 1:50 or 50:1 and 100:1 (Figure 5-3). The
data obtained also provides insight into the tremendous variability in drug sensitivity that is
possible in cell lines with specific characteristics. For example, in cell monolayer studies with
the OV-90 cell line, it was found that, despite having wild-type BRCA status and relatively high
resistance to both drugs as single agents, synergy was observed for nearly all ratios of the
DOX-OLP combination. OV-90 was also the only cell line in this study that was derived from
ascites obtained from a patient with previously untreated HGSOC. The UWB1.289 and
UWB1.289+BRCA1 cell line pair is illustrative of the restoration of BRCA1 status that can
occur, for instance, following adaptation to chemotherapy. In this study, the BRCA1-restored
cells were found to be three to four fold more resistant to both DOX and OLP monotherapy as
compared to the parental, BRCA1-deficient cell line (Figure 5-1). However, the DOX-OLP
combination appeared to be more synergistic in the BRCA1-restored cell line
UWB1.289+BRCA1 (Figure 5-3). The PEO1 and PEO4 cell pair, derived from the same patient
before and after chemotherapy, respectively, showed a more expected behavior, where
restoration of BRCA2 in response to treatment yielded lower sensitivity to both single drugs,
and a trend towards antagonism for the drug combination (Figure 5-1, Figure 5-3). The
disparity in cell doubling time between the UWB1.289 and UWB1.289+BRCA1 cells (54 h vs
38 h) and the difference in the morphologies of the PEO1 and PEO4 MCTS (Figure 5-4)
suggest that there may be other differences beyond BRCA status between these cell lines.
This calls into question the utility of these cell “pairs” for evaluation of effects that are
assumed to be solely related to BRCA status. Furthermore, variations in proliferation rates
have been identified as the most relevant associative factor to chemotherapy response in
ovarian cancer cell lines in comparison to the expression of oncogenic protein markers,33 in
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 188
that higher rates of proliferation render cells more susceptible to chemotherapeutics that
target cell cycle mechanisms. This effect was indeed observed in the present study. Despite
being BRCA wild-type, the HEYA8 cell line was found to be most sensitive to single drug
treatment with both DOX and OLP, possibly due to its short T2 of 16 h. As well, combination
treatment resulted in overall synergistic effects in this cell line. Correspondingly, the BRCA1-
deficient COV362 cell line, which had the slowest proliferation rate (T2 = 99 h), appeared to
be relatively resistant to both DOX and the combination treatment.
Evaluation of drug sensitivity in MCTS can provide additional important insight, as the
3D structure enables evaluation of effects of tissue microenvironment, cellular organization
and drug penetration in addition to cytotoxicity. However, these effects depend on the nature
of the MCTS and only manifest in MCTS that present as a coherent spheroid nodule. None of
the BRCA-mutated cell lines formed usable MCTS. It was observed that the effective dose of
DOX in OV-90 MCTS was similar to the IC50 value obtained for DOX in cell monolayers. In
contrast, the effective concentrations of DOX in OVCAR8 and HEYA8 MCTS were 20-fold
greater than the respective IC50 values for the drug in cell monolayers. The sensitivity of the
OV-90 MCTS to DOX may be explained by the heterogeneous structure within these spheroids
as well as their loss of tumor spheroid integrity during treatment (Figure 5-5, C, D). The
differences observed in the CI values obtained for the DOX:OLP combinations in cell
monolayers did translate, to some extent, to comparable differences in the more treatment
resistant MCTS. For the OVCAR8 MCTS, the drug combination of DOX:OLP at all ratios tested
did not appear to be substantially more effective in reducing MCTS volume than treatment
with DOX alone, which is analogous to findings in the monolayer studies where an overall
additive effect was observed for both Fa = 0.5 and Fa = 0.75. For HEYA8, strong synergy was
observed at high molar ratios of DOX:OLP (1:100, 100:1) in both 2D monolayers and the 3D
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 189
MCTS, and a similar trend was observed for OV-90 (Figure 5-6). Thus, results obtained from
both the monolayer CI and MCTS growth inhibition studies successfully demonstrate the
importance of selecting the appropriate ratio for a drug combination, and the significant
impact that the optimal ratio can have in terms of enhancing the growth inhibitory effect of
the two drugs.
Collectively, these observations illustrate that no singular characteristic of ovarian cancer
cells can predict their sensitivity to a therapeutic agent. Cellular response to chemotherapy
and/or molecular therapy is complex and is determined by an aggregate of the genetic,
morphological and physiological traits of a specific tumor. In this study, it was expected that
the combination of DOX and OLP would result in overall synergistic effects, especially in the
BRCA deficient ovarian cancer cell lines. However, in general it was found that BRCA-
deficiency alone did not predict a cell line’s sensitivity to the DOX:OLP combination.
Additionally, some cell lines recognized as BRCA proficient were found to be sensitive to the
combination. The full potential of DOX and OLP may have not yet been achieved in the clinic
due to lack of optimized combination strategies. The identification and administration of
synergistic drug ratios of these drugs may offer a solution for full exploitation of this
promising combination. It is noteworthy that the bolus dosing regimen employed in this study
is unlike that used clinically, which includes twice-daily administration of OLP during
chemotherapy and beyond as maintenance therapy. In future in vivo preclinical evaluations, it
will be interesting and clinically important to determine the impact of OLP maintenance
therapy following DOX-OLP combination treatment at synergistic ratios, and the impact of the
combination dosing schedule, on treatment efficacy and toxicity.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 190
5.6. Conclusions
In summary, this report highlights the challenges that tumor heterogeneity poses to the
effective treatment of recurrent HGSOC. Cell line characteristics including BRCA status,
doubling time and single-drug sensitivity were not predictive of response to the DOX-OLP
combination. Overall, the effectiveness of the DOX:OLP combination therapy in monolayers
and MCTS was found to be ratio-dependent, such that more equimolar ratios (1:2 and 2:1
DOX:OLP) resulted in additive effects, while a greater level of synergy was observed with more
extreme ratios (1:100 and 100:1 DOX:OLP), especially at the higher effect level of 0.75. Future
studies by our group will examine the impact of treatment with the synergistic molar ratios
identified in this study, with our without OLP maintenance therapy, as well as dosing regimens,
on efficacy of DOX and OLP in relevant animal models of HGSOC.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 191
5.7. Acknowledgements
S.E. is funded by the NSERC CREATE Biointerfaces Training Program and holds an Ontario
Trillium Scholarship. C.A. acknowledges GlaxoSmithKline for an endowed chair in
Pharmaceutics and Drug Delivery. The authors acknowledge Dr. Tara Paton from the Genetic
Analysis Facility at The Hospital for Sick Children, for providing technical expertise in the STR
analysis and BRCA sequencing presented herein, and Dr. Tara Spence for critical review of
the manuscript.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 192
5.8. Supplementary Information
Table 5-5: Short Tandem Repeat Analysis for the cell lines used. All cell lines were identified
according to their respective references.
STR Analysis Results Reference Profiles Reference Profile Source
Sample Name Marker Allele 1 Allele 2 Allele 1 Allele 2
OV90
AMEL X X X X
ATCC
CSF1PO 12 13 12 13
D13S317 11 12 11 12
D16S539 11 11 11 11
D5S818 11 15 11 15
D7S820 10 10 10 10.1
TH01 9.3 9.3 9.3 9.3
TPOX 8 10 8 10
vWA 16 17 16 17
OVCAR8
AMEL X X X X
See Reference 70
CSF1PO 11 11 11 11
D13S317 12 12 12 12
D16S539 13 13 13 13
D21S11 28 28 28 28
D5S818 12 12 12 12
D7S820 12 12 12 12
TH01 7 7 7 7
TPOX 8 8 8 8
vWA 16 17 16 17
UWB1.289
AMEL X X X X
ATCC CSF1PO 11 11 11 11
D13S317 9 9 9 9
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 193
D16S539 12 12 12 12
D5S818 13 13 13 13
D7S820 7 10 7 10
TH01 9 9 9 9
TPOX 9 11 9 11
vWA 16 19 16 19
UWB1.289+BRCA1
AMEL X X X X
ATCC
CSF1PO 11 11 11 11
D13S317 9 9 9 9
D16S539 12 12 12 12
D5S818 13 13 13 13
D7S820 7 10 7 10
TH01 9 9 9 9
TPOX 9 11 9 11
vWA 16 19 16 19
COV362
AMEL X X X X
Public Health England Culture
Collections
CSF1PO 11 12 11 12
D13S317 14 14 14 14
D16S539 11 11 11 11
D5S818 13 13 13 13
D7S820 10 14 10 14
TH01 9.3 9.3 9.3 9.3
TPOX 11 11 11 11
vWA 15 18 15 18
PEO1
AMEL X X X X
Public Health England Culture
Collections
CSF1PO 10 12 10 12
D13S317 10 10 10 10
D16S539 9 9 9 9
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 194
D5S818 11 12 11 12
D7S820 10 10 10 10
TH01 9.3 9.3 9.3 9.3
TPOX 9 11 9 11
vWA 15 16 15 16
PEO4
AMEL X X X X
Public Health England Culture
Collections
CSF1PO 10 12 10 12
D13S317 10 10 10 10
D16S539 9 9 9 9
D5S818 11 12 11 12
D7S820 10 10 10 10
TH01 9.3 9.3 9.3 9.3
TPOX 9 11 9 11
vWA 15 16 15 16
HEYA8
AMEL X X X X
See Reference 71
CSF1PO 10 11 10 11
D13S317 11 11 11 11
D16S539 8 12 8 12
D21S11 30 30 30 30
D5S818 11 12 11 12
D7S820 12 12 12 12
TH01 8 9.3 8 9.3
TPOX 11 11 11 11
vWA 16 17 16 17
SKOV3
AMEL X X X X
ATCC CSF1PO 11 11 11 11
D13S317 8 11 8 11
D16S539 12 12 12 12
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 195
D5S818 11 11 11 11
D7S820 13 14 13 14
TH01 9 9.3 9 9.3
TPOX 8 11 8 11
vWA 17 18 17 18
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 196
Figure 5-7: Effect of OLP treatment on growth inhibition of MCTS formed from OVCAR8 (A),
OV-90 (B) and HEYA8 (C) ovarian cancer cells. MCTS growth inhibition was assessed
following a 72h treatment with OLP. Every other day, 50 % of the media was replaced prior to
imaging and determination of MCTS volume
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 197
5.9. References
1. Jayson GC, Kohn EC, Kitchener HC, Ledermann JA. Ovarian cancer. Lancet. Oct 11 2014;384(9951):1376-1388.
2. Berns EM, Bowtell DD. The changing view of high-grade serous ovarian cancer. Cancer research. Jun 1 2012;72(11):2701-2704.
3. Vargas AN. Natural history of ovarian cancer. Ecancermedicalscience. 2014;8:465.
4. Colombo PE, Fabbro M, Theillet C, Bibeau F, Rouanet P, Ray-Coquard I. Sensitivity and resistance to treatment in the primary management of epithelial ovarian cancer. Critical reviews in oncology/hematology. Feb 2014;89(2):207-216.
5. Ahmed AA, Etemadmoghadam D, Temple J, et al. Driver mutations in TP53 are ubiquitous in high grade serous carcinoma of the ovary. J Pathol. May 2010;221(1):49-56.
6. Syrios J, Banerjee S, Kaye SB. Advanced Epithelial Ovarian Cancer: From Standard Chemotherapy to Promising Molecular Pathway Targets - Where Are we Now? Anticancer Res. May 2014;34(5):2069-2077.
7. Gonzalez-Martin A, Sanchez-Lorenzo L, Bratos R, Marquez R, Chiva L. First-line and maintenance therapy for ovarian cancer: current status and future directions. Drugs. Jun 2014;74(8):879-889.
8. Schmid BC, Oehler MK. New perspectives in ovarian cancer treatment. Maturitas. Feb 2014;77(2):128-136.
9. Staropoli N, Ciliberto D, Botta C, et al. Pegylated liposomal doxorubicin in the management of ovarian cancer: a systematic review and metaanalysis of randomized trials. Cancer biology & therapy. Jun 1 2014;15(6):707-720.
10. Safra T, Borgato L, Nicoletto MO, et al. BRCA Mutation Status and Determinant of Outcome in Women with Recurrent Epithelial Ovarian Cancer Treated with Pegylated Liposomal Doxorubicin. Molecular cancer therapeutics. Oct 2011;10(10):2000-2007.
11. Adams SF, Marsh EB, Elmasri W, et al. A high response rate to liposomal doxorubicin is seen among women with BRCA mutations treated for recurrent epithelial ovarian cancer. Gynecologic oncology. Dec 2011;123(3):486-491.
12. Deeks ED. Olaparib: first global approval. Drugs. Feb 2015;75(2):231-240.
13. Davis A, Tinker AV, Friedlander M. "Platinum resistant" ovarian cancer: What is it, who to treat and how to measure benefit? Gynecologic oncology. Jun 2014;133(3):624-631.
14. Cooke SL, Brenton JD. Evolution of platinum resistance in high-grade serous ovarian cancer. The Lancet. Oncology. Nov 2011;12(12):1169-1174.
15. Hennessy BTJ, Timms KM, Carey MS, et al. Somatic Mutations in BRCA1 and BRCA2 Could Expand the Number of Patients That Benefit From Poly (ADP Ribose) Polymerase Inhibitors in Ovarian Cancer. Journal of Clinical Oncology. Aug 1 2010;28(22):3570-3576.
16. Bell D, Berchuck A, Birrer M, et al. Integrated genomic analyses of ovarian carcinoma. Nature. Jun 30 2011;474(7353):609-615.
17. O'Sullivan CC, Moon DH, Kohn EC, Lee JM. Beyond Breast and Ovarian Cancers: PARP Inhibitors for BRCA Mutation-Associated and BRCA-Like Solid Tumors. Frontiers in oncology. 2014;4:42.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 198
18. Do K, Chen AP. Molecular pathways: targeting PARP in cancer treatment. Clinical cancer research : an official journal of the American Association for Cancer Research. Mar 1 2013;19(5):977-984.
19. Liu JF, Konstantinopoulos PA, Matulonis UA. PARP inhibitors in ovarian cancer: Current status and future promise. Gynecologic oncology. May 2014;133(2):362-369.
20. Helleday T. The underlying mechanism for the PARP and BRCA synthetic lethality: Clearing up the misunderstandings. Mol Oncol. Aug 2011;5(4):387-393.
21. Audeh MW, Carmichael J, Penson RT, et al. Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and recurrent ovarian cancer: a proof-of-concept trial. Lancet. Jul 24 2010;376(9737):245-251.
22. Kaye SB, Lubinski J, Matulonis U, et al. Phase II, open-label, randomized, multicenter study comparing the efficacy and safety of olaparib, a poly (ADP-ribose) polymerase inhibitor, and pegylated liposomal doxorubicin in patients with BRCA1 or BRCA2 mutations and recurrent ovarian cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. Feb 1 2012;30(4):372-379.
23. Gelmon KA, Tischkowitz M, Mackay H, et al. Olaparib in patients with recurrent high-grade serous or poorly differentiated ovarian carcinoma or triple-negative breast cancer: a phase 2, multicentre, open-label, non-randomised study. The lancet oncology. Sep 2011;12(9):852-861.
24. Oza AM, Cibula D, Benzaquen AO, et al. Olaparib combined with chemotherapy for recurrent platinum-sensitive ovarian cancer: a randomised phase 2 trial. The lancet oncology. Jan 2015;16(1):87-97.
25. Mason KA, Valdecanas D, Hunter NR, Milas L. INO-1001, a novel inhibitor of poly(ADP-ribose) polymerase, enhances tumor response to doxorubicin. Invest New Drugs. Feb 2008;26(1):1-5.
26. Munoz-Gamez JA, Martin-Oliva D, Aguilar-Quesada R, et al. PARP inhibition sensitizes p53-deficient breast cancer cells to doxorubicin-induced apoptosis. The Biochemical journal. Feb 15 2005;386(Pt 1):119-125.
27. Lee JM, Ledermann JA, Kohn EC. PARP Inhibitors for BRCA1/2 mutation-associated and BRCA-like malignancies. Ann Oncol. Jan 2014;25(1):32-40.
28. Magan N, Isaacs RJ, Stowell KM. Treatment with the PARP-inhibitor PJ34 causes enhanced doxorubicin-mediated cell death in HeLa cells. Anticancer Drugs. Jul 2012;23(6):627-637.
29. Plummer R. Poly(ADP-ribose)polymerase (PARP) inhibitors: from bench to bedside. Clin Oncol (R Coll Radiol). May 2014;26(5):250-256.
30. Carol H, Fan MMY, Harasym TO, et al. Efficacy of CPX-351, (Cytarabine: Daunorubicin) Liposome Injection, Against Acute Lymphoblastic Leukemia (ALL) Xenograft Models of the Pediatric Preclinical Testing Program. Pediatr Blood Cancer. Jan 2015;62(1):65-71.
31. Dicko A, Mayer LD, Tardi PG. Use of nanoscale delivery systems to maintain synergistic drug ratios in vivo. Expert Opin Drug Del. Dec 2010;7(12):1329-1341.
32. Harasym TO, Liboiron BD, Mayer LD. Drug ratio-dependent antagonism: a new category of multidrug resistance and strategies for its circumvention. Methods in molecular biology. 2010;596:291-323.
33. Beaufort CM, Helmijr JC, Piskorz AM, et al. Ovarian cancer cell line panel (OCCP): clinical importance of in vitro morphological subtypes. PloS one. 2014;9(9):e103988.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 199
34. Domcke S, Sinha R, Levine DA, Sander C, Schultz N. Evaluating cell lines as tumour models by comparison of genomic profiles. Nat Commun. Jul 2013;4.
35. Stordal B, Timms K, Farrelly A, et al. BRCA1/2 mutation analysis in 41 ovarian cell lines reveals only one functionally deleterious BRCA1 mutation. Molecular oncology. Jun 2013;7(3):567-579.
36. Mikhail AS, Eetezadi S, Allen C. Multicellular tumor spheroids for evaluation of cytotoxicity and tumor growth inhibitory effects of nanomedicines in vitro: a comparison of docetaxel-loaded block copolymer micelles and Taxotere(R). PloS one. 2013;8(4):e62630.
37. Hirschhaeuser F, Menne H, Dittfeld C, West J, Mueller-Klieser W, Kunz-Schughart LA. Multicellular tumor spheroids: An underestimated tool is catching up again. Journal of biotechnology. Jul 1 2010;148(1):3-15.
38. Zietarska M, Maugard CM, Filali-Mouhim A, et al. Molecular description of a 3D in vitro model for the study of epithelial ovarian cancer (EOC). Molecular carcinogenesis. Oct 2007;46(10):872-885.
39. Shield K, Ackland ML, Ahmed N, Rice GE. Multicellular spheroids in ovarian cancer metastases: Biology and pathology. Gynecol Oncol. Apr 2009;113(1):143-148.
40. Elattar A, Bryant A, Winter-Roach BA, Hatem M, Naik R. Optimal primary surgical treatment for advanced epithelial ovarian cancer. Cochrane Db Syst Rev. 2011(8).
41. Chou TC. Drug Combination Studies and Their Synergy Quantification Using the Chou-Talalay Method. Cancer Res. Jan 15 2010;70(2):440-446.
42. Chou TC. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol Rev. Sep 2006;58(3):621-681.
43. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Advances in enzyme regulation. 1984;22:27-55.
44. Mikhail AS, Eetezadi S, Allen C. Multicellular Tumor Spheroids for Evaluation of Cytotoxicity and Tumor Growth Inhibitory Effects of Nanomedicines In Vitro: A Comparison of Docetaxel-Loaded Block Copolymer Micelles and Taxotere (R). PloS one. Apr 23 2013;8(4).
45. Kenny PA, Lee GY, Myers CA, et al. The morphologies of breast cancer cell lines in three-dimensional assays correlate with their profiles of gene expression. Molecular oncology. Jun 2007;1(1):84-96.
46. DelloRusso C, Welcsh PL, Wang WX, Garcia RL, King MC, Swisher EM. Functional characterization of a novel BRCA1-Null ovarian cancer cell line in response to ionizing radiation. Mol Cancer Res. Jan 2007;5(1):35-45.
47. Wolf CR, Hayward IP, Lawrie SS, et al. Cellular heterogeneity and drug resistance in two ovarian adenocarcinoma cell lines derived from a single patient. International journal of cancer. Journal international du cancer. Jun 15 1987;39(6):695-702.
48. Sakai W, Swisher EM, Jacquemont C, et al. Functional restoration of BRCA2 protein by secondary BRCA2 mutations in BRCA2-mutated ovarian carcinoma. Cancer research. Aug 15 2009;69(16):6381-6386.
49. van den Berg-Bakker CA, Hagemeijer A, Franken-Postma EM, et al. Establishment and characterization of 7 ovarian carcinoma cell lines and one granulosa tumor cell line: growth features and cytogenetics. Int J Cancer. Feb 20 1993;53(4):613-620.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 200
50. Schilder RJ, Hall L, Monks A, et al. Metallothionein gene expression and resistance to cisplatin in human ovarian cancer. Int J Cancer. Mar 15 1990;45(3):416-422.
51. Stordal B, Timms K, Farrelly A, et al. BRCA1/2 mutation analysis in 41 ovarian cell lines reveals only one functionally deleterious BRCA1 mutation. Molecular oncology. Jun 2013;7(3):567-579.
52. Geisler JP, Hatterman-Zogg MA, Rathe JA, Buller RE. Frequency of BRCA1 dysfunction in ovarian cancer. Journal of the National Cancer Institute. Jan 2 2002;94(1):61-67.
53. Baldwin RL, Nemeth E, Tran H, et al. BRCA1 promoter region hypermethylation in ovarian carcinoma: a population-based study. Cancer Res. Oct 1 2000;60(19):5329-5333.
54. Baldwin RL, Nemeth E, Tran H, et al. BRCA1 promoter region hypermethylation in ovarian carcinoma: A population-based study. Cancer research. Oct 1 2000;60(19):5329-5333.
55. Ruscito I, Dimitrova D, Vasconcelos I, et al. BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients--a study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD). European journal of cancer. Aug 2014;50(12):2090-2098.
56. Provencher DM, Lounis H, Champoux L, et al. Characterization of four novel epithelial ovarian cancer cell lines. In vitro cellular & developmental biology. Animal. Jun 2000;36(6):357-361.
57. Moore DH, Allison B, Look KY, Sutton GP, Bigsby RM. Collagenase expression in ovarian cancer cell lines. Gynecologic oncology. Apr 1997;65(1):78-82.
58. Buick RN, Pullano R, Trent JM. Comparative properties of five human ovarian adenocarcinoma cell lines. Cancer Res. Aug 1985;45(8):3668-3676.
59. Selby PJ, Thomas JM, Monaghan P, Sloane J, Peckham MJ. Human tumour xenografts established and serially transplanted in mice immunologically deprived by thymectomy, cytosine arabinoside and whole-body irradiation. Br J Cancer. Jan 1980;41(1):52-61.
60. Fogh J, Fogh JM, Orfeo T. One hundred and twenty-seven cultured human tumor cell lines producing tumors in nude mice. Journal of the National Cancer Institute. Jul 1977;59(1):221-226.
61. Chaluvally-Raghavan P, Zhang F, Pradeep S, et al. Copy number gain of hsa-miR-569 at 3q26.2 leads to loss of TP53INP1 and aggressiveness of epithelial cancers. Cancer cell. Dec 8 2014;26(6):863-879.
62. Astanehe A, Arenillas D, Wasserman WW, et al. Mechanisms underlying p53 regulation of PIK3CA transcription in ovarian surface epithelium and in ovarian cancer. J Cell Sci. Mar 1 2008;121(Pt 5):664-674.
63. DelloRusso C, Welcsh PL, Wang W, Garcia RL, King MC, Swisher EM. Functional characterization of a novel BRCA1-null ovarian cancer cell line in response to ionizing radiation. Mol Cancer Res. Jan 2007;5(1):35-45.
64. Chou TC. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacological Reviews. Sep 2006;58(3):621-681.
65. Ocana A, Amir E, Yeung C, Seruga B, Tannock IF. How valid are claims for synergy in published clinical studies? Ann Oncol. Aug 2012;23(8):2161-2166.
Chapter 5: Combination Index Studies of Doxorubicin and Olaparib 201
66. Tardi P, Johnstone S, Harasyrn N, et al. In vivo maintenance of synergistic cytarabine:daunorubicin ratios greatly enhances therapeutic efficacy. Leukemia Res. Jan 2009;33(1):129-139.
67. Tardi PG, Dos Santos N, Harasym TO, et al. Drug ratio-dependent antitumor activity of irinotecan and cisplatin combinations in vitro and in vivo. Molecular cancer therapeutics. Aug 2009;8(8):2266-2275.
68. Patch AM, Christie EL, Etemadmoghadam D, et al. Whole-genome characterization of chemoresistant ovarian cancer. Nature. May 28 2015;521(7553):489-494.
69. Leung AW, Kalra J, Santos ND, Bally MB, Anglesio MS. Harnessing the potential of lipid-based nanomedicines for type-specific ovarian cancer treatments. Nanomedicine. Mar 2014;9(3):501-522.
70. Lorenzi PL, Reinhold WC, Varma S, et al. DNA fingerprinting of the NCI-60 cell line panel. Mol Cancer Ther. Apr 2009;8(4):713-724.
71. Korch C, Spillman MA, Jackson TA, et al. DNA profiling analysis of endometrial and ovarian cell lines reveals misidentification, redundancy and contamination. Gynecologic oncology. Oct 2012;127(1):241-248.
202
Conclusions and Future Directions
Sina Eetezadi, Larissa C. F. da Gama and Christine Allen
Experiments by L.C.F. da Gama under supervision of S. Eetezadi. Written
by S. Eetezadi. Edited by C. Allen.
Chapter 6: Conclusions and Future Directions 203
6.1. Thesis Conclusion and Summary of Findings
The overarching goal of this thesis was to develop innovative chemotherapeutic treatment
approaches with a focus on ovarian cancer. Specifically, it was hypothesized that the delivery
of doxorubicin (DOX) in block copolymer micelles (BCMs) would enhance both the
intratumoral distribution and tumor bioavailability of DOX resulting in a significant
improvement in tumor growth inhibition, relative to PLD, in clinically relevant 3D cell models
of ovarian cancer (MCTS). An additional goal of this research is to examine the impact of
combined administration of OLP and DOX on the therapeutic outcomes in ovarian cancer +/-
BRCA mutations.
The BCM-DOX formulation developed in Chapter 4, was found to include high drug loading
of DOX encapsulated in stable BCMs of 50 nm in diameter (Figure 4-2). Ultra-filtration was
employed as a means to increase the drug concentration within the formulation. In the case
of BCM-DOX this resulted in drug loading levels of 2 to 7.6 mg/ml with a copolymer
concentration of 50 mg/mL. The DOX concentration of 7.6 mg/mL is the highest DOX drug
loading level reported to date for a micellar formulation (Table 4-1). Moreover, the BCM-DOX
formulation demonstrated increased free drug penetration of DOX in HEYA8 and OV90 MCTS
(Figure 4-8) and as a result MCTS growth inhibition which was comparable to that obtained
for DOX alone and significantly increased in comparison to PLD (Figure 4-10).
In Chapter 3 the applicability of an easy to use 3D cell culture platform based on
multicellular tumor spheroids (MCTS) was demonstrated as a means to evaluate efficacy and
tissue penetration of therapeutic agents in vitro (Error! Reference source not found.). Using
this tool, methods were developed to (1) analyze cell cytotoxicity similar to 2D monolayer
cultures (Figure 3-6), to (2) follow growth inhibitory effects of agents over time similar to in
vivo efficacy studies (Figure 3-7, Figure 4-10, Figure 5-5), to (3) evaluate free DOX penetration
Chapter 6: Conclusions and Future Directions 204
qualitatively and quantitatively (Figure 4-7, Figure 4-8) and to (4) assess histology and tissue
structure of MCTS (Figure 3-8, Figure 4-6, Table 5-4). A software program was written for the
ImageJ software package that performed automated MCTS volume calculations based on
analysis of light microscopy images of MCTS (Figure 3-11). This enabled the analysis of
MCTS volume changes in response to treatment in thousands of MCTS in a semi-automatic
way, which was previously done manually. Without this method many future experiments
involving large numbers of MCTS would not have been possible.
The MCTS experimental platform was employed as a model of avascular tumor nodules
and small tumor lesions that result from metastasis or remain following cytoreductive
surgery in ovarian cancer. Collectively, it was shown that sensitivity to drug is significantly
reduced for MCTS in comparison to 2D monolayers. This was shown for docetaxel (DTX) in
HeLa and HT29 cell lines (Figure 3-6), but also for DOX in OVCAR8 and HEYA8 cell lines (Figure
4-4, Figure 4-10). Importantly, it was also shown that cells gain more resistance against DTX
even when treated as monolayers following disaggregation of MCTS (Figure 3-10). Moreover,
tissue the structure of the spheroid was found to override the inherent sensitivity of cells. As
such, HT29 cells were more sensitive in 2D cell culture in comparison to HeLa cells, yet less
sensitive as MCTS, which was related to their more dense tissue structure and higher cell
packing density (Figure 3-6). Similarly, effective DOX doses were 20-times greater for OVCAR8
and HEYA8 MCTS than their respective monolayer IC50 values (Figure 5-1, Figure 5-5).
However, for OV-90 MCTS effective drug doses were similar to cell monolayer IC50s, which
can be explained by the less structured morphology of the OV90 MCTS tissue and the loss of
tissue integrity during treatment (Figure 5-1, Figure 5-5).
Collectively, the findings also indicate that tissue penetration of drug is indeed an issue
even for “invisible”, microscopic tumor lesions, such as avascular metastatic cell aggregates.
Chapter 6: Conclusions and Future Directions 205
As such, the higher the cell density within the MCTS, in terms of number of cells per volume,
and the overall cell organization within the tissue (Figure 4-6) reduced penetration of DOX
(Figure 4-8). However, specifically for DOX a major challenge was the dependence of
fluorescence intensity to the cellular environment of the drug, which is a major limitation for
the use of fluorescence methods to reliably quantify drug penetration, even in controlled and
homogenous environments such as MCTS.
Tumor heterogeneity was discussed as one of the key challenges for the clinical
translation of block copolymer micelles in chapter 2 (Figure 2-3). The evaluation of the
combination of DOX together with Olaparib (OLP) in a panel of 9 ovarian cancer cell lines
(Figure 5-3) illustrated that consideration of the genomic characteristics of a tumor (e.g. BRCA
mutation status) do not always enable reliable prediction of sensitivity against therapeutic
agents. Rather it is the aggregate effect of all genes, proteins and regulatory pathways
involved that determine response to treatment. In this setting each cell line may be viewed as
representative of a patient. Theoretically BRCA mutations should confer more sensitivity to
treatment with DOX, OLP or the combination of both. Yet, response to treatment was found
to be highly heterogeneous and not predictable by BRCA mutation status. As such,
comparison of IC50s for BRCA deficient cell lines vs BRCA proficient ones, did not find any
significant difference in sensitivity to either DOX or OLP (Figure 5-2). Similarly, also the
combination of DOX and OLP was not found to be more effective in all BRCA deficient cell
lines, but rather a heterogeneous pattern emerged (Figure 5-3).This work also illustrates that
molar drug ratios in combination therapy are another factor that has to be considered,
contrary to the simplified view of just combining two drugs at each of their maximum
tolerated doses. For example, the DOX:OLP combination appeared to be most synergistic
where there was a high difference in molar concentrations such as 1:100, 1:50 or 50:1 and
Chapter 6: Conclusions and Future Directions 206
100:1 in monolayers (Figure 5-3) and MCTS (Figure 5-6). The synergistic effect observed bears
promise for future evaluation in vivo which warrants an appropriate delivery method to ensure
that the determined molar ratios of both drugs accumulate at the tumor as such, despite
differences in the pharmacokinetic profile of each drug, respectively.
Chapter 6: Conclusions and Future Directions 207
6.2. Future Directions
This project has laid a strong foundation for further development of innovative treatment
approaches in ovarian cancer. From the studies presented within this thesis there are at least
two avenues that warrant further exploration: 1) expansion of the MCTS model and 2) further
evaluation of combination therapies of cytotoxic agents together with PARP inhibitors.
6.2.1. Expanding on the MCTS model for Ovarian Cancer
Still in early stages, advanced MCTS techniques have the potential to become a new
standard testing platform in ovarian cancer where MCTS generation, treatment and analysis
can be fully automated. For example, Friedrich et. al. proposed a MCTS-based drug screen to
be fully integrated into large-scale drug testing routines.1 Such a platform would also help us
to reduce our usage of animal models and may lead to more predictive clinical trials.2,3
In this work MCTS were formed exclusively from tumor cells, using the liquid overlay
technique on 96-well plates where each MCTS was formed in a single well (Chapter 1). A semi-
automatic analysis method developed for the ImageJ software package enabled analysis of
MCTS volume changes in response to treatment. This proved to be an easy-to-use approach
and had much versatility, however also limitations emerged that exposed the weaknesses of
this oversimplified system. As such, the culture system lacked vascularization, which means
that accumulation of drugs or other agents in MCTS is solely based on diffusion, whereas in
vivo a combination of convection and diffusion both contribute to overall drug accumulation.
Additionally, a clearance system analogous to lymphatic vessels is also lacking, which would
simulate the addition of nutrients and oxygen as well as the removal of cell waste and
especially dead cell debris following drug treatment. Microfluidic technology may enable
generation of MCTS from a single cell suspension following settlement into predefined
Chapter 6: Conclusions and Future Directions 208
wells.4,5 Additionally, these devices enable simulation of blood flow and clearance and can be
used together with a drug gradient generator, to simulate pharmacokinetic changes of drug
levels at the tumor site.6,7 However, the current systems appear to be at an early stage and
more work is needed to build a robust platform that can create high-quality MCTS with
reproducible, homogenous growth patterns.
Furthermore, it is known that solid tumors act as organs and not only the aberrant cancer
cells, but also the surrounding stroma play an imperative role in tumor progression and
malignancy. Specifically, because of their inherent genomic instability, and thus mutability,
tumors may be less predictable than healthy tissue organs.8,9 Development of more advanced
in vitro models could likely add to improving the understanding of the disease and foster
development of novel therapeutics. Specifically, as discussed in chapter 2 and demonstrated
in chapter 5 of this thesis, a tumor’s response to treatment is highly heterogeneous and
unpredictable. High-throughput testing platforms that can be used to evaluate the many
manifestations of the targeted cancer type within a controlled environment may be able to
unfold new insights. MCTS formed by co-culture of cancer cells with other cells of the tumor
microenvironment may provide a more accurate model of the disease. As of yet, this has not
been widely investigated for ovarian cancer. One example is provided by Bilandzic and
Stenvers in the Journal of Visual Experiments, where they show a method to co-culture
ovarian cancer cells with LP9 human mesothelial cells and studied the invasive capacity of
ovarian cancer cell lines.10 In another example, Lawrenson et al used fallopian tube secretory
epithelial cells to generate an MCTS model that better reflects HGSOC, which is known to
originate from the fallopian tube.11 In this study, it was shown in gene expression profiles that
the MCTS in vitro better reflect the genetic profile in vivo than traditional 2D cell. Both methods
Chapter 6: Conclusions and Future Directions 209
could be adapted to develop an ovarian cancer co-culture MCTS system for characterization
of chemotherapeutic treatment modalities.
As described in chapter 5 all of the BRCA mutated HGSOC cell lines did not form MCTS
and were therefore not usable for further evaluation, which limited the scope of the study. One
potential reason could be the lack of a stabilizing extracellular matrix. Detailed protocols for
3D cell culture of breast cancer cells embedded in matrigel are available.12 Indeed, Loessner
et al reported growth of the SKOV3 cell line, exclusively in PEG hydrogels. Potentially a similar
approach could be used to yield MCTS for UWB1.289, UWB1.289+BRCA, PEO1, PEO4 and
COV362, which in addition to SKOV3 also did not form MCTS with the methods employed
herein. Success of this method would be particularly promising for HGSOC, since a MCTS
model for BRCA mutated ovarian cancer has still not been reported to date. This could also
change if increasing efforts are made to establish novel cell lines of BRCA mutated ovarian
cancers from primary patient samples. Only one report has been found that established
ovarian MCTS from ascites-derived primary patient cells and demonstrated prolonged
viability of MCTS in vitro in comparison to 2D monolayers.13
6.2.2. Delivering localized drug combinations
The results obtained in chapter 5 indicate that molar drug ratios of DOX:OLP can indeed
influence the effectiveness of the drug combination. It was found that drug combinations with
high molar ratios, such as 100:1 or 1:100 are more effective than low molar ratios such as 1:2
or 2:1 in monolayers (Figure 5-3) and in MCTS (Figure 5-6). Translation of these observations
in vivo and ultimately to the patient will depend on achieving the determined ratios at the
tumor site following administration. This cannot be achieved by simple administration of both
drugs in free form, as DOX and OLP have different pharmacokinetic profiles which in turn
Chapter 6: Conclusions and Future Directions 210
would result in varying drug ratios over time. As such, the reported clinical half-life for free
DOX which is administered intravenously is around 9 h and for PLD 45 h,14 while OLP which is
given orally is reported to have a half-life of approximately 5 h.
Three strategies can be envisioned to overcome this challenge: 1) optimizing treatment
schedules; 2) delivering the drug combination as closely as possible to the tumor tissue; or 3)
packaging the combination together in a delivery vehicle that carries the cargo to the tumor
site. The first strategy, while theoretically possible is likely to be unfeasible in a clinical setting,
especially because different schedules for every patient would be needed. The second
strategy would be to deliver both drugs together directly into the peritoneal cavity, with the
assumption that the differences in clearance would have less of an effect due to the proximity
to the tumor tissue. Finally, nanotechnology-based drug delivery, with for example a micellar
or liposomal formulation, has the unique capability of delivering several drugs together and
releasing them selectively at the target tumor site. For example, Hasenstein et. al. reported
the combination of the three drugs paclitaxel, rapamycin and tanespimycin in a single micellar
formulation for treatment of ovarian cancer.15 Recently, CPX-351 a liposomal formulation of
a fixed-ratio combination of cytarabine and daunorubicin reported improved outcomes in a
phase II clinical trial in acute myeloid leukemia and is progressing towards phase III.16,17
A possible development strategy would be to first confirm the beneficial effect of molar
ratios in vivo using animal xenograft models of ovarian cancer. As a first step, based on the
second strategy outlined above DOX:OLP could be delivered with an Alzet® osmotic micro-
infusion pump placed into the intraperitoneal cavity of mice for local delivery.18 Alzet® pumps
are of the size of small capsules and are implantable in laboratory animals such as mice, but
are not for human use. The high osmolality of the outer surface of the pump causes water to
flux into the pump and compresses a flexible reservoir which contains the drug solution. Upon
Chapter 6: Conclusions and Future Directions 211
compression the solution inside the reservoir is slowly pushed out of the pump into the
surrounding environment at a controlled, predetermined rate for durations ranging from 1 day
to 6 weeks. The technology has been used by our group previously for continuous
administration of carboplatin to SKOV-3 mouse xenograft models.19
Preliminary work has been conducted as part of this project, which demonstrates the
general feasibility of this approach. The challenge was to solubilize both DOX and OLP at
sufficiently high concentrations while ensuring drug stability over the 14-day period. While the
pump itself is relatively easy to use, it requires that DOX and OLP are formulated as a liquid
solution at sufficient drug concentrations. This is particularly challenging for OLP, which has
a water solubility of 0.002 mg/ml according to the manufacturer’s documentation. DOX has
a water solubility of 10 mg/ml and has an MTD of around 5-10 mg/kg for every 3 weeks
depending on the mouse model used.20,21 For a 20 g mouse, this means that at least 100 µg
Figure 6-1: Formulation stability of the DOX:OLP solution incubated at 37 °C over a period of
14 days. Concentration of both drugs reduces slightly, however both remain over 85 % of
initial levels after a two week period.
Chapter 6: Conclusions and Future Directions 212
of DOX has to be delivered which amounts to a concentration of 1 mg/ml in the 100 µl
reservoir of a standard Alzet® pump. DOX is water soluble at this concentration, however
given the molecular weights of DOX (543 g/mol) and OLP (435 g/mol), this also means that
at the extreme 1:100 (DOX: OLP) molar ratio it would be required to solubilize OLP at roughly
80 mg/ml.
A formulation was developed that contained 1.25 mg/mL DOX and 10 mg/mL OLP
molecularly dissolved in a clear solution. This was achieved using a three times molar excess
of 2-hydroxypropyl-β-cyclodextrin to OLP and 300 μg/ml methylparabene in a 0.9 % sodium
chloride solution containing 10 % DMSO at pH 5.5. During the preparation process heat,
sonication and vortexing were used in order to enhance drug interaction and solubility. The
needed quantities of all excipients were extensively tested and optimal concentrations were
determined experimentally. The quantities correspond to a molar ratio of 1:10 of DOX:OLP,
which includes the maximum concentration of OLP that could be solubilized using this
Figure 6-2: Representative peaks for OLP, DOX and DAN (internal standard for DOX) after
extraction from in vitro release media containing BSA.
0
20
40
60
80
100
120
140
0 5 10 15 20 25
Ab
sorb
ance
mA
U
Time (min)
UV 279 nm
DOX
DAN
OLP
FLU Ex 480 nmEm 560 nm
Chapter 6: Conclusions and Future Directions 213
formulation strategy. The solution was shown to be generally stable for 14 days at 37 °C, only
the concentration of DOX seemed to be diminish slightly over time (Figure 6-1).
Drug release was assessed in vitro by suspending the ALZET pump filled with the DOX/OLP
formulation in PBS containing 50 mg/mL BSA over a 14 day period. At specific timepoints an
aliquot of release media was removed and liquid-liquid extraction was employed to isolate
DOX and OLP from BSA. Specifically, aliquots of the release media were mixed with
chloroform:1-propanol (2:1, v/v) and daunorubicin (DAN) as an internal standard. The organic
phase was then removed, the solvent was dried down and the sample was dissolved in HPLC
mobile phase prior to analysis. The mobile phase was composed of 71% pH 2 phosphate
buffer, 3% 1-propanol and 27% Acetonitrile (ACN) with a flow rate of 1 ml/min. The
compounds were detected using a UV detector at 279 nm for OLP and a fluorescence
detector at 480 nm excitation and 560 nm emission for DOX and DAN. Peaks were found to
be well separated using this method (Figure 6-2). Limit of detection was found to be 250 ng /
mL for OLP and 25 ng / mL for DOX and DAN.
HPLC analysis confirmed sustained release of drugs over the 14-day period with an
average molar ratio of OLP to DOX at 12.5 (Figure 6-3).
Chapter 6: Conclusions and Future Directions 214
In summary, it was shown that it is possible to deliver DOX:OLP continuously over 14 days
at a maximum molar ratio of 1:10 with a clinically relevant concentration of DOX. However,
since combination index studies have shown that higher ratios are more effective, future
development should focus on delivering DOX:OLP at molar ratios of 50:1 or 100:1, which
would be less challenging in terms of drug solubility.
Figure 6-3: In vitro release study of OLP and DOX using an ALZET pump immersed in
physiological BSA solution at 37 °C. Both drugs release at the same percentage and therefore
keep their relative ratio over the course of the study.
Chapter 6: Conclusions and Future Directions 215
For further evaluation in vivo, animal xenograft models of the ovarian cancer cell line panel
used in chapter 5 will also be required (Table 5-2). Of the 9 cell lines, so far only the BRCA wt
cell lines SKOV-3 and HEYA8 have been previously established in our laboratory.22,23 For OV-
90 and PEO1/4 animal models have been reported in the literature.24,25 The tumorigenicity for
the other cell lines is unknown and evaluation is warranted.
Following a successful proof-of-principle study with an Alzet® pump, an appropriate
nanotechnology-based drug delivery system for clinical use could be developed, that ensures
delivery of the determined optimal drug ratio at the tumor site. Possible platforms could
include a block copolymer micelle formulation based on BCM-DOX, a liposomal formulation
based on PLD or a solidifying injectable gel, for sustained local delivery.15,26,27
Chapter 6: Conclusions and Future Directions 216
6.3. References
1. Friedrich J, Seidel C, Ebner R, Kunz-Schughart LA. Spheroid-based drug screen: considerations and practical approach. Nature protocols. 2009;4(3):309-324.
2. White EA, Kenny HA, Lengyel E. Three-dimensional modeling of ovarian cancer. Advanced drug delivery reviews. Dec 15 2014;79-80:184-192.
3. Eetezadi S, Ekdawi SN, Allen C. The challenges facing block copolymer micelles for cancer therapy: In vivo barriers and clinical translation. Advanced drug delivery reviews. Oct 13 2014.
4. Kwapiszewska K, Michalczuk A, Rybka M, Kwapiszewski R, Brzozka Z. A microfluidic-based platform for tumour spheroid culture, monitoring and drug screening. Lab on a chip. 2014;14(12):2096-2104.
5. Sun DP, Lu J, Chen ZG, Yu YY, Li YB. A novel three-dimensional microfluidic platform for on chip multicellular tumor spheroid formation and culture. Microfluid Nanofluid. Nov 2014;17(5):831-842.
6. Das T, Meunier L, Barbe L, et al. Empirical chemosensitivity testing in a spheroid model of ovarian cancer using a microfluidics-based multiplex platform. Biomicrofluidics. Jan 2013;7(1).
7. Yang JP, Chen ZG, Ching PY, Shi QJ, Li XC. An integrated microfluidic platform for evaluating in vivo antimicrobial activity of natural compounds using a whole-animal infection model. Lab on a chip. 2013;13(17):3373-3382.
8. Michor F, Polyak K. The origins and implications of intratumor heterogeneity. Cancer Prev Res (Phila). Nov 2010;3(11):1361-1364.
9. Egeblad M, Nakasone ES, Werb Z. Tumors as Organs: Complex Tissues that Interface with the Entire Organism. Dev Cell. Jun 15 2010;18(6):884-901.
10. Bilandzic M, Stenvers KL. Assessment of Ovarian Cancer Spheroid Attachment and Invasion of Mesothelial Cells in Real Time. Jove-J Vis Exp. May 2014(87).
11. Lawrenson K, Notaridou M, Lee N, et al. In vitro three-dimensional modeling of fallopian tube secretory epithelial cells. Bmc Cell Biol. Sep 27 2013;14.
12. Lee GY, Kenny PA, Lee EH, Bissell MJ. Three-dimensional culture models of normal and malignant breast epithelial cells. Nature methods. Apr 2007;4(4):359-365.
13. Lam JT, Bauerschmitz GJ, Kanerva A, et al. Replication of an integrin targeted conditionally replicating adenovirus on primary ovarian cancer spheroids. Cancer Gene Ther. May 2003;10(5):377-387.
14. Gabizon A, Catane R, Uziely B, et al. Prolonged circulation time and enhanced accumulation in malignant exudates of doxorubicin encapsulated in polyethylene-glycol coated liposomes. Cancer Res. Feb 15 1994;54(4):987-992.
15. Hasenstein JR, Shin HC, Kasmerchak K, Buehler D, Kwon GS, Kozak KR. Antitumor activity of Triolimus: a novel multidrug-loaded micelle containing Paclitaxel, Rapamycin, and 17-AAG. Molecular cancer therapeutics. Oct 2012;11(10):2233-2242.
16. Cortes JE, Goldberg SL, Feldman EJ, et al. Phase II, Multicenter, Randomized Trial of CPX-351 (cytarabine: daunorubicin) Liposome Injection Versus Intensive Salvage Therapy in Adults With First Relapse AML. Cancer-Am Cancer Soc. Jan 15 2015;121(2):234-242.
Chapter 6: Conclusions and Future Directions 217
17. Lancet JE, Cortes JE, Hogge DE, et al. Phase 2 trial of CPX-351, a fixed 5:1 molar ratio of cytarabine/daunorubicin, vs cytarabine/daunorubicin in older adults with untreated AML. Blood. May 22 2014;123(21):3239-3246.
18. ALZA Corporation. ALZET - Products Overview. www.alzet.com. Accessed 1st July 2015.
19. Zhidkov N, De Souza R, Ghassemi AH, Allen C, Piquette-Miller M. Continuous intraperitoneal carboplatin delivery for the treatment of late-stage ovarian cancer. Mol Pharm. Sep 3 2013;10(9):3315-3322.
20. Pastorino F, Di Paolo D, Piccardi F, et al. Enhanced antitumor efficacy of clinical-grade vasculature-targeted liposomal doxorubicin. Clinical cancer research : an official journal of the American Association for Cancer Research. Nov 15 2008;14(22):7320-7329.
21. Boven E, Schluper HM, Erkelens CA, Pinedo HM. Doxorubicin compared with related compounds in a nude mouse model for human ovarian cancer. European journal of cancer. 1990;26(9):983-986.
22. Vassileva V, Moriyama EH, De Souza R, et al. Efficacy assessment of sustained intraperitoneal paclitaxel therapy in a murine model of ovarian cancer using bioluminescent imaging. Brit J Cancer. Dec 9 2008;99(12):2037-2043.
23. Zahedi P, De Souza R, Huynh L, Piquette-Miller M, Allen C. Combination drug delivery strategy for the treatment of multidrug resistant ovarian cancer. Mol Pharm. Feb 7 2011;8(1):260-269.
24. Zietarska M, Maugard CM, Filali-Mouhim A, et al. Molecular description of a 3D in vitro model for the study of epithelial ovarian cancer (EOC). Molecular carcinogenesis. Oct 2007;46(10):872-885.
25. Langdon SP, Lawrie SS, Hay FG, et al. Characterization and properties of nine human ovarian adenocarcinoma cell lines. Cancer Res. Nov 1 1988;48(21):6166-6172.
26. Zahedi P, Stewart J, De Souza R, Piquette-Miller M, Allen C. An injectable depot system for sustained intraperitoneal chemotherapy of ovarian cancer results in favorable drug distribution at the whole body, peritoneal and intratumoral levels. Journal of Controlled Release. Mar 28 2012;158(3):379-385.
27. Barenholz Y. Doxil(R)--the first FDA-approved nano-drug: lessons learned. Journal of controlled release : official journal of the Controlled Release Society. Jun 10 2012;160(2):117-134.