research pharmacokinetic/pharmacodynamic modeling ...ing hypoxic cytotoxins from two different...
TRANSCRIPT
Can
PhaSN3ImpCel
KevinH.D. S
Abst
Hypa cons(1). Inprogre
AuthorUniversRadiatStanfor
Note: SResear
CurrentBiology
CurrentAucklan
CorresResearland, Nwr.wilso
doi: 10
©2010
Clin C4946
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
Clinical
Canceresearchcer Therapy: Preclinical
rmacokinetic/Pharmacodynamic Modeling Identifies0000 and SN29751 as Tirapazamine Analogues withroved Tissue Penetration and Hypoxic
R
l Killing in Tumors
O. Hicks1, Bronwyn G. Siim1, Jagdish K. Jaiswal1, Frederik B. Pruijn1, Annie M. Fraser1, Rita Patel1, Alison Hogg1,
arath Liyanage1, Mary Jo Dorie2, J. Martin Brown2, William A. Denny1, Michael P. Hay1, and William R. Wilson1ractPur
ited canalosidersExp
througtures,diffusmice)Res
ing hy1-oxidthat ocell cuproviddictiohuma
a numbssion and
s' Affiliatioity of Auckion Oncolod University
upplementach Online (h
address fo, University
address fd, New Zea
ponding Ach Centre, Tew Zealandn@aucklan
.1158/1078-
American A
ancer Res
Downloa
pose: Tirapazamine (TPZ) has attractive features for targeting hypoxic cells in tumors but has lim-linical activity, in part because of poor extravascular penetration. Here, we identify improved TPZgues by using a spatially resolved pharmacokinetic/pharmacodynamic (SR-PKPD) model that con-tissue penetration explicitly during lead optimization.erimental design: The SR-PKPD model was used to guide the progression of 281 TPZ analoguesh a hierarchical screen. For compounds exceeding hypoxic selectivity thresholds in single-cell cul-SR-PKPD model parameters (kinetics of bioreductive metabolism, clonogenic cell killing potency,ion coefficients in multicellular layers, and plasma pharmacokinetics at well tolerated doses inwere measured to prioritize testing in xenograft models in combination with radiation.ults: SR-PKPD–guided lead optimization identified SN29751 and SN30000 as the most promis-poxic cytotoxins from two different structural subseries. Both were reduced to the correspondinge selectively under hypoxia by HT29 cells, with an oxygen dependence quantitatively similar tof TPZ. SN30000, in particular, showed higher hypoxic potency and selectivity than TPZ in tumorltures and faster diffusion through HT29 and SiHa multicellular layers. Both compounds alsoed superior plasma PK in mice and rats at equivalent toxicity. In agreement with SR-PKPD pre-ns, both were more active than TPZ with single dose or fractionated radiation against multiplen tumor xenografts.clusions: SN30000 and SN29751 are improved TPZ analogues with potential for targeting tumor
Conhypoxia in humans. Novel SR-PKPD modeling approaches can be used for lead optimization during an-ticancer drug development. Clin Cancer Res; 16(20); 4946–57. ©2010 AACR.
sistanisms.
oxia, frequently considered a hallmark of cancer, isequence of the inefficient vascularization of tumors
er of tumor types, hypoxia contributes tois associated with poor prognosis and re-
hypoxbreakand mof meAdditactionulatinrelativa stroing pathrouunderThe
tirapaactiviimprocispla(16).
ns: 1Auckland Cancer Society Research Centre, Theland, Auckland, New Zealand and 2Department ofgy, Division of Radiation and Cancer Biology,, Stanford, California
ry data for this article are available at Clinical Cancerttp://clincancerres.aacrjournals.org/).
r B.G. Siim: Gray Institute for Radiation Oncology andof Oxford, Oxford OX4 4GA, United Kingdom.
or A.M. Fraser: Wolters Kluwer Pharma Solutions,land.
uthor: William R. Wilson, Auckland Cancer Societyhe University of Auckland, Private Bag 92019, Auck-. Phone: 64-9-9236883; Fax: 64-9-3737571; E-mail:d.ac.nz.
0432.CCR-10-1439
ssociation for Cancer Research.
; 16(20) October 15, 2010
Research. on July 21, 20clincancerres.aacrjournals.org ded from
ce to therapeutic agents through multiple mechan-For example, in the context of radiation therapy,ic cells are less sensitive to radiation-induced DNAage and cell killing (2) and have increased invasiveetastatic potential (3), leading to failure becausetastatic disease outside the radiation field (4, 5).ional hypoxia, induced through the antiangiogenicof radiation (6), enhances tumor regrowth by stim-g vasculogenesis (7). These features, along with thee absence of hypoxia in normal tissues (8), provideng rationale for targeting hypoxia, either by inhibit-thways required for hypoxic cell survival (9, 10) orgh the metabolic activation of bioreductive prodrugshypoxic conditions (11–13).most thoroughly investigated bioreductive prodrug,zamine (TPZ), showed encouraging indications ofty in early clinical studies (14, 15) but failed tove overall survival in a recent phase III trial with
tin/radiotherapy for advanced head and neck cancerThis disappointing outcome was shown to be at21. © 2010 American Association for Cancer
least ptherapstratifthereprovetion tFur
the beadvanhypoxtial freneous(21, 21-oxidtotoxithe raprodrotherquinoelectrbe gensuch(25),featurhypoxthe intivatioreductto inhsmalleof lessThe
activitStudiecludin
is mehypoxsizedwouldetratiopoorlsues. Mhas mmacoTPZ,transpsitionhas bshowagainBTOsel to gwith(D; remodeno-BTtargetof the42), aphilicincrea(43) sence wicochbioredHer
TTOSR-PKterizecytotoand (we shprovegrafts, making them potential development candidatesfor hu
Mate
CompTPZ
and SNsis (4liquidDMSOum, o tanol/water partition coefficients at pH 7.4, andmous plasma protein binding were measured as described
Translational Relevance
Hypoxia is a ubiquitous feature of tumors and argu-ably one of the most important therapeutic targets inoncology that has yet to be exploited. Tirapazamine(TPZ) is the best-studied hypoxia-activated prodrug,but it has limited clinical activity, in part because ofinefficient penetration into hypoxic tumor tissue. Wehave used a novel spatially resolved pharmacokinetic/pharmacodynamic (SR-PKPD) modeling approachto guide optimization of TPZ analogues for improvedtissue penetration and hypoxic cell killing in tumors.The resulting compounds, SN29751 and SN30000,show clear improvement over TPZ in therapeuticactivity in multiple xenograft models in combinationwith either single dose or fractionated radiation. TheSR-PKPD tools developed in this study will assist inthe clinical development of SN30000 as a hypoxia-targeted prodrug.
3 HayM ,Hicks KO, Pruijn FB, Pchalek K, Yang S, Blaser A, LeeHH, SiimBG,
Hypoxic Cytotoxicity of Improved Tirapazamine Analogues
www.a
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
artially due to failure in compliance with radiationy protocol (17) and is also likely to reflect lack ofication for the most hypoxic tumors (18). Thus,is a reasonable expectation that TPZ, or an im-d analogue, could have a major impact on radia-herapy if developed appropriately.ther exploration of TPZ analogues is merited becausenzotriazine di-N-oxide (BTO) class has two uniquetages over other bioreductive prodrugs for targetingia. First, the active cytotoxin is derived from the ini-e radical reduction product (19, 20), which sponta-ly transforms into DNA-damaging oxidizing radicals2), whereas its subsequent reduction products (thee SR4317 and nor-oxide SR4330) are much less cy-c (19, 23). Hypoxic selectivity of TPZ derives frompid reoxidation of the initial radical to the parentug by O2 (20). In contrast, the active cytotoxins fromclasses of bioreductive prodrugs (nitro compounds,nes, and tertiary amine N-oxides) are 2-, 4-, or 6-on reduction products, which, in many cases, canerated by oxygen-insensitive two-electron reductasesas DT-diaphorase (24) or aldo-keto reductase 1C3compromising hypoxic selectivity. A second uniquee of TPZ is that it is activated under relatively mildia (26, 27); its oxygen sensitivity is approximatelyverse of that for radiotherapy, with half-maximal ac-n at ∼1 μmol/L O2 in solution (27). One-electronion of other bioreductive prodrugs is more sensitiveibition by O2 (28–30), which restricts activation to ar subset of essentially anoxic cells that are arguablyer importance for radiation therapy (31).se features led us to ask whether the limited clinicaly of TPZ could be improved by rational drug design.
s with three-dimensional cell culture models, in-g multicellular layers (MCL), have shown that TPZDennytumorIn prep
acrjournals.org
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
tabolized too rapidly to penetrate optimally intoic tumor tissue (27, 32–35). We therefore hypothe-that improving the extravascular transport of TPZincrease its therapeutic selectivity given that its pen-n limitations will have a larger effect on activity iny perfused tumors than in well-perfused normal tis-easurement of tissue transport parameters in MCLs
ade it possible to develop a spatially resolved phar-kinetic/pharmacodynamic (SR-PKPD) model forbased on Green's function models of O2 and TPZort, which describes its activity as a function of po-within a microvascular network (36). This modeleen validated as a tool for lead optimization bying that it provides reliable prediction of activityst hypoxic cells in HT29 xenografts for a set of 16(36). We have subsequently used the SR-PKPD mod-uide lead optimization, focusing on lipophilic BTOspotentially improved tissue diffusion coefficientsfs. 37, 38) but also holistically evaluating all otherl parameters. The compounds include simple 3-ami-O derivatives (39, 40), 3-amino-BTOs with DNA-ing functionality (41), 3-alkyl BTOs in which removal3-amino H-bond donor usefully increases D (40,nd tricyclic triazine di-oxides (TTO) in which a lipo-saturated ring fused to the 3-amino-BTO core alsoses D (43). The corresponding 3-alkyl-TTO subclasseeks to combine both features and builds on experi-ith the BTO series (39, 42) by adjusting other phys-emical properties to optimize the kinetics ofuctive metabolism, solubility, and systemic PK.e, we identify a 3-alkyl BTO (SN29751) and a 3-alkyl(SN30000) as the preferred compounds from thisPD–guided lead optimization program. We charac-their physicochemical properties, hypoxia-selectivexicity, cellular pharmacology, and PK in plasmaby modeling) in hypoxic regions of tumors. Further,ow that these novel TPZ analogues provide im-d therapeutic activity against human tumor xeno-
man therapy.
rials and Methods
ounds(44) and SN29751 (45)were synthesized as reported,30000 bymodification3 of amethod for TTO synthe-
3). Compounds [purity >95% by high-performancechromatography (HPLC)] were stored at −20°C andstock solutions at −80°C. Solubility in culture medi-
ceP
WA, Wilson WR. A new class of hypoxia-selective agents with anti-activity: 7,8-dihydro-6H-indeno[5,6-e][1,2,4]triazine 1,4-dioxides.aration.Clin Cancer Res; 16(20) October 15, 2010 4947
21. © 2010 American Association for Cancer
previoPhysCInc.).at leasstudie
Cell cCel
lectionfrozenheat-iby wein spisinglegenicigrown
In vitAs
expos(Bactrby sulwas dHill exicity
In vitMet
killingstirredmediu(<100ouslyfugedappar(kmet)by linmediuand vplatedfittedwhichativeportioparen
whereunit ctermi(35),tial ming dintegr
whereis giv
E wasefficie(35)
OxygClo
of oxprevio(3 ×respirwhichnescetratiointervas aboHill eby reg
DiffuDif
minechamequilitabolimentsassaycountdrugconceusing
High-Dru
2.1 m1100studietion (ing prwithconceammowere i
AnimAll
as prenu/nuwereFreshi.p. t
Hicks et al.
Clin C4948
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
usly (36). pKa values were calculated using ACD/hem v. 8.0 (Advanced Chemistry Development,For in vitro experiments, DMSO stocks were dilutedt 100-fold into culture medium, whereas for in vivos, compounds were formulated in normal saline.
ulturel lines, obtained from the American Type Culture Col-, were cultured as monolayers from Mycoplasma-freestocks at <3-month intervals in αMEM with 5%
nactivated fetal bovine serum without antibioticsekly passage. Multicellular spheroids were grownnner flasks with 10% FCS and dissociated to give-cell suspensions for drug metabolism and clono-ty assays or for inoculation into mice. MCLs wereas described (36).
ro growth inhibition assaysdescribed previously (36, 39), log-phase cells wereed to drugs for 4 hours under 20% O2 or anoxiaon anaerobic chamber), and growth was assessedforhodamine B staining 4 to 5 days later. The IC50
etermined by nonlinear regression to the standardquation, and the intra-experiment hypoxic cytoto-ratio (HCR) calculated as aerobic IC50/anoxic IC50.
ro drug metabolism and clonogenic cell killingabolic consumption of drugs and clonogenic cellwere measured simultaneously using magneticallysingle-cell suspensions (1-2 × 106 cells/mL) inm without serum, equilibrated with 5% CO2/N2
ppm O2) or 5% CO2/20% O2 as described previ-(35). Samples were removed at intervals and centri-; supernatants were stored at −80°C for HPLC. Theent first-order rate constant for drug metabolismwas determined from the concentration-time dataear regression. Cell pellets were resuspended in freshm; cell number was determined by Coulter counter,iability by 0.4% trypan blue exclusion. Cells wereto determine clonogenic survival. The data wereusing the previous (36) cellular PKPD model inthe rate of log cell kill [LCK; defined as the neg-
log of the surviving fraction (SF)] at time t is pro-nal to both the rate of drug metabolism and thet drug concentration C:
d log SF
dt¼ γC
dMdt
ðAÞ
dM/dt = kmet × C/ϕ is the metabolized drug perell volume, ϕ is the HT29 cell volume fraction de-ned from the cell count as previously describedγ is the proportionality constant, and C0 is the ini-easured drug concentration. To allow for decreas-rug concentrations during exposure, Eq. A wasated to give
−log SF ¼ γE ðBÞ10 mLments
ancer Res; 16(20) October 15, 2010
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
the exposure integral E, at each sampling time, t,en by:
E ¼ C20
2 ϕ1−e−2 ϕkm e tT� � ðCÞ
plotted against log SF to obtain the potency co-nt γ by regression as previously validated for TPZand TPZ analogues (36).
en dependence of cytotoxicity in vitronogenic cell killing was measured over a rangeygen concentrations in stirred cell suspensions asusly described (27), using lower cell densities105 cells/mL) to minimize the effect of cellularation on solution oxygen concentrations (Cs),were monitored using an OxyLite 2000 O2 lumi-
nt probe (Oxford Optronix Ltd). Drug concen-ns were also monitored, at 30- to 60-minuteals, by HPLC, and the cell survival data were fittedve. The potency as a function of Cs was fitted to aquation to estimate KO2
(Cs to halve anoxic potency)ression.
sion in multicellular layersfusion through HT29 and SiHa MCLs was deter-d as described previously (36, 37) using a two-ber diffusion apparatus (46). The medium wasbrated with 95% O2 to suppress bioreductive me-sm. Compounds were added to the donor compart-with 14C-urea and samples taken at intervals todrug (by HPLC) and 14C-urea (liquid scintillationing) in both compartments. MCL thicknesses anddiffusion coefficients (DMCL) were fitted to thentration-time profiles of urea and drug, respectively,a Fick's second law mathematical model (36).
performance liquid chromatographyg concentrations were determined with a 150 mm ×m Alltima C8 reverse phase column and AgilentHPLC using photodiode array detection. For in vitros, culture supernatants were analyzed by direct injec-35). Plasma samples (below) were analyzed follow-evious methods (36) after precipitation of proteinsacetonitrile, evaporation to dryness in a Speed-Vacntrator, and reconstitution in 125 μL of 45 mmol/Lnium formate buffer (pH 4.5), of which 100 μLnjected.
al toxicityanimal experiments followed institutional protocolsviously described (36). Specific pathogen-free CD-1mice (∼25 g) or Sprague-Dawley rats (∼200 g)
ear-tagged and randomized to treatment groups.ly prepared solutions of compounds were giveno mice at 20 mL/kg body weight or to rats at
/kg (25 mL/kg for TPZ), using 1.33-fold dose incre-. The maximum tolerated dose (MTD) was definedClinical Cancer Research
21. © 2010 American Association for Cancer
as thebodyin a gtime o
PlasmMic
eachunderat 1740 μLwitho(3,00for HPmined
XenogTum
CD-1inocumorsdosing9 a.mschedMice60 gasinglestrainmor rsuspethe enscribein logStatistone-wparisotimestimes
SpatiaOxy
latedcrovasR3230the tidrugdrugPK. DmicroPKPDwas ael pardrugmicrobetweand tpredichypox
Resu
PKPDAn
proveand Tturalhypoxeightyagainsevaluathe pinclumetabrate calso cfromhydropreviodimenkmet,0
diverslyingrior ptive toA fi
predicthe raaerobHT29(AUCpasseand ttwicemousthe M75%the mat 75pointtestinin HTMTD.activefurthetestedbe inaof theloguecandireducgues showed ≥5-fold higher aqueous solubility than TPZdespite increased lipophilicity at neutral pH (Table 1).
4 HayMP, Hicks KO, Pruijn FB, Pchalek K, Yang S, Blaser A, LeeHH, SiimBG,
Hypoxic Cytotoxicity of Improved Tirapazamine Analogues
www.a
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
highest dose that caused no drug-related deaths,weight loss of more than 15%, or severe morbidityroup of three to six animals, with an observationf 28 days.
a pharmacokineticse were injected i.p. or i.v. at 75% of the MTD forcompound, with sampling by cardiac punctureterminal CO2 anesthesia. Rats were injected i.p.
8 and 316 μmol/kg, and blood samples of 20 towere obtained serially from the saphenous vein
ut anesthesia. Blood was immediately centrifuged0 × g, 3 minutes), and plasma stored at −80°CLC. Noncompartmental PK parameters were deter-by WinNonLin v 5 (Pharsight Corp.).
raft modelsor xenografts were grown s.c. in the dorsal flank ofnude mice, 1.5 cm from the base of the tail, bylation of 107 cells. Treatment was initiated when tu-reached a mean diameter of 9 to 11 mm, using i.p.with either a single dose or multidose [bidaily at. and 3 p.m. for 4 consecutive days (BID1-4)]ule, at various times before or after irradiation.were irradiated without anesthesia using a cobalt-mma source, either whole-body (15 or 20 Gy) fordose studies or locally using a custom-built re-
ing jig for multidose studies (2 or 2.5 Gy × 8). Tu-esponse was assessed by clonogenic assay of cellnsions from xenografts removed 18 hours afterd of treatment (excision assay) as previously de-d (36). LCK was calculated from the difference(clonogens/g) for treated and control groups.
ical significance of drug effects was tested usingay ANOVA with Dunnett's test for multiple com-ns. Alternatively, tumors were measured threeweekly using calipers until tumors reached threethe pretreatment volume (regrowth assay).
lly resolved PKPD modelinggen and drug concentration gradients were calcu-using Green's function methods in a mapped mi-cular network (500 × 500 × 230 μm) from a ratAc tumor as previously described (36), based onssue diffusion coefficients and rate constants formetabolism measured in vitro above with inflowconcentration defined by the measured plasmarug-induced cell kill at each point in the tumorregion was then calculated using the above cellularmodel. Radiation-induced cell kill at each point
lso calculated using reported linear-quadratic mod-ameters (36). The predicted surviving fraction forand radiation was averaged over the whole tumorregion to calculate the overall LCK. The differenceen the surviving fraction for drug with radiationhe surviving fraction for radiation alone gave the
ted drug-induced LCK in the radiation-resistantic cell population.DennytumorIn prep
acrjournals.org
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
lts
-guided lead optimizationoverview of the algorithm we used to identify im-d TPZ analogues is shown in Fig. 1A. Of 281 BTOTO compounds synthesized from five different struc-classes, 225 had sufficient solubility to determineia-selective cytotoxicity in IC50 assays. One hundred-two passed decision point “A” in Fig. 1A (HCR >20t both HT29 and SiHa cells), and 173 of these wereted in additional tissue culture models to determinearameters required for SR-PKPD modeling. Thisded measurement of clonogenic cell killing andolic drug consumption under anoxia (first-orderonstant kmet,0) in stirred HT29 cell suspensions. Wealculated tumor tissue diffusion coefficients (D)physicochemical parameters (MW, log P at pH 7,gen bond donors and acceptors) using relationshipsusly established with a training set of BTOs in three-sional MCL cultures (38). The ranges of values forand D are shown in Fig. 1B, which illustrates theity of extravascular transport properties; compoundsbelow the diagonal line are predicted to have supe-enetration distances into hypoxic HT29 tissue rela-TPZ (40).
rst iteration of the SR-PKPD model was then run tot the hypoxic cytotoxicity differential [HCD = LCK indiobiologically hypoxic region (<4 μmol/L O2)/ic region (>30 μmol/L O2)] of each compound inxenografts and the plasma area under the curve) required for a hypoxic LCK of >0.5. Compoundsd this decision point (“B” in Fig. 1A) if HCD >1he required AUC was <333 μmol/L h, which isthat of TPZ at its MTD in initial PK studies in thise strain (36). For 115 compounds, we determinedTD in CD-1 nude mice and the plasma AUC atof MTD. The SR-PKPD model was then rerun usingeasured AUC to identify compounds with LCK >0.3% MTD. Compounds passing this third decision(“C” in Fig. 1A) were then evaluated in vivo by
g their ability to kill radiation-resistant hypoxic cells29 xenografts following a single drug dose at 75% ofSixteen of the 18 compounds predicted to beshowed statistically significant activity, whereas ar 12 compounds predicted to be inactive were alsoto evaluate the algorithm and were all found toctive. SN29751 was identified as the most promisingBTO series (42) and SN30000 as the preferred ana-in the new TTO series.4 The structures of these leaddates are shown in Fig. 1C, along with their stable 2etion products (1-oxides 1 and 2). These new analo-
WA, Wilson WR. A new class of hypoxia-selective agents with anti-activity: 7,8-dihydro-6H-indeno[5,6-e][1,2,4]triazine 1,4-dioxides.aration.
Clin Cancer Res; 16(20) October 15, 2010 4949
21. © 2010 American Association for Cancer
Bioreand SA m
the punderculturloss ooxideidentibanceFig. Sductiv(1.53diateSN297SN2
selecti
feratiwereselecteightthanpotencompTPZ afor SNcommhighehigheacrosconfi(Fig.
Fig. 1.optimizkmet,0, aSN2975cells (2 met
reduced metabolites. Points, mean from triplicate cultures; bars, SEM.
Hicks et al.
Clin C4950
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
ductive metabolism and cytotoxicity of SN29751N30000ore detailed investigation of each component of
harmacology of these new lead compounds wastaken. The metabolism of the prodrugs in HT29es, illustrated in Fig. 1D, showed hypoxia-dependentf the parent compounds with formation of the 1-s (and for SN30000 a trace of the nor-oxide 3)fied by comparison of retention time and absor-spectra with authentic standards (Supplementary
1). For all experiments, the rate constant for biore-e consumption (kmet,0) was highest for SN30000± 0.21 min−1; errors are SEM throughout), interme-for TPZ (1.30 ± 0.04 min−1), and lowest for51 (0.87 ± 0.03 min−1).
9751 and SN30000, like TPZ, showed hypoxia-ve cytotoxicity in HT29 and SiHa IC50 (antiproli-(coefSN30
ancer Res; 16(20) October 15, 2010
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
ve) assays, whereas the reduced metabolitessubstantially less cytotoxic and lacked hypoxicivity (Fig. 2A). When this assay was extended tocell lines, SN30000 was consistently more potentTPZ under anoxia, whereas SN29751 was lesst (Fig. 2B). Notably, the anoxic potency of bothounds was highly correlated with that ofcross the cell lines (Fig. 2B; R2 = 0.97 and 0.9329751 and SN30000, respectively), suggesting aon mechanism of action. Consistent with its
r potency under anoxia, the HCR was consistentlyr for SN30000 than for TPZ and SN29751s all cell lines (Fig. 2C). These findings werermed for HT29 cells by clonogenic assay2D), which showed the highest anoxic potency
Optimization of TPZ analogues and bioreductive metabolism of the identified lead candidates SN29751 and SN30000. A, SR-PKPD–guided leadation algorithm. See the text for explanation. B, range of extravascular transport parameters (rate constant for reductive metabolism under anoxia,nd calculated tissue diffusion coefficient, D) for compounds passing decision point “A.” C, structures of TPZ, the lead 3-alkyl-BTO analogue1, the lead 3-alkyl-TTO SN30000, and their reduced metabolites. D, metabolism of TPZ, SN29751, and SN30000 in stirred suspensions of HT29× 106/mL) under aerobic and hypoxic conditions, determined by HPLC. Left, parent di-N-oxides with regression lines to estimate k . Right,
ficient γ in Table 1) and HCR values for000.
Clinical Cancer Research
21. © 2010 American Association for Cancer
PenetThe
in HTpropepredicthrougen cservedas simlues f2.5-fotrendMCL
OxygeWe
SN300descriSR-PK
TPZ. Asurvivcreasiwere uoxygeto ha(KO2
∼(one-wconsushowewith 5
ToxicFol
CD-1SN29that i
Tabfor
Par
PhyMSElopH
SR-k
KγDϕD
CA
t1θ
NOT*Fro†Fro‡Frohyp§Fro∥Fro¶Fro
Hypoxic Cytotoxicity of Improved Tirapazamine Analogues
www.a
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
ration of multicellular layer culturesdiffusion coefficients of SN30000 and SN2975129 MCL, calculated from their physicochemicalrties, were higher than that of TPZ (Fig. 1B). Thesetions were confirmed by measuring penetrationgh aerobic HT29 MCL (Fig. 3A). At these high oxy-oncentrations, no reduced metabolites were ob-and diffusion coefficients (DMCL) could be fittedple diffusion without metabolism (Fig. 3B); the va-or SN30000 and SN29751 were respectively 3- andld higher than that for TPZ (Table 1). This same(SN30000 > SN29751 > TPZ) was seen in SiHa(Supplementary Fig. S2).
n dependence of cytotoxicity and metabolismdefined the oxygen dependence of SN29751 and00 activation using the same approach as previously
bed (27), thus testing the assumption in the initialPD screen (Fig. 1A) that this was the same as forTPZ (rized
lepr
am
sicWolu(1)gKa
D,
PK
met
O2
[×(×
os
ma
UC
m ref. 36.
acrjournals.org
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
s illustrated for SN30000 (Fig. 3C), the slope of theal versus exposure integral plots decreased with in-ng oxygen concentration in solution. These slopessed to determine cytotoxic potency as a function ofn (Fig. 3D). The oxygen concentration requiredlve the anoxic potency of SN30000 and SN297511 μmol/L; Table 1) was not significantly differentay ANOVA, P = 0.26) from that for TPZ. Metabolicmption of the prodrugs, in the same experiments,d similar oxygen dependence (Supplementary Fig. S3)0% inhibition at approximately 0.6 to 1 μmol/L O2.
ity and plasma pharmacokinetics in mice and ratslowing single i.p. administration, the MTD in malenude mice was 1,000 μmol/kg (370 mg/kg) for751 and 750 μmol/kg (275 mg/kg) for SN30000,s, 5.6- and 4.2-fold higher molar doses than for
178 μmol/kg, 31.7 mg/kg), respectively. As summa-in Supplementary Table S1, these differences in host0a
1. Physicochemicalediction of hypoxic c
properties of TPZ, SN29751, and SN30ell killing in HT29 xenografts in CD-1 m
±
±±±0
.
21. © 20
00 and SR-PKle nude mice
Clin Cance
10 American A
PD model parat 75% of MT
r Res; 16(20) Oct
ssociation for C
ametersD
eter Description TPZ SN29751 SN30000
ochemical properties
(Da) Molecul ar weight 178 3 71 3 67 bility (mmol/L) Solubilit y in culture medium at 37°C 8.9* >4 6.7† 4 8.5‡(mV) One-ele
ctron reduction potential −45 6 ± 8* −440 ± 9† −39 9 ± 8‡P, pH 7.4 Octanol
/water partition coefficient at pH 7.4 −0.3 4 ± 0.02* 0.13 ± 0.01† 0.50 ± 0.05‡Calculated amine pKa NA 7.3 7.1‡
HA Number of hydrogen bond donors, acceptors 2, 6 0, 8 0, 7
PD model parameters
,0 (min−1) First-ord
er rate constant for prodrug metabolism 1.30 0.04 (74) 0.87 ± 0.03 (4) 1.53 ± 0.21 (8) unde r anoxia(μmol/L) O2 conc
entration to halve anoxic cytotoxic potency 1.21 0.09§ 0.81 ± 0.19 1.14 ± 0.24 10−5 (μmol/L)−2] Proport ionality constant in PD model 2.41 0.13 (72) 1.25 ± 0.31 (4) 5.34 ± 0.89 (13) 10−6 cm2 s−1) Diffusio n coefficient in HT29 MCLs and tumors 0.40lume fraction of HT29 MCLs and tumors
0.02 (12) 1.07 ± 0.08 (4) 1.26 ± 0.11 (6)Cell voe (μmol/kg) Dose co
rresponding to 75% of MTD, used for PK.517∥
133 7
50 5 62 and P D evaluation in miceum plasma concentration in mice 61
x (μmol/L) Maximp∞ (μmol-h/L) Area und
er the plasma (total drug) concentration- 7 ± 0.1 149.649 1± 25.6 85.162
± 7.879
time c
urve extrapolated to infinity in mice/2 (min) Plasma terminal half-life in mice 27.9 34.8 32.1(%) Free fraction of drug in plasma in mice 100 ± 5.3¶ 98 ± 2 83 ± 1
E: Values are mean and SEM, with the number of determinations given in parentheses.m ref. 39.m ref. 40.m: Hay MP, Hicks KO, Pruijn FB, Pchalek K, Yang S, Blaser A, Lee HH, Siim BG, Denny WA, Wilson WR. A new class ofoxia-selective agents with anti-tumor activity: 7,8-dihydro-6H-indeno[5,6-e][1,2,4]triazine 1,4-dioxides. In preparation.m ref. 27.m ref. 35.
ober 15, 2010 4951
ancer
toxicithe thC57/Bin CDdosin421 μand 7His
dosesitativSN30dosesand anounctopathfor SNfindininmicafter trInit
MTD
half-lifor TPHowefor thand 7tively,i.v. dSN30Com
Dawlegues (in bo421 μtive plives,for th
SR-PKcomb
Fig. 2.experimcell line*, P < 0
Hicks et al.
Clin C4952
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
ty (SN29751 < SN30000 < TPZ) were also seen inree other mouse strains investigated (C3H/HeN,l6, and Rag-1−/−) and with other dosing schedules-1 nude mice. For example, the MTD for twice-dailyg (9 a.m. and 3 p.m.) for 4 consecutive days wasmol/kg for SN29751, 237 μmol/kg for SN30000,5 μmol/kg for TPZ.topathology of CD-1 nude mice following single drugor the BID1-4 schedule at 1-1.3 ×MTD showed a qual-ely similar organ toxicity profile for SN29751,000, and TPZ (Supplementary Table S2). For single, bone marrow hypoplasia, gastrointestinal toxicity,irway epithelium vacuolation were the most pro-ed acute findings. Following the BID1-4 schedule, his-ologic changes were similar to the single-dose study29751, but gastrointestinal toxicity was the only acuteg for SN30000 andTPZ.As reported previously for TPZe (47), significant retinal toxicity was observed 28 dayseatment with all three agents (Supplementary Fig. S4).
ial screening of PK following single i.p. doses at 75% Thes. Lines are linear regressions. C, hypoxic cytotoxicity ratio (HCR; aerobic IC50/hyp.05; **, P < 0.01. D, clonogenic survival of HT29 cells as a function of exposure
ancer Res; 16(20) October 15, 2010
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
ves of all three compounds (28, 35, and 32 minutesZ, SN29751, and SN30000, respectively; Table 1).ver, increased plasma concentrations were achievede analogues, resulting in AUC values of 49, 162,9 μmol-h/L for TPZ, SN29751, and SN30000, respec-at these equitoxic doses (Table 1). Comparison withosing showed an i.p. bioavailability of 77% for000 (Supplementary Table S3).parison of toxicity and plasma PK in Sprague-y rats also showed higher MTD values for the analo-178 μmol/kg for TPZ and 316 μmol/kg for SN30000th sexes; 316 μmol/kg for SN29751 in females andmol/kg in males) after single i.p. doses. Representa-lasma PK is shown in Fig. 4B; Cmax, terminal halfand AUC (Supplementary Table S4) were all highere analogues than for TPZ.
PD modeling of tumor cell killing inination with radiation
measured parameters of the SR-PKPD model, sum-in CD-1 nude mice (Fig. 4A) indicated similar plasma marized in Table 1, allow a complete description of PK and
Hypoxia-selective cytotoxicity of TPZ, SN29751, and SN30000 in human tumor cell cultures. Values are means and SEM for 2 to 10ents. A, IC50 for HT29 and SiHa cells following 4-h drug exposure under aerobic or hypoxic conditions. B, hypoxic IC50 in eight human tumor
oxic IC50) in the same cell lines. Differences from TPZ by ANOVA:under hypoxia (0% O2) and oxic conditions (20% O2).
Clinical Cancer Research
21. © 2010 American Association for Cancer
PD (cthe mpoundprediSN30host tplemeradia(dashdrugsumedall tisfor ramodea cleaTPZ au`latetumoron thsolvedshowthe anand ththerapmarke
ActiviThe
mentgraftsLCK fisteredradiatlargewas ssignifout rawith snot shto TPtion otimetion aactivitcompas forkillingThe
also e
Fig. 3.parameSN3000of an Hdrug trachambeby theTPZ (gr(open s(black freceivechambethe MCexpressexpecteequilibrthe 14C(insteaddifferenreplicat(with dicompodata plointegralMethodsuspenat the oindicateof cyto(open s(closedthe preTPZ demethodfits with
2
parameter.
Hypoxic Cytotoxicity of Improved Tirapazamine Analogues
www.a
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
ell killing) at each point within HT29 xenografts;odel output is shown in Fig. 4C for all three com-s at 75% of MTD in CD-1 nude mice. The modelcts a large increase in hypoxic cell killing by000 and SN29751 relative to TPZ at equivalentoxicity. In both cases, killing shows excellent com-ntarity to that calculated for a large single dose oftion (20 Gy), which spares the hypoxic cellsed line in Fig. 4C). To model overall LCK by eachwhen used in combination with radiation, we as-independent action of both agents and summed
sue regions for the combination, subtracting LCKdiation alone. This additional LCK and thel-predicted hypoxic selectivity (HCD) both showedr increase for SN30000 and SN29751 relative tot equivalent host toxicity (Fig. 4D). We also sim-d the activity of the compounds against an HT29notionally grown in Sprague-Dawley rats, based
e measured plasma PK at MTD. The spatially re-PD predictions (Supplementary Fig. S5) again
a large improvement in hypoxic cell killing foralogues relative to TPZ at equivalent host toxicity,e combined activity with radiation for this “virtual
eutic trial” (Fig. 4D) is again predicted to bedly superior to TPZ.endpoto rad
acrjournals.org
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
ty against human tumor xenograftsSR-PKPD model predictions were in good agree-with the measured killing of HT29 cells in xeno-by excision assay (Fig. 5A), which showed greateror SN30000 and SN29751 than TPZ when admin-at 75% of MTD 5 minutes after a single dose of
ion (to sterilize well-oxygenated cells). A similarincrease (∼3-fold) in hypoxic tumor cell killingeen with SiHa and H460 xenografts (Fig. 5A). Noicant activity was seen with the drugs alone (with-diation) against any of the xenografts, consistentelective killing of the hypoxic subpopulation (dataown). The analogues were also generally superiorZ when administered 30 minutes before each frac-f a BID1-4 radiotherapy schedule (Fig. 5A). Thecourse of interaction between the drugs and radia-gainst SiHa tumors (Fig. 5B) also showed highery of SN29751 and SN30000 and showed that theounds were active both before and after irradiation,TPZ, showing that the mechanism is hypoxic cellrather than direct radiosensitization.activity of SN30000 with fractionated radiation wasvaluated in SiHa tumors using tumor regrowth as the
Extravascular transportters for TPZ, SN29751, and0. A, H&E-stained sectionT29 MCL. The direction ofnsport in the diffusionr experiments is indicatedarrow. B, concentrations ofay symbols), SN29751ymbols), and SN30000illed symbols) in ther compartment of diffusionrs after flux throughL. Concentrations areed as fractions of thed concentration atium and plotted against-urea internal standardof time) to account forces in MCL thickness. Twoe experiments are shownfferent symbols) for eachund. C, surviving fractiontted against drug exposure(see Materials ands) for HT29 single-cellsions exposed to SN30000xygen concentrationsd. D, oxygen dependencetoxic potency for SN29751ymbols) and SN30000symbols) compared withviously reported values fortermined using the same(27). Curves are modelthe KO as the fitted
int (Fig. 5C), showing significant activity additionaliation at the two highest drug doses tested. Comparison
Clin Cancer Res; 16(20) October 15, 2010 4953
21. © 2010 American Association for Cancer
withwhereweighSN30inhibiactivit
Discu
Thiprovecytotopenetagainsmostdepardurin
and sselectIn conassess
Fig. 4.male CHT29 tuinput intumorsfor an H
Hicks et al.
Clin C4954
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
TPZ in the same experiment is shown in Fig. 5D,the time for tumor regrowth is compared with bodyt loss (as a measure of toxicity). Both TPZ and000 showed dose-dependent body weight loss and
tion of tumor regrowth, but with greater antitumor to anTPZ aprodrtionpharmand tin tumaspecents ipredicthe tuThe
y for SN30000 relative to toxicity.
ssion
s study identifies novel TPZ analogues with im-d formulation properties (aqueous solubility),xic potency against hypoxic cells in culture, tissueration characteristics, and therapeutic activityt hypoxic cells in multiple tumor models. Perhapsnotably, the approach we have taken represents a
ture from the usual strategy for lead optimization confirbased on simulations in C. Selectivity is indicated by the hypoxic cytotoxicity diT29-like tumor in male rats is based on the plasma PK in B.
ancer Res; 16(20) October 15, 2010
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
electivity in monolayer cell cultures are used tocompounds for evaluation in in vivo models.trast, we use SR-PKPD modeling to make a holisticment of the molecular features that contributetitumor activity in vivo. The SR-PKPD model fornalogues incorporates the relationships betweenug metabolism, cytotoxicity, and oxygen concentra-as determined in vitro; the systemic (plasma)acokinetics of the compounds at tolerated doses;heir extravascular transport (tissue penetration)ors as assessed with the MCL model. The latter
t enables calculation of drug concentration gradi-n a representative microvascular network and, thus,tion of cell killing probability at each position inmor tissue.utility of this SR-PKPD approach is shown by
med hypoxia-selective cell killing in HT29 xeno-
g anticancer drug development in which potency grafts for 16 of 18 compounds predicted by the modelPlasma PK of i.p. administered TPZ, SN29751, and SN30000 in mice and rats and SR-PKPD model predictions. A, plasma PK at 75% MTD forD-1 nude mice (n = 3 at each time). B, plasma PK at MTD for male Sprague-Dawley rats (n = 2). C, predictions of the SR-PKPD model for cell killing inmors (plotted as a function of oxygen concentration at each of the 4,000 grid positions in the microvascular network) based on plasma PKthe mice described in A. D, predicted whole-tumor average LCK by the drugs when combined with radiation (additional to radiation only) in HT29
fferential (HCD) as defined in the text. The analogous simulation
Clinical Cancer Research
21. © 2010 American Association for Cancer
to behypoxsignifmodeHT29parepound(despas truadoptseriesrat R3functitive osensitand tuNot
potenrange
showe0.16)correlFig. S6are sysistenTPZ itrepresprovevasculhigheas formakesnograSN30and rfeatur
Fig. 5.xenogrradiatiocompomean oof tumoradiatio263 μm[P = 0.0 wth dedata ar t 45, 90
Hypoxic Cytotoxicity of Improved Tirapazamine Analogues
www.a
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
active (one of the two false positives showed aic LCK of 0.60 ± 0.36, which did not reach statisticalicance) and the quantitative agreement betweenl prediction and measured hypoxic cell killing inxenografts for TPZ, SN29751, and SN30000 [com-Figs. 4A and 5A (left)]. In contrast, 12 of 12 com-s predicted by the SR-PKPD model to be inactive
ite good hypoxic selectivity in vitro) were confirmede negatives. This suggests that the assumptionsed in the modeling are acceptable for this drug, including that the microvascular geometry of the230Ac tumor (used as the basis for our Green'son oxygen and drug transport models) is representa-f human tumor xenografts and that the intrinsicivity of HT29 cells is the same in single-cell culturemors.ably, the inactive compounds in vivo showed hypoxic
004 and P = 0.0058, respectively (log-rank test)]. D, comparison of regroe for the same experiments described in C, which included TPZ groups a
cy and selectivity for HT29 cells in culture in the sameas for the active compounds, and linear regression
coefficytoto
acrjournals.org
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
d that neither hypoxic potency (IC50; r2 = 0.07, P =
nor selectivity (log10 HCR; r2 = 0.02, P = 0.47) in vitroated with hypoxic LCK in tumors (Supplementary). Thus, the key determinants of therapeutic activitystemic PK relative to toxicity and intratumor PK [con-t with the evidence that this is a limiting feature forself (32); refs. 27, 33–36]; these features are explicitlyented in the model. In the case of SN30000, the im-ment over TPZ is mainly driven by its superior extra-ar penetration (Fig. 3B; Supplementary Fig. S2) andr cytotoxic potency under hypoxia (Fig. 2B-D), where-SN29751, improved systemic (plasma) PK (Fig. 4A)a larger contribution to its higher activity against xe-fts inmice. The improved extravascular penetration of000 is consistent with its higher lipophilicity (Table 1)emoval of the H-bond donor 3-NH2 group, thesees being the major determinants of MCL diffusion
lay of SiHa tumors and body weight loss at nadir. The SN30000, and 135 μmol/kg/dose using the same eight-dose schedule.
Activities of TPZ, SN29751, and SN30000 against human tumor xenografts in combination with radiation. A, comparison of cell killing in threeaft lines by tumor excision and clonogenic assay 18 h after the end of treatment (columns, mean of five mice; bars, SEM). Left, single-dosen (15 Gy for SiHa, 20 Gy for HT29 and H460), with compounds administered 5 min after irradiation. Right, fractionated radiation (2.5 Gy × 8), withunds administered 30 min before each radiation dose. B, time course of interaction with radiation against SiHa tumors by excision assay (points,f five mice; bars, SEM). Left, single radiation dose (15 Gy). Right, fractionated radiation (2.5 Gy × 8). Values above the points show the numberrs excluded from the analysis because <3 colonies were recovered. C, activity of SN30000 alone and 30 min before each dose of fractionatedn (2 Gy × 8) against SiHa xenografts by tumor regrowth assay. Pooled data from two experiments; 7 to 11 mice per group. SN30000 alone atol/kg/dose was not significantly active but gave highly significant increases in radiation-induced tumor regrowth delay at both 263 and 176 μmol/kg/dose
cients for TPZ analogues (38). The higher hypoxicxicity potency of SN30000 presumably reflects its
Clin Cancer Res; 16(20) October 15, 2010 4955
21. © 2010 American Association for Cancer
57-mVquentIn effemits fpenetStu
(∼3-foTPZ wmodealso inover TSN29with fthe mto be lreflectdrugssuggevesselhypoxThe
SN300compcorrestingui(Fig. 2strongtaboliadditiclonoSN297in hom
sensitcytototion foxyge(Fig. 3histopobservpositiing thprospPK
mentprofilwithinhere poxic rand tobasedWe plcellulavirtuawill pbiomover T
Discl
W.RM.P. HinventoW.A. D
Refe1. Pri
hetCo
2. WafreBr
3. Luntas
4. Bridicsar
5. Fyldecer
6. RotumindCa
7. Kio
5 HayMDennytumorIn prep6 Patel
Hicks et al.
Clin C4956
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
higher one-electron reduction potential5 and conse-faster bioreductive metabolism (Fig. 1D; Table 1).ct, the higher diffusion coefficient of SN30000 per-aster bioreductive activation without compromisingration into the hypoxic target zone.dies with three xenograft models show a substantialld) therapeutic gain for SN30000 and SN29751 overith single-dose radiation (Fig. 5A), and SR-PKPDl predictions based on plasma PK achievable in ratsdicate a large therapeutic advantage for the analoguesPZ in this second species (Fig. 4D). SN30000 and751 are also clearly superior to TPZ in combinationractionated radiation (Fig. 5A, B, and D), althoughagnitude of improvement with HT29 tumors seemsess than for single dose radiation (Fig. 5A). Thismighta lesser penetration problem for bioreductive pro-in combination with fractionated radiation if, assted (31), moderately hypoxic cells closer to bloodsmake a larger contribution to outcome than severelyic cells in this setting.present study shows that the mechanism of action of00 and SN29751 is similar to that of TPZ. All threeounds are metabolized under hypoxia to theponding nontoxic 1-oxides (Fig. 1D), with an indis-shable cell line dependence as hypoxic cytotoxinsB). Consistent with this, we have recently shown acorrelation between TPZ and SN30000 reductive me-sm under hypoxia in a panel of 16 cell lines (48). Inon, the relationship between γH2AX formation andgenic cell killing is the same for TPZ, SN30000, and
51, and Chinese hamster ovary cell lines defectiveologous recombination repair show similar hyper- AcknWeWang a
Thepaymenadvertiindicate
Rece
uced up-regulation of metastasis-promoting gene products.ncer Res 2005;65:2387–96.i M, Vogel H, Schultz G, Hoffman RM, Harsh GR, Brown JM. Inhibi-
tionglio
8. Vatum9:1
9. Seca
10. RoresregInv
11. BroNa
12. Mcco
13. ChMe
14. vocis
P,Hicks KO, Pruijn FB, Pchalek K, Yang S, Blaser A, LeeHH, SiimBG,WA, Wilson WR. A new class of hypoxia-selective agents with anti-activity: 7,8-dihydro-6H-indeno[5,6-e][1,2,4]triazine 1,4-dioxides.aration.
ancer Res; 16(20) October 15, 2010
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
ivity to all three compounds,6 consistent with thexicity of the new analogues occurring through replica-ork arrest as for TPZ (49). Coupled with the similarn dependence of the three agents in HT29 culturesD; Supplementary Fig. S3) and similar normal tissueathology in mice (Supplementary Table S2), theseations suggest that SN30000 and SN29751 are welloned to leverage clinical experience with TPZ, includ-e use of positron emission tomography imaging toectively identify hypoxic tumors (18).evaluationof anticancer drugs during clinical develop-is typically limited to describing concentration-timees in plasma, rather than in the effect compartmenttumors. The spatially resolved PK model describedrovides a unique opportunity to simulate PK in hyp-egions of tumors during the phase I trial of SN30000compare this with an analogous simulation for TPZon its reported plasma PK at MTD in humans (50).an to link this to PD prediction in humans using ther PKPD model described here, essentially as for thel therapeutic trial in rats represented by Fig. 4D. Thisrovide an early indication, independent of responsearkers, whether SN30000 offers a therapeutic gainPZ as a hypoxic cytotoxin for use in humans.
osure of Potential Conflicts of Interest
. Wilson, J.M. Brown, and W.A. Denny: consultants, Proacta, Inc.;ay, K.O. Hicks, F.B. Pruijn, B.G. Siim, W.A. Denny, and W.R. Wilson:rs on patents related to SN30000 and SN29751; W.R. Wilson andenny: commercial research grant, Proacta, Inc.
owledgments
thank Dianne Ferry for assistance with bioanalysis and Drs. Jinglind Thorsten Melcher for critical comments on the manuscript.costs of publication of this article were defrayed in part by thet of page charges. This article must therefore be hereby markedsement in accordance with 18 U.S.C. Section 1734 solely tothis fact.
ived 05/31/2010; revised 07/19/2010; accepted 07/21/2010;
R, Hicks KO, Wilson WR. Unpubli published OnlineFirst 08/20/2010.shed data.renceses AR, Cornelissen AJ, Sloot AA, et al. Structural adaptation anderogeneity of normal and tumor microvascular networks. PLoSmputational Biology 2009;5:e1000394.rdman P. The importance of radiation chemistry to radiation ande radical biology. (The 2008 Silvanus ThompsonMemorial Lecture).J Radiol 2009;82:89–104.t SJ, Chaudary N, Hill RP. The tumor microenvironment and me-tatic disease. Clin Exp Metastasis 2009;26:19–34.zel DM, Scully SP, Harrelson JM, et al. Tumor oxygenation pre-ts for the likelihood of distant metastases in human soft tissuecoma. Cancer Res 1996;56:941–3.es A, Milosevic M, Hedley D, et al. Tumor hypoxia has indepen-nt predictor impact only in patients with node-negative cervix can-. J Clin Oncol 2002;20:680–7.fstad EK, Mathiesen B, Henriksen K, Kindem K, Galappathi K. Theor bed effect: increased metastatic dissemination from hypoxia-
of vasculogenesis, but not angiogenesis, prevents the recurrenceofblastoma after irradiation in mice. J Clin Invest 2010;120:694–705.upel P, Hockel M, Mayer A. Detection and characterization ofor hypoxia using pO2 histography. Antioxid Redox Signal 2007;221–35.menza GL. Defining the role of hypoxia-inducible factor 1 inncer biology and therapeutics. Oncogene 2010;29:625–34.uschop KM, van den BT, Dubois L, et al. The unfolded proteinponse protects human tumor cells during hypoxia throughulation of the autophagy genes MAP1LC3B and ATG5. J Clinest 2010;120:127–41.wn JM, Wilson WR. Exploiting tumor hypoxia in cancer treatment.t Rev Cancer 2004;4:437–47.Keown SR, Cowen RL, Williams KJ. Bioreductive drugs: fromncept to clinic. Clin Oncol 2007;19:427–42.en Y, Hu L. Design of anticancer prodrugs for reductive activation.
d Res Rev 2009;29:29–64.n Pawel J, von Roemeling R, Gatzemeier U, et al. Tirapazamine plusplatin versus cisplatin in advanced non-small-cell lung cancer:Clinical Cancer Research
21. © 2010 American Association for Cancer
a rtirasm
15. RisdiacaltheOn
16. RisradcarIII t201
17. Peaphea201
18. Rismisin ptoTas24:
19. Babycity
20. Ladhypdio
21. Juncleandio
22. ShRFtio3-aSo
23. Siimhyp431
24. Knica
25. Gudruald
26. KoSR
27. Hicthetion200
28. Mainflin
29. Siimcitydu
30. Hictrathedia
31. Womospo
32. Dupa
33. Hicdifusi42
34. Kytiramo
35. Hiclulatratic
36. Hictissac20
37. PruWRphtur
38. Pruofdru68
39. Hashigu
40. Hanabeext
41. Habese
42. Habeon20
43. Hadio68
44. BomeMa
45. Pchyag
46. Hictra[2-anco
47. Letive50
48. WaofanceRe20
49. Evcoind25
Hypoxic Cytotoxicity of Improved Tirapazamine Analogues
www.a
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439
eport of the international CATAPULT I study group. Cisplatin andpazamine in subjects with advanced previously untreated non-all-cell lung tumors. J Clin Oncol 2000;18:1351–9.chin D, Peters L, Fisher R, et al. Tirapazamine, cisplatin, and ra-tion versus fluorouracil, cisplatin, and radiation in patients with lo-ly advanced head and neck cancer: a randomized phase II trial ofTrans-Tasman Radiation Oncology Group (TROG 98.02). J Clincol 2005;23:79–87.chin D, Peters LJ, O'Sullivan B, et al. Tirapazamine, cisplatin, andiation versus cisplatin and radiation for advanced squamous cellcinoma of the head and neck (TROG 02.02, HeadSTART): a phaserial of the Trans-Tasman Radiation Oncology Group. J Clin Oncol0, doi: 10.1200/JCO.2009.27.4449.ters LJ, O'Sullivan B, Giralt J, et al. Critical impact of radiother-y protocol compliance and quality in the treatment of advancedd and neck cancer: results from TROG 02.02. J Clin Oncol0, doi: 10.1200/JCO.2009.27.449.chin D, Hicks RJ, Fisher R, et al. Prognostic significance of [18F]-onidazole positron emission tomography-detected tumor hypoxiaatients with advanced head and neck cancer randomly assigned
chemoradiation with or without tirapazamine: a substudy of Trans-man Radiation Oncology Group Study 98.02. J Clin Oncol 2006;2098–104.ker MA, Zeman EM, Hirst VK, Brown JM. Metabolism of SR 4233Chinese hamster ovary cells: basis of selective hypoxic cytotoxi-. Cancer Res 1988;48:5947–52.eroute K, Wardman P, Rauth AM. Molecular mechanisms for theoxia-dependent activation of 3-amino-1,2,4-benzotriazine-1,4-xide (SR 4233). Biochem Pharmacol 1988;37:1487–95.notula V, Sarkar U, Sinha S, Gates KS. Initiation of DNA strandavage by 1,2,4-benzotriazine 1,4-dioxide antitumor agents: mech-istic insight from studies of 3-methyl-1,2,4-benzotriazine 1,4-xide. J Am Chem Soc 2009;131:1015–24.inde SS, Maroz A, Hay MP, Patterson AV, Denny WA, Anderson. Characterization of radicals formed following enzymatic reduc-n of 3-substituted analogues of the hypoxia-selective cytotoxinmino-1,2,4-benzotriazine 1,4-dioxide (tirapazamine). J Am Chemc 2010;132:2591–9.
BG, Pruijn FB, Sturman JR, et al. Selective potentiation of theoxic cytotoxicity of tirapazamine by its 1-N-oxide metabolite SR7. Cancer Res 2004;64:736–42.ox RJ, Chen S. Quinone reductase-mediated nitro-reduction: clin-l applications. Methods Enzymol 2004;382:194–221.ise CP, Abbattista M, Singleton RS, et al. The bioreductive pro-g PR-104A is activated under aerobic conditions by humano-keto reductase 1C3. Cancer Res 2010;70:1573–84.ch CJ. Unusual oxygen concentration dependence of toxicity of-4233, a hypoxic cell toxin. Cancer Res 1993;53:3992–7.ks KO, Siim BG, Pruijn FB, Wilson WR. Oxygen dependence ofmetabolic activation and cytotoxicity of tirapazamine: implica-s for extravascular transport and activity in tumors. Radiat Res4;161:656–66.rshall RS, Rauth AM. Oxygen and exposure kinetics as factorsuencing the cytotoxicity of porfiromycin, a mitomycin C analogue,Chinese hamster ovary cells. Cancer Res 1988;48:5655–9.
BG, Atwell GJ, Wilson WR. Oxygen dependence of the cytotoxi-and metabolic activation of 4-alkylamino-5-nitroquinoline biore-
ctive drugs. Br J Cancer 1994;70:596–603.ks KO, Myint H, Patterson AV, et al. Oxygen dependence and ex-vascular transport of hypoxia-activated prodrugs: comparison ofdinitrobenzamide mustard PR-104A and tirapazamine. Int J Ra-t Oncol Biol Phys 2007;69:560–71.uters BG, Brown JM. Cells at intermediate oxygen levels can bere important than the “hypoxic fraction” in determining tumor re-
nse to fractionated radiotherapy. Radiat Res 1997;147:541–50.rand RE, Olive PL. Physiologic and cytotoxic effects of tira-zamine in tumor-bearingmice. RadiatOncol Investig 1997;5:213–9.50. Senewe
acrjournals.org
Research. on July 21, 20clincancerres.aacrjournals.org Downloaded from
ks KO, Fleming Y, Siim BG, Koch CJ, Wilson WR. Extravascularfusion of tirapazamine: effect of metabolic consumption assessedng the multicellular layer model. Int J Radiat Oncol Biol Phys 1998;:641–9.le AH, Minchinton AI. Measurement of delivery and metabolism ofpazamine to tumour tissue using the multilayered cell culturedel. Cancer Chemother Pharmacol 1999;43:213–20.ks KO, Pruijn FB, Sturman JR, Denny WA, Wilson WR. Multicel-r resistance to tirapazamine is due to restricted extravascularnsport: a pharmacokinetic/pharmacodynamic study in HT29 mul-ellular layer cultures. Cancer Res 2003;63:5970–7.ks KO, Pruijn FB, Secomb TW, et al. Use of three-dimensionalue cultures to model extravascular transport and predict in vivotivity of hypoxia-targeted anticancer drugs. J Natl Cancer Inst06;98:1118–28.ijn FB, Sturman JR, Liyanage HDS, Hicks KO, Hay MP, Wilson. Extravascular transport of drugs in tumor tissue: effect of lipo-ilicity on diffusion of tirapazamine analogs in multicellular layer cul-es. J Med Chem 2005;48:1079–87.ijn FB, Patel K, Hay MP, Wilson WR, Hicks KO. Predictiontumour tissue diffusion coefficients of hypoxia-activated pro-gs from physicochemical parameters. Aust J Chem 2008;61:7–93.y MP, Gamage SA, Kovacs MS, et al. Structure-activity relation-ps of 1,2,4-benzotriazine 1,4-dioxides as hypoxia-selective analo-es of tirapazamine. J Med Chem 2003;46:169–82.y MP, Hicks KO, Pruijn FB, et al. Pharmacokinetic/pharmacody-mic model-guided identification of hypoxia-selective 1,2,4-nzotriazine 1,4-dioxides with antitumor activity: the role ofravascular transport. J Med Chem 2007;50:6392–404.y MP, Pruijn FB, Gamage SA, et al. DNA-targeted 1,2,4-nzotriazine 1,4-dioxides: potent analogues of the hypoxia-lective cytotoxin tirapazamine. J Med Chem 2004;47:475–88.y MP, Pchalek K, Pruijn FB, et al. Hypoxia-selective 3-alkyl 1,2,4-nzotriazine 1,4-dioxides: the influence of hydrogen bond donorsextravascular transport and antitumor activity. J Med Chem07;50:6654–64.y MP, Hicks KO, Pchalek K, et al. Tricyclic [1,2,4]triazine 1,4-xides as hypoxia selective cytotoxins. J Med Chem 2008;51:53–65.yd M, Hay MP, Boyd PD. Complete 1H, 13C and 15N NMR assign-nt of tirapazamine and related 1,2,4-benzotriazine N-oxides.gn Reson Chem 2006;44:948–54.halek K, Hay MP. Stille coupling reactions in the synthesis ofpoxia-selective 3-alkyl-1,2,4-benzotriazine 1,4-dioxide anticancerents. J Org Chem 2006;71:6530–5.ks KO, Pruijn FB, Baguley BC, Wilson WR. Extravascularnsport of the DNA intercalator and topoisomerase poison N-(dimethylamino)ethyl]acridine-4-carboxamide (DACA): diffusiond metabolism in multicellular layers of tumor cells. J Pharma-l Exp Ther 2001;297:1088–98.e AE, Wilson WR. Hypoxia-dependent retinal toxicity of bioreduc-anticancer prodrugs in mice. Toxicol Appl Pharmacol 2000;163:
–9.ng J, Ferry D, Hay MP, Patel R, Koch CJ, Wilson WR. Comparisonbioreductive metabolism of tirapazamine, its second generationalogue SN 30000 and hypoxia imaging agent EF5 in human tumorll lines [Abstract]. In: Proc 101st Annual Meeting Am Assoc Cancers 2010 Apr 17–21; Washington, DC. Philadelphia (PA): AACR;10, abstract 2282.ans JW, Chernikova SB, Kachnic LA, et al. Homologous re-mbination is the principal pathway for the repair of DNA damageuced by tirapazamine in mammalian cells. Cancer Res 2008;68:7–65.nan S, Rampling R, Graham MA, et al. Phase I and pharmacoki-
tic study of tirapazamine (SR 4233) administered every threeeks. Clin Cancer Res 1997;3:31–8.Clin Cancer Res; 16(20) October 15, 2010 4957
21. © 2010 American Association for Cancer
2010;16:4946-4957. Published OnlineFirst August 20, 2010.Clin Cancer Res Kevin O. Hicks, Bronwyn G. Siim, Jagdish K. Jaiswal, et al. TumorsImproved Tissue Penetration and Hypoxic Cell Killing inSN30000 and SN29751 as Tirapazamine Analogues with Pharmacokinetic/Pharmacodynamic Modeling Identifies
Updated version
10.1158/1078-0432.CCR-10-1439doi:
Access the most recent version of this article at:
Material
Supplementary
http://clincancerres.aacrjournals.org/content/suppl/2010/10/15/1078-0432.CCR-10-1439.DC1
Access the most recent supplemental material at:
Cited articles
http://clincancerres.aacrjournals.org/content/16/20/4946.full#ref-list-1
This article cites 49 articles, 17 of which you can access for free at:
Citing articles
http://clincancerres.aacrjournals.org/content/16/20/4946.full#related-urls
This article has been cited by 10 HighWire-hosted articles. Access the articles at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://clincancerres.aacrjournals.org/content/16/20/4946To request permission to re-use all or part of this article, use this link
Research. on July 21, 2021. © 2010 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Published OnlineFirst August 20, 2010; DOI: 10.1158/1078-0432.CCR-10-1439