obstacles to dose optimization in early stage...

Obstacles to Dose Optimization in Early Stage Cancer Drug Development

René Bruno and Laurent Claret Pharsight Consulting Services

International Workshop on Dose Optimization Strategies

for Targeted Drugs 23 - 24 March 2015, Amsterdam

© Copyright 2015 Certara, L.P. All rights reserved.

Outline

• Oncology drug development and dose selection • A drug-disease modeling framework

– Longitudinal tumor size models – Survival models

• New proposed endpoints based on continuous tumor growth inhibition metrics for early oncology studies

• Conclusions

2


• Expedited programs, huge competition • Empirical selection of dose and dosing schedules in Phase I

– Maximum tolerated dose (MTD) • Not appropriate for dose selection with targeted therapies?

– Pharmacologically active dose based on biomarker responses specific to the mechanism of action

• Fine to establish proof of mechanism … – i.e. is the target impacted, is there some tumor growth inhibition…

• … but not mature for dose selection

• Phase II program not informative – Design

• Limited to establish proof of concept • Very few randomized Phase IIb dose-ranging studies

– Primary clinical endpoints poorly informative (ORR, PFS) and somewhat subjective (PFS)

• Phase III: High failure rate – > 50%

3

Oncology drug development and dose selection


Modeling Framework Tumor Growth Inhibition Metrics


A modeling framework to support the Phase II-III transition

Bruno et al. Clin Pharmacol Ther, 93:303-5, 2013

5

Tumor growth inhibition (TGI) metrics to assess exposure-response in early clinical studies

and predict expected OS


Drug-specific tumor size models

• Semi-mechanistic exposure-driven tumor growth inhibition (TGI) models – Tumor growth, exposure driven drug effect, resistance

appearance1-6

• Empirical models

– Simplified TGI model (assumes constant exposure)7-8

– Linear growth plus exponential shrinkage9-10

– Exponential growth and shrinkage11

6

1Claret et al. PAGE 2006; 2Claret et al. J. Clin. Oncol. 27:4103-8, 2009 3Tham et al. Clin. Cancer Res. 14:4213-8, 2008 4Stein et al. BMC Cancer 12:311, 2012 5Ribba et al. Clin. Cancer Res. Published online (Jul-03, 2012) 6Hansson et al. CPT:PSP, 2, e84, 2013

7Claret et al. PAGE 2012; 8Claret et al. J. Clin. Oncol. 31:2110-14, 2013 9Wang et al. Clin. Pharmacol. Ther. 86:167-74, 2009 10Maitland et al. Clin Pharmacol Ther, 93:345-51, 2013 11Stein et al. Clin. Cancer Res, 17:907-17, 2011


Models for clinical endpoints (overall survival)

• Survival time distribution is estimated (parametric model) as a function of prognostic factors and treatment effect

• Drug independent, disease specific model – TGI metric is used as a biomarker to capture treatment effect – Historical Phase III studies can be used to develop the models – Overall survival models have been developed for MBC1, CRC2,3,

pancreatic cancer, ovarian cancer4, H&N carcinoma, multiple myeloma5, non-hodgkin lymphoma, gastric cancer6, renal cell carcinoma7 (Pharsight collaborators)

– and NSCLC8-10 (FDA, Pharsight) • A few cases of external evaluations are available2,5,11

– More are needed

7

1Claret et al. Proc ASCO 2006 (abstract 2530) 2Claret et al. J. Clin. Oncol. 27:4103-8, 2009 3Claret et al. J. Clin. Oncol. 31:2110-14, 2013 4Lindborn et al. ACoP, 2009 5Bruno et al. Blood 118(21):1881 (Abstract), 2011 6Quartino et al. PAGE 2013 7Mercier et al. ESMO 2014

8Wang et al Clin.Pharmacol. Ther. 86:167-74, 2009 9Claret et al. Clin. Pharmacol. Ther. 95, 446-451, 2014 10Bruno et al. Proc ASCO 2013, abstract e19103 11Claret et al. Clin. Pharmacol. Ther. 92:631-4, 2012

Recently reviewed in Bruno et al. Clin. Pharmacol. Ther. 95, 386-393, 2014.


Tumor growth inhibition metrics

8

Claret et al. J. Clin. Oncol., 31:2110-2114, 2013

Week 8 ECTS

Recently reviewed in Bruno et al. Clin. Pharmacol. Ther. 95, 386-393, 2014


Optimal endpoint for randomized phase II trials Sharma et al, JCO 2014: Resampling the N9741 Trial to Compare Tumor Dynamic Versus Conventional End Points in Randomized Phase II Trials

9


Optimal endpoint for randomized phase II trials

• N9741, a randomized phase III trial of chemotherapy for metastatic colorectal cancer

• Compared the power of various endpoints to detect the superior therapy – In this 3 arm study FOLFOX demonstrated longer OS than IROX or

IFL

• Simulated two-arm, randomized phase II trials of 20 to 80 patients per arm

• Resampled (1002 patients, 5000 replications): – Week 6, 12, and 18 ECTS (observed) – TTG (model-based) – PFS

10


Resampling simulation results

11

Sharma et al, JCO, 33, 36-41, 2015

Power of randomized phase II trials of : (A) FOLFOX vs. IFL

(B) FOLFOX vs. IROX with various endpoints

Log ratio = ECTS


Conclusions

• Supports the consideration of TTG estimated from nonlinear mixed-effects modeling of tumor measurements as a powerful endpoint for randomized phase II trials – Should be measured as a secondary endpoint – More studies should compare TGI metrics with PFS (or even ORR)

• Which phase II endpoint will most often lead to the correct go/no-go decision?

• TGI metrics – Might show a distinct advantage when treatment benefit is smaller

(e.g. FOLFOX vs. IROX) – Might benefit from optimized study designs (e.g. in term of tumor

size observation schedule)

12


Clinical Trial Simulations to Support to Phase II Decisions: Metastatic Renal Cell Carcinoma Framework Mercier et al, ESMO 2014; Claret et al, ASCPT 2015 A model relating overall survival to tumor growth inhibition in renal cell carcinoma patients treated with sunitinib, axitinib or temsirolimus 13


Metastatic renal cell carcinoma (mRCC) OS model

• Historical Phase II-III studies with a variety of treatments – Over 2500 patients

14

Mercier et al. ESMO 2014, Ann Oncol 25 (Supplement 4): iv146–iv164, 2014



• OS model incorporates 7 baseline prognostic factors – Drug effect captured by week 8 ETS

15


SE: standard error, p: wald test (χ2) + sign favorable; - sign not favorable

Drug effect Prognostic factors



• Model qualification – Predictive check of the sunitinib to INF-α HR in the first-line sunitinib study (1034)

16




• Model simulation – Predictive distribution of HR comparing an investigational treatment to sunitinib in a 600 patient study

(300 per arm) as a function of difference in tumor growth inhibition (delta in week 8 ECTS)

17


• According to the simulations – An investigational treatment that

would induce a 20% week 8 ETS difference from reference may result in an improved OS with a HR of ∼ 0.75

– A 300 patients per arm Phase III study would have a 80% probability of success to show a HR < 0.80 (target product profile)

N

Reference HROS=1

Median, 80%, 90%, 95% PI

Target HROS=0.8


Discussion - Conclusions


Value of new endpoints and model-based simulations • New endpoints based on continuous longitudinal tumor size data

may offer powerful alternatives – To establish proof of concept based on early clinical data (cohort

extensions in Phase I or Phase II studies) – To assess dose-response by enabling randomized dose-ranging Phase

II studies • Combined with model-based simulations, expected survival (PFS)

probability distribution for an investigational treatment (and possibly HR vs. SOC) can be predicted, bridging Phase II TGI data to Phase III outcome1,2

• Phase III clinical trials can be simulated to assess probability of success in support of – Go-no go decision – Trial design – Interim analyses

• Predictions of expected outcome can also be made for unstudied doses and schedules

19

1Claret et al. J. Clin. Oncol. 27:4103-8, 2009 2Claret et al. Clin. Pharmacol. Ther. 92:631-4, 2012


Acknowledgements • Pharsight

– P. Chanu, C. Falcoz, F. Mercier, P. Girard (now with Merck-Serono, Geneva, Switzerland) • F. Hoffmann-La Roche, Basel, Switzerland and Shanghai, China

– K. Zuideveld, K. Jorga, J. Fagerberg, M. Abt – F. Schaedeli Stark, F. Sirzen, R. Gieschke, N. Frey, N. Jonsson – Bob Powell

• Genentech, South San-Francisco, CA – JY Jin and colleagues

• Genmab, Copenhagen, Denmark – N. Losic

• Amgen, Thousands Oak, California – J.F. Lu, C.P. Hsu, T. Sun

• Celgene, Summit, NJ – C. Jacques

• Progenics, Tarrytown, NY – Y. Rotshteyn, V. Wong

• Lilly, ErlWood, UK – D. Cronier

• FDA pharmacometrics team – Y. Wang and J. Gobburu

20


References • Claret L, Girard P, O'Shaughnessy J, et al. Model-based predictions of expected anti-tumor response and survival

in phase III studies based on phase II data of an investigational agent. J Clin Oncol 2006;24:307s (suppl, abstract 2530).

• Claret L, Andre V, de Alwis D, Bruno R. Modeling and simulation to assess the use of change in tumor size as primary endpoint in Phase II studies in oncology. PAGE 17 2008. [http://www.page meeting.org/?abstract=1386].

• Claret L et al. A modelling framework to simulate Xeloda dose intensity and survival in colorectal cancer. PAGE 17 2008. [http://www.page-meeting.org/?abstract=1312].

• Bruno R and Claret L. FDA Clin. Pharmacol. Advisory Committee meeting, Rockville, March 18, 2008 http://www.fda.gov/ohrms/dockets/ac/cder08.html#PharmScience

• Lindbom L, Claret L, Andre V, Cleverly A, de Alwis D, Bruno R. A drug independent tumor size reduction-survival model in advanced ovarian cancer to support early clinical development decisions. American Conference on Pharmacometrics, ACoP 2009. http://2009.go-acop.org/sites/all/assets/webform/ACoP%202009%20Poster%20-%20Lars%20Lindbom.pdf

• Bruno et al. Simulation of survival with first- and second-line non-small cell lung cancer (NSCLC) therapy using a public domain drug-disease modeling framework. J Clin Oncol 2009;27(15s)(abstr 8087)

• Claret L, Girard P, Hoff PM et al. Model-based prediction of Phase III overall survival in colorectal cancer based on Phase II tumor dynamics. J Clin Onco. 2009;27:4103-4108.

• Bruno R, Claret L. On the use of change in tumor size to predict survival in clinical oncology studies: Toward a new paradigm to design and evaluate Phase II studies. Clin Pharmacol Ther 2009;86:136-138.

• Jonsson F, Claret L, Knight R, et al. A longitudinal tumor growth inhibition model based on serum M-protein levels in patients with multiple myeloma treated by dexamethasone. PAGE 19. 2010. [www.page meeting.org/?abstract=1705].

• Claret L, Jonsson F, Knight R, et al. A drug independent tumor burden reduction-survival model in patients with multiple myeloma to support early clinical development decisions. 6th International Symposium on Measurement and Kinetics of In Vivo Drug Effects. 2010, Noordwijkerhout, the Netherlands.

• Claret L, Lu J-F, Sun Y-N, Bruno R. Development of a modeling framework to simulate efficacy endpoints for motesanib in thyroid cancer patients. Cancer Chemother Pharmacol 2010;66:1141-1149.

21

http://www.fda.gov/ohrms/dockets/ac/cder08.html

http://2009.go-acop.org/sites/all/assets/webform/ACoP%202009%20Poster%20-%20Lars%20Lindbom.pdf

http://2009.go-acop.org/sites/all/assets/webform/ACoP%202009%20Poster%20-%20Lars%20Lindbom.pdf


References (cont.)

• Claret L, Lu J, Bruno R, Sikorski RS et al. Simulation of phase III studies of motesanib 125 mg once daily plus carboplatin/paclitaxel or bevacizumab plus carboplatin/paclitaxel versus carboplatin/paclitaxel in first-line non-small cell lung cancer using a public domain drug-disease modeling framework and phase II data. Journal of Clinical Oncology 28, 2010 (suppl; abstr e18089)

• Bruno R, Jonsson F, Zaki M et al. Simulation of clinical endpoints (survival, PFS) in patients with refractory multiple myeloma treated with pomalidomide based on interim week 8 M-protein response. The 2011 European Multidisciplinary Cancer Congress, Stockholm (ECCO 16/ESMO 36), 23-27 September, 2011. Eur J Cancer 2011:47 (suppl. 1), S647 (abstract 9227).

• Frances N, Claret L, Bruno R, Iliadis A. Tumor growth modeling from clinical trials reveals synergistic anticancer effect of the capecitabine and docetaxel combination in metastatic breast cancer. Cancer Chemother Pharmacol 2011;68:1413-1419.

• Bruno R, Jonsson F, Zaki M et al. Simulation of clinical outcome for pomalidomide plus low-dose dexamethasone in patients with refractory multiple myeloma based on week 8 M-protein response. Blood 2011;118 (21):1881 (Abstract).

• Claret L, Lu J-F, Bruno R, Hsu CP et al. Simulations using a public domain drug-disease modeling framework and Phase II data predict Phase III survival outcome in first-line non-small-cell lung cancer (NSCLC). Clin Pharmacol Ther 2012:92: 631-634.

• Bruno R, Lindbom L, Schaedeli Stark F, Chanu P, Gilberg F, Frey N, Claret L. Simulations to assess Phase II non-inferiority trials of different doses of capecitabine in combination with docetaxel for metastatic breast cancer. Clin Pharmacol Ther: Pharmacometry and System Pharmacology, 1, e19, doi:10.1038/psp.2012.20, published online 26 December 2012.

• Claret L, Gupta M, Han K, Joshi A, Sarapa N, He J, Powell B, Bruno R. Evaluation of tumor-size response metrics to predict overall survival in Western and Chinese patients with first-line metastatic colorectal cancer. J Clin Oncol, 31, 2110-2114, 2013.

• Bruno R, Mercier F, Claret L. Model-based drug development in oncology: What’s next. Clin Pharmacol Ther, 93, 303-305, 2013 (Invited Commentary).

• Marchand M, Claret L, Losic N, Puchalski TA, Bruno R. Population pharmacokinetics and exposure-response analyses to support dose selection of daratumumab in multiple myeloma patients. PAGE 22 (2013) Abstr 2668 [www.page-meeting.org/?abstract=2668].

• Quartino AL, Claret L, Li J, Lum B, Visich J, Bruno R, Jin J. Evaluation of Tumor Size Metrics to Predict Survival in Advanced Gastric Cancer. PAGE 22 (2013) Abstr 2812 [www.page-meeting.org/?abstract=2812].

22


References (cont.)

• Chanu P, Claret L, Bruno R, Radtke DB, Carpenter SP, Wooldridge JE, Cronier DM. PK/PD relationship of the monoclonal anti-BAFF antibody tabalumab in combination with bortezomib in patients with previously treated multiple myeloma: comparison of serum M-protein and serum free light chains as predictor of progression free survival. PAGE 22 (2013) Abstr 2732 [www.page-meeting.org/?abstract=2732].

• Rotshteyn Y, Mercier F, Bruno R, Stambler N, Israel RJ, Wong V. Correlation of PSMA ADC exposure with reduction in tumor growth rate determined using serial PSA measurements from a phase I clinical trial. Journal of Clinical Oncology, 31, 2013 (suppl; abstr e16047).

• Claret L, Mancini P, Sebastien B, Veyrat-Follet C, Bruno R. Model-based estimates of tumor growth inhibition (TGI) metrics to predict for overall survival (OS) in first-line non-small cell lung cancer (NSCLC). Journal of Clinical Oncology, 31, 2013 (suppl; abstr e19049).

• Bruno R, Hsu C-P, Claret L, Lu J, Sun Y-N. Exploratory modeling and simulation to support development of motesanib in Asian patients with non-small cell lung cancer (NSCLC) based on MONET1 study results. Journal of Clinical Oncology, 31, 2013 (suppl; abstr e19103).

• Claret L., Gupta M., Han K., Joshi A., Sarapa N., He J., Powell B., Bruno R. Prediction of overall survival or progression free survival by disease control rate at week 8 is independent of ethnicity: Western versus Chinese patients with first‐line non‐small cell lung cancer treated with chemotherapy with or without bevacizumab. J Clinical Pharmacol, 54, 253-257, 2014.

• Claret L., Bruno R., Lu J.F., Sun Y.N. & Hsu C.P. Exploratory modeling and simulation to support development of motesanib in Asian patients with non-small cell lung cancer based on MONET1 study results. Clin. Pharmacol. Ther. 95, 446-451, 2014.

• Bruno R., Mercier F., Claret L. Evaluation of tumor-size response metrics to predict survival in oncology clinical trials. Clin Pharmacol Ther, 95, 386-393, 2014 (Invited State of the Art paper).

• Mercier F, Houk B, Claret L, Milligan P, Bruno R. A model relating overall survival to tumor growth inhibition in renal cell carcinoma patients treated with sunitinib, axitinib or temsirolimus. ESMO 2014, Ann Oncol 25 (Supplement 4): iv146–iv164, 2014.

23

obstacles to dose optimization in early stage...

Documents