<<

Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1 Characterization of an Abiraterone Ultraresponsive Phenotype in Castration-Resistant

2 Prostate Cancer Patient-Derived Xenografts

3

4 Hung-Ming Lam1,2*, Ryan McMullin3*†, Holly M. Nguyen1, Ilsa Coleman4, Michael Gormley5,

5 Roman Gulati4, Lisha G. Brown1, Sarah K. Holt1, Weimin Li5, Deborah S. Ricci6, Karin

6 Verstraeten7, Shibu Thomas5, Elahe A. Mostaghel4, 8, Peter S. Nelson4, 8, Robert L. Vessella1,

7 9, and Eva Corey1

8

9 Affiliation of authors: 1Department of Urology, University of Washington School of

10 Medicine, Seattle, Washington; 2State Key Laboratory of Quality Research in Chinese

11 Medicine, Macau Institute for Applied Research in Medicine and Health, Macau University of

12 Science and Technology, Macau (SAR), China; 3LabConnect, Seattle, Washington; 4Fred

13 Hutchinson Cancer Research Center, Seattle, Washington; 5Janssen Research &

14 Development, Spring House, Pennsylvania; 6Janssen Research & Development, Raritan,

15 New Jersey; 7Janssen Research & Development, Beerse, Belgium; 8Department of

16 Medicine, University of Washington, Seattle, Washington; 9Department of Veterans Affairs

17 Medical Center, Seattle, Washington

18

19 *Co-primary lead authors

20 †Current affiliation: Janssen Research & Development, Spring House, Pennsylvania 21 22

23 Running title (60 characters max): Abiraterone Response and Resistance

24

25 Keywords: , prostate cancer, xenografts, biomarkers,

26 receptor, glucocorticoid receptor, castration resistance

27

28 Grant/funding support: The work was supported by The Richard M. Lucas Foundation, the

29 Prostate Cancer Foundation, SU2C, a NIH PO1 CA085859, and the PNW Prostate Cancer

1

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

30 SPORE NIH P50 CA097186. HML is a recipient of the Young Investigator Award from the

31 Prostate Cancer Foundation, an Idea Development Award from the Department of Defense

32 (W81XWH-14-1-0271), and a FHCRC/UW Cancer Consortium New Investigator Grant of

33 NIH P30 CA015704. Janssen Research & Development provided funding support for some

34 of the molecular analyses reported herein.

35

36

37 Corresponding author:

38 Eva Corey, PhD 39 Department of Urology 40 Mailstop 356510 41 University of Washington 42 Seattle, WA 98195 43 Phone: +1 206 543 1461 44 Fax: +1 206 543 1146 45 Email: [email protected] 46

47 Conflict of interest:

48 Peter S. Nelson has served on advisory boards for Janssen and received compensation.

49 Ryan McMullin, Michael Gormley, Weimin Li, Deborah S. Ricci, Karin Verstraeten, and

50 Shibu Thomas are employees of Janssen Research & Development and own stock in

51 Johnson & Johnson.

52

53

54 Word count

55 Text (excluding references, tables/figures): 4328

56 Figures: 5

57

2

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

58 Translational Relevance 59 60 Abiraterone acetate (AA) improves survival in patients with metastatic castration-resistant

61 prostate cancer (mCRPC); however, not all tumors respond and responding tumors

62 eventually develop resistance. Currently there is no information available regarding how to

63 stratify patients for durable AA therapy, and the mechanisms underlying AA resistance are

64 diverse. We used patient-derived xenograft models that recapitulated the diverse clinical

65 response of CRPC to AA and identified a molecular signature of secreted proteins

66 associated with the AA ultraresponsive phenotype. The signature will provide the much

67 needed information on noninvasive biomarker development to select AA-responsive

68 patients. Upon resistance, our results suggested reduced (AR) signaling

69 and sustainably low nuclear glucocorticoid receptor (nGR) localization in the AA

70 ultraresponders. In contrast, sustained AR signaling and increased nGR localization was

71 observed in the intermediate and minimal responders. Further inhibition along the AR/GR

72 signaling axis may be effective in AA-resistant patients who are intermediate or minimal

73 responders.

74

75

76

3

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

77 Abstract

78 Purpose: To identify the molecular signature associated with abiraterone acetate (AA)

79 response and mechanisms underlying AA resistance in castration-resistant prostate cancer

80 patient-derived xenografts (PDXs).

81 Experimental Design: SCID mice bearing LuCaP 136CR, 77CR, 96CR, and 35CR PDXs

82 were treated with AA. Tumor volume and prostate-specific antigen were monitored, and

83 tumors were harvested 7 days post-treatment or at end of study for gene expression and

84 immunohistochemical studies.

85 Results: Three phenotypic groups were observed based on AA response. An

86 ultraresponsive phenotype was identified in LuCaP 136CR with significant inhibition of tumor

87 progression and increased survival, intermediate responders LuCaP 77CR and LuCaP

88 96CR with a modest tumor inhibition and survival benefit, and LuCaP 35CR with minimal

89 tumor inhibition and no survival benefit upon AA treatment. We identified a molecular

90 signature of secreted proteins associated with the AA ultraresponsive phenotype. Upon

91 resistance, AA ultraresponder LuCaP 136CR displayed reduced androgen receptor (AR)

92 signaling and sustainably low nuclear glucocorticoid receptor (nGR) localization,

93 accompanied by alteration and epithelial-mesenchymal transition

94 phenotype enrichment with increased expression of NF-κB-regulated genes; intermediate

95 and minimal responders maintained sustained AR signaling and increased tumoral nGR

96 localization.

97 Conclusions: We identified a molecular signature of secreted proteins associated with AA

98 ultraresponsiveness and sustained AR/GR signaling upon AA resistance in intermediate or

99 minimal responders. These data will inform development of noninvasive biomarkers

100 predicting AA response, and suggest further inhibition along the AR/GR signaling axis may

101 be effective only in AA-resistant patients who are intermediate or minimal responders. These

102 findings require verification in prospective clinical trials.

4

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

103 Introduction

104 Androgen-deprivation therapy (ADT) has been the mainstay therapy for patients with

105 advanced prostate cancer (1). Abiraterone acetate (AA), the prodrug of abiraterone, is a

106 specific CYP17A1 inhibitor that blocks androgen biosynthesis, resulting in effective reduction

107 of serum and intratumoral (2-4). AA was the first second-generation ADT shown

108 to improve survival in metastatic castration-resistant prostate cancer (mCRPC) patients (5-

109 9). While dramatic decline in prostate specific antigen (PSA) was achieved in some patients,

110 others exhibited a subtle PSA response or de novo resistance, and disease progression is

111 universal (1, 5, 9).

112 Predictive biomarkers that distinguish ultraresponders from intermediate or minimal

113 responders to AA are critically needed. Early attempts using circulating tumor cells (CTCs)

114 showed that TMPRSS2-ERG fusion did not predict the response to AA in CRPC patients

115 (10). However, Antonarakis et al. recently showed that CRPC patients with positive

116 androgen receptor transcript variant (ARv7) in their pretreatment CTC did not demonstrate

117 PSA decline and 68% of a small cohort of patients with negative ARv7 demonstrated > 50%

118 PSA decline after receiving AA (11), suggesting that the detection of positive ARv7 in CTCs

119 may predict AA sensitivity.

120 De novo and acquired resistance to AA is emerging clinically, and there are

121 preclinical and clinical efforts to investigate the mechanisms of resistance. In preclinical

122 studies, resistance to AA was associated with an induction of full-length AR , ARv7, and

123 CYP17A1 (12). In clinical studies, the presence of ARv7 in CTCs was associated with

124 resistance to AA and shorter overall survival (11). In addition, acquired resistance to AA has

125 been associated with the emergence of AR mutations that have been reported in up to 20%

126 of patients who progressed (13-15). Recently, upregulation of glucocorticoid receptor (GR)

127 has been shown to be a possible bypass mechanism to ADT, and CRPC patients with

128 positive GR in their bone marrow biopsies were less likely to have a durable response to

129 , another second-generation ADT (16).

5

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

130 Currently, there is little information about biomarkers to identify patients who will

131 durably respond to AA, and the mechanisms of resistance are diverse. In the present study

132 we evaluated the AA response in a panel of LuCaP CRPC patient-derived xenografts

133 (PDXs) that displayed differential responsiveness to AA and identified a molecular signature

134 associated with AA ultraresponsiveness. We also provided evidence to support diverse

135 resistance mechanisms upon AA treatment. This study highlights potential noninvasive

136 biomarkers that may be used to select patients for durable AA therapy, and potential

137 targeting of the epithelial-mesenchymal transition (EMT)/nuclear factor κB (NF-κB) pathway

138 in AA ultraresponsive or AR/GR pathways in AA intermediate- or minimally responsive

139 CRPC.

140

141 Materials and Methods

142 Prostate cancer PDX models

143 Animal procedures were carried out in accordance with National Institutes of Health

144 guidelines and upon University of Washington Institutional Animal Care and Use Committee

145 approval. Four different LuCaP human CRPC PDXs (LuCaP 136CR, LuCaP 77CR, LuCaP

146 96CR, and LuCaP 35CR) were used. All four PDXs express wild-type AR but exhibit

147 differential expression of PSA, PTEN, and ERG (corresponding patient information is

148 summarized in Supplementary Table S1). Two additional PDX models (LuCaP 70CR and

149 LuCaP 86.2CR) were used for survival analysis upon AA treatment and assessment of gene

150 signature.

151 Intact male CB-17 SCID mice (aged ~6 weeks; Charles River Laboratories, San

152 Diego, CA) were implanted subcutaneously with tumor bits of LuCaP 136 or LuCaP 77. Mice

153 were castrated when tumor volume was ≥100 mm3. When tumor regrew to 1.5-fold the

154 original volume, tumors were referred to as LuCaP 136CR or LuCaP 77CR (Fig. 1). LuCaP

155 96CR and LuCaP 35CR are castration-resistant PDXs that are propagated in castrated male

156 mice. Castrated male CB-17 SCID mice were implanted subcutaneously with LuCaP 96CR

157 or LuCaP 35CR tumor bits, and enrolled when tumor volume reached ≥100 mm3 (Fig. 1).

6

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

158 Upon enrollment, mice were randomized to vehicle (20% HPbCD/0.37N HCl/PBS) or AA

159 treatment groups (0.5 mmol/kg; Janssen Pharmaceutical Companies, Beerse, Belgium).

160 Animals were treated by oral gavage on a weekly schedule of 5 days on, 2 days off. Tumor

161 volume and body weight were measured twice weekly, and blood samples were drawn

162 weekly for PSA measurements using AxSym Total PSA Assay (Abbott Laboratories, Abbott

163 Park, IL). Five animals in each group were sacrificed 7 days after the initiation of treatment

164 (D7) and the remaining animals were followed and sacrificed when tumors exceeded 1,000

165 mm3 (end of study, EOS) or sacrificed if animals became compromised. At sacrifice (D7 or

166 EOS), half of the tumor was harvested for paraffin embedding and half was frozen for

167 subsequent analyses. Treatment schemes for LuCaP 70CR and LuCaP 86.2CR are

168 illustrated in Supplementary Fig. S1.

169

170 Intratumoral androgen measurement

171 Intratumoral androgen levels were measured using mass spectrometry as described

172 previously (17, 18). Vehicle-treated tumors and AA-resistant tumors harvested at EOS were

173 used for these analyses.

174

175 Immunohistochemistry

176 Hematoxylin and eosin staining of paraffin-embedded tissues was used to identify

177 viable tumor cells in the tissues. Two cores (5–8 tumors per group) were punched and

178 placed in tissue microarrays. The tissue microarray slides were stained for AR (F39.4.1,

179 1:100, BioGenex, Fremont, CA), GR (D6H2L, 1:100, Cell Signaling, Danvers, MA),

180 chromogranin A (DAK-A3, 1:100, DAKO, Carpinteria, CA), and synaptophysin (D-4, 1:200,

181 Santa Cruz Biotechnology, Dallas, TX) using standard procedures as described previously

182 (19-21). All evaluations were performed in a blinded fashion and a quasi-continuous

183 immunohistochemical (IHC) score was calculated by multiplying each intensity level (0 for no

184 stain, 1 for faint stain, and 2 for intense stain) by the corresponding percentage of cells (0–

7

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

185 100%) at the corresponding intensity and totaling the results. IHC scores ranged from 0 (no

186 staining in any cell) to 200 (intense staining in 100% of the cells).

187

188 RNA extraction

189 Frozen pieces of tumor were embedded in Optimal Cutting Temperature Compound

190 and 5-µm sections were stained with hematoxylin and eosin. Areas of viable tumor cells

191 were identified and macro-dissected for RNA extraction using a standard procedure with

192 RNA STAT 60 (Tel-Test, Friendswood, TX). RNA was then purified using RNeasy Mini kit

193 utilizing the optional DNase digestion in solution prior to purification (Qiagen, Hilden,

194 Germany) for subsequent gene expression analyses. RNA integrity number was determined

195 using the Agilent Bioanalyzer system (Agilent, Santa Clara, CA).

196

197 Gene expression analyses

198 For Affymetrix microarray analyses, biotin-labeled, amplified RNA (aRNA) was

199 synthesized from 200 ng total RNA using the 3ʹ IVT Express Kit (Affymetrix, Santa Clara,

200 CA). The aRNA was purified using Agencourt RNAClean XP beads (Beckman Coulter Inc.,

201 Brea, CA) on the BioMek FX Workstation (Beckman Coulter Inc.). Biotin-labeled aRNA was

202 fragmented using the 3ʹ IVT Express Kit. A total of 4.5 µg fragmented biotin-labeled aRNA

203 was hybridized on an HT Human Genome (HG)-U219 96-array plate. The plate was

204 washed, stained, and scanned with the GeneTitan Instrument. All reagents were from

205 Affymetrix. Gene expression microarray data were normalized to minimize systematic

206 technical variation using robust multichip average (22) and represented in the log2 scale.

207 Data were filtered to remove probes with mean signal intensities below the 25th percentile of

208 signal intensities for all probes. The Significance Analysis of Microarrays (SAM) program

209 (http://www-stat.stanford.edu/~tibs/SAM/) (23) was used to analyze expression differences

210 between groups using unpaired, two-sample t tests and controlled for multiple testing by

211 estimating q-values using the false discovery rate method. Gene family was manually

212 curated from Gene Ontology and Uniprot databases. The AR score was determined by the

8

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

213 expression of a 21-gene signature and calculated as described previously (24). Microarray

214 data are deposited in the Gene Expression Omnibus database under the accession number

215 GSE85672.

216

217 Ingenuity pathway analysis

218 The differentially expressed genes between vehicle-treated and AA-resistant tumors

219 at the EOS from each of the four LuCaP models were imported into Ingenuity Pathway

220 Analysis (Ingenuity Systems; https://www.ingenuity.com) to identify molecular and cellular

221 functions and regulator effect network involved in AA resistance as previously described (25,

222 26).

223

224 Gene set enrichment analysis

225 Gene set enrichment analysis (GSEA) (27) was conducted to evaluate enrichment of

226 differential expression patterns in canonical signaling pathways (Reactome) (28) or

227 predefined gene signatures of prostate cancer core gene expression modules representing

228 distinct biological programs (Compendia Bioscience, Ann Arbor, MI) and annotated

229 signatures associated with EMT, AR activity, GR activity, and AA response.

230

231 Quantitative real-time polymerase chain reaction

232 Total RNA was reverse-transcribed to cDNA, and real-time polymerase chain

233 reaction (PCR) was carried out as described previously (29). Species-specific primer

234 sequences are presented in Supplementary Table S2. PCR reactions with SYBR GreenER

235 PCR Master-Mix (Invitrogen, Carlsbad, CA) were monitored with the 7900HT Fast Real-time

236 PCR System (Applied Biosystems, Foster City, CA). Individual mRNA levels were

237 normalized to human RPL13a.

238

239 AR sequencing

9

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

240 Genomic DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen) and PCR

241 amplified using primer AR_exon8_c1-589_F: ATTGCGAGAGAGCTGCATCA and

242 AR_exon8_c1-589_R: TGCTTGTTTTTGTTTTGATTTCC. Sanger Sequencing was

243 performed using BigDye Terminator v3.1 Cycle Sequencing Kit (# 4337454, Life

244 Technologies, Carlsbad, CA) according to manufacturer’s recommendations. Sequences

245 were aligned to human AR genomic sequence NC_000023.11 and mRNA RefSeq NM_0044

246 using Sequencher Software (version 5.1, Gene Codes, Ann Arbor, MI). Mutations were

247 verified using The Androgen Receptor Gene Mutations Database (McGill University).

248

249 Statistical analyses

250 Survival was determined using Kaplan-Meier estimation of time from start of

251 treatment (vehicle or AA) to sacrifice and compared by log-rank (Mantel-Cox) test. Statistical

252 analyses of tumor volume and PSA responses were performed as described previously (19).

253 Briefly, longitudinal tumor measurements and PSA serum levels were log-transformed and

254 modeled using linear mixed models conditional on the treatment group with random effects

255 for each animal. Following standard diagnostic assessment of model fit, we simulated 1,000

256 datasets from each fitted model, calculated the empirical mean and 95% confidence limits at

257 each time point, and refitted the models to these datasets. The final results represented

258 means and 95% confidence limits of 1,000 bootstrap replicates. In addition, the rate of

259 change in serum PSA and tumor volume upon AA treatment was tested using estimated

260 fixed effects for each LuCaP line. Student’s t test and Pearson correlation coefficients were

261 used for statistical comparisons between the groups in the intratumoral androgen

262 measurements, gene expression analysis, and IHC analyses. For GSEA, a gene set that

263 displayed false discovery rate <25% is considered significantly enriched.

264 265 Results

266 Heterogeneous AA responses and identification of an AA ultraresponder in LuCaP

267 PDX models

10

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

268 CRPC was developed using four different models of LuCaP PDXs (Fig. 1). AA

269 treatment improved survival and inhibited tumor progression in three of the four models. In

270 mice bearing LuCaP 136CR tumors, survival was substantially improved in AA-treated

271 compared with vehicle-treated mice (P < 0.001) and the median survival improved from 6.8

272 weeks (vehicle) to 21.8 weeks (AA; denoted as AA ultraresponder; 220% gain in survival;

273 Fig. 2A). AA treatment resulted in statistically significant but modest improvement in survival

274 in mice bearing LuCaP 77CR (P = 0.05) and LuCaP 96CR (P = 0.02) — both denoted as

275 intermediate responders (36–74% gain in survival; Fig. 2A). AA did not significantly extend

276 survival in mice bearing LuCaP 35CR (12% gain in survival; P = 0.52; denoted as minimal

277 responder; Fig. 2A).

278 Both the AA ultraresponder LuCaP 136CR and intermediate responders LuCaP

279 77CR and LuCaP 96CR, but not the minimal responder LuCaP 35CR, demonstrated

280 significantly delayed tumor and PSA progression (except for LuCaP 136CR, which has

281 undetectable levels of serum PSA; Fig. 2, panels B and C), followed by both tumor and PSA

282 recurrence. These results suggested the PDX models recapitulated clinical AA response

283 phenotypes comprising ultraresponders with inhibition of tumor progression and a significant

284 extension of survival followed by tumor recurrence, and intermediate and minimal

285 responders with brief or limited AA effect on tumor growth inhibition followed by disease

286 progression.

287

288 Gene expression associated with LuCaP 136CR ultraresponsiveness to AA

289 To identify the gene expression profiles associated with AA ultraresponsiveness, we

290 conducted global transcriptome analyses of the PDX lines. We identified 531 differentially

291 expressed genes between the AA ultraresponder LuCaP 136CR versus the intermediate

292 responder LuCaP 96CR and minimal responder LuCaP 35CR at D7 (P < 0.0001, fold

293 change ≥3; Fig. 3A). LuCaP 77CR D7 tumors were not included in the global analysis

294 because the specimens were not available, but their EOS tumors were included in the gene

295 expression validation. Of the 156 genes that were successfully mapped into known gene

11

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

296 families, 68 (44%) were secreted proteins (Supplementary Fig. S2A). We observed that the

297 differential expression of these 68 secretory proteins in LuCaP 136CR were consistent

298 between early time point (D7; Supplementary Fig. S2B) and EOS (Supplementary Fig. S2C),

299 suggesting the expression of these markers was not dependent on age of mice or tumor

300 size. We then selected the top 10 upregulated and downregulated genes of secreted

301 proteins (total 20 genes) in the AA ultraresponder LuCaP 136CR compared with the

302 intermediate and minimal responders for qPCR validation (Fig. 3B and Supplementary Fig.

303 S3). Primers for 18 genes were available, and qPCR confirmed all of the eight upregulated

304 genes (CEL, ARMCX1, TNC, BMP7, IER3, FSTL5, SNTB1, and FBN2; Fig. 3C) and 10

305 downregulated genes (IL17RB, GDF15, ST6GAL1, SPOCK1, MSMB, INHBB, MINPP1,

306 GALS3BP, C15orf48, and PLA2G2A; Supplementary Fig. S4). However, the downregulated

307 genes showed more variable expression in the intermediate (LuCaP 77CR, LuCaP 96CR)

308 and minimal (LuCaP 35CR) responders and therefore were not included in the development

309 of a stringent gene signature for AA ultraresponsiveness.

310 We next validated the highly consistent eight-gene signature that was upregulated in

311 the AA ultraresponder LuCaP 136CR in an independent cohort of six LuCaP models that

312 displayed different responses to AA. As expected, the signature positively correlated with the

313 percentage gained survival on AA (R = 0.95, P = 0.0002; Fig. 3D and E), supporting the

314 potential of this eight-gene signature in predicting AA ultraresponsiveness.

315

316 Mechanisms associated with the acquired resistance of individual AA-responsive

317 phenotypes

318 To identify response and resistance mechanisms specific to different AA response

319 phenotypes, we conducted global transcriptome analyses on the AA-treated (D7) and AA-

320 resistant (EOS) tumors. Interestingly, upon AA resistance, a distinct set of genes was

321 differentially expressed in each of the four models (vehicle vs. AA, P < 0.01, fold change ≥2),

322 and there was virtually no overlap of genes between ultraresponders and

323 intermediate/minimal responders or within the intermediate and minimal responders (Fig. 4A

12

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

324 and Supplementary Table S5), suggesting that the AA-induced resistance mechanisms are

325 largely diverse. Next, we conducted Ingenuity Pathway Analysis to identify molecular and

326 cellular function involved in the AA resistance in individual models. For both ultraresponder

327 LuCaP 136CR and the intermediate responder LuCaP 77CR, cell growth and proliferation

328 represented 40% to 45% of genes that were associated with AA resistance. In LuCaP 96CR,

329 a majority of AA differentially expressed genes were related to cell morphology (30%),

330 whereas in the minimal responder, AA differentially expressed genes were principally

331 mapped to cell-to-cell signaling (20%) or cellular death and survival (20%) (Fig. 4B).

332 GSEA analysis showed that AA treatment of LuCaP 77CR was negatively associated

333 with signatures of cell growth and androgen-regulated genes upon resistance at EOS

334 (Supplementary Fig. S5). Similarly, AA treatment of LuCaP 96CR was negatively associated

335 with a cell cycle–associated signature that was previously reported to be decreased in a cell

336 line–derived xenograft model of AA resistance (Supplementary Fig. S4) (30). Interestingly, in

337 the AA ultraresponder LuCaP 136CR, we identified steroid metabolism as the top altered

338 regulator effect network upon AA resistance (Fig. 4C), which, together with the high basal

339 expression of the esterase CEL, implies that alterations in the steroid availability

340 and usage may contribute to the development of AA resistance in this model. Importantly,

341 GSEA analysis showed that AA treatment of LuCaP 136CR was initially negatively

342 associated with signatures of proliferation, cell growth, and a selected AR transcriptional

343 program at D7, and this negative proliferation signature persisted but with fewer genes

344 represented at the leading edge at EOS (Supplementary Fig. S5). Despite the negative

345 association with the specific proliferation markers, LuCaP 136CR acquired AA resistance

346 that was enriched with genes associated with NF-κB transcriptional activity, EMT,

347 extracellular matrix, and prostate basal cells (Supplementary Fig. S5). These results suggest

348 the diversity of resistance mechanisms to AA and specifically indicate potential mechanisms

349 that drive AR-independent resistance in the AA ultraresponsive phenotype.

350

13

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

351 Low basal AR signaling and a further reduction of androgen signaling upon

352 resistance in the AA ultraresponder LuCaP 136CR

353 We examined the AR signaling axis to gain insight into its role in AA resistance and

354 tumor progression. Previous reports showed that AA treatment elevated serum levels of

355 progesterone and other upstream that activated mutant AR (e.g., L701H and

356 T878A) leading to AA resistance (31-34). To elucidate whether AR mutation was involved in

357 the differential AA responsiveness observed in our models, we sequenced the ligand-binding

358 domain of AR and detected no mutation in the AA-treated LuCaP PDXs (data not shown),

359 suggesting that the differential AA responsiveness was not due to AR mutation.

360 We next conducted targeted analysis on intratumoral androgens and androgen

361 signaling pathways in AA-resistant tumors. We used a sensitive liquid chromatography–

362 mass spectrometry method to detect intratumoral androgens that are sensitive to AA

363 inhibition. In the ultraresponder LuCaP 136CR, AA treatment significantly reduced

364 intratumoral levels of (P = 0.009), (P = 0.04),

365 (P = 0.03; Fig. 5A), and (P = 0.04; Supplementary Fig. S6).

366 Interestingly, LuCaP 136CR demonstrated the lowest basal AR signaling among the LuCaP

367 lines tested, depicted by a low AR activity score (Fig. 5B) and a low AR signature score (Fig.

368 5C). Upon AA resistance, the decrease in intratumoral androgens was accompanied by a

369 general downregulation of steroidogenic enzymes, including LDLR (P = 0.004), STARD4 (P

370 = 0.005), and DUSP1 (P = 0.01) (Supplementary Table S3) (12), a further downregulation of

371 AR activity (Fig. 5B), and a reduced AR signature score (Fig. 5C). These results suggested

372 reduced AR signaling in the AA ultraresponder LuCaP 136CR upon resistance.

373 In contrast, despite decreasing testosterone in the intermediate responders LuCaP

374 77CR (P = 0.03) and LuCaP 96CR (P = 0.02) upon AA treatment, high variability in

375 dihydrotestosterone levels was observed in LuCaP 77CR and a statistically insignificant

376 reduction was observed in LuCaP 96CR (P = 0.11) (Fig. 5A). Upstream steroids, including

377 pregnenolone (P = 0.02) and (DHEA; P = 0.056), were increased in

378 the intermediate responder LuCaP 77CR upon AA resistance (Supplementary Fig. S5),

14

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

379 whereas progesterone was decreased in the intermediate responder LuCaP 96CR (P =

380 0.02; Supplementary Fig. S6). Consistent with the sustained level of intratumoral androgens,

381 no reduction in the enrichment in AR-responsive genes (Fig. 5B) and AR signature (Fig. 5C)

382 was detected upon AA resistance in the intermediate responders LuCaP 77CR and LuCaP

383 96CR. Similarly, in the AA minimal responder LuCaP 35CR, AA treatment showed an initial

384 negative association with GSEA signatures of AR- and GR-regulated genes at D7

385 (Supplementary Fig. S5) and a reduction in our selected AR signature (Fig. 5C). However,

386 the negative association was not observed upon AA resistance at EOS (Supplementary Fig.

387 S5), and the AA-resistant tumor demonstrated a persistent expression of steroidogenic

388 enzymes (Supplementary Table S3), AR-responsive genes (Fig. 5B), and AR signature (Fig.

389 5C). Due to the limited number of LuCaP 35CR AA-resistant tumors available, statistically

390 significant change in the intratumoral androgens was not observed in these tumors upon AA

391 resistance (Fig. 5A). Collectively, these results pointed to sustained AR signaling in the AA

392 intermediate and minimal responders upon resistance. In all models, we also tested whether

393 the AA-resistant tumors acquired a neuroendocrine phenotype. Our results showed that both

394 neuroendocrine markers (chromogranin A and synaptophysin) were absent or minimally

395 expressed (<0.1% in LuCaP 77CR) in the vehicle-treated tumors and the expression did not

396 change upon AA resistance (data not shown).

397 Finally, we questioned whether AR and GR levels in the tumor may contribute to the

398 downregulation of AR signaling in the AA-resistant tumors in the ultraresponder LuCaP

399 136CR and the sustained AR signaling in the intermediate or minimal responders. In the

400 ultraresponder LuCaP 136CR, the gene expression of AR and ARv7 was increased upon

401 castration (Supplementary Table S4) but remained unchanged upon further androgen

402 ablation by AA (Fig. 5D), and the nuclear AR and GR localization was not altered upon AA

403 resistance (Fig. 5, panels E and F). The nuclear GR level remained low even upon AA

404 resistance in the ultraresponder LuCaP 136CR (Fig. 5F). In the intermediate and minimal

405 responders, increased expression of AR and its variants was observed upon castration in

406 LuCaP 77CR (Supplementary Table S4), but the expression of AR and ARv7 generally

15

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

407 remained unchanged upon AA resistance except for LuCaP 96CR (Fig. 5D). Nuclear

408 localization of AR remained high (i.e., H-score > 100) in the intermediate and minimal

409 responders although a slight decrease in nuclear AR localization for LuCaP 77CR was

410 observed upon AA resistance (Fig. 5, panels D and F). Collectively, these findings

411 suggested active AR signaling in these AA-resistant tumors. Importantly, we observed a high

412 basal level of nuclear GR in the AA minimal responder LuCaP 35CR (Fig. 5F) and a

413 consistent upregulation of both GR gene expression (NR3C1, except for LuCaP 35CR) and

414 nuclear localization for all intermediate and minimal responders (Fig. 5, panels D and F).

415 These GR results may suggest that high basal nuclear GR localization is associated with AA

416 minimal responsiveness, and that an increase in nuclear GR upon AA treatment is

417 associated with rapid, acquired resistance. In summary, upon AA resistance, the

418 ultraresponder LuCaP 136CR displayed lower intratumoral androgens and AR signaling

419 accompanied by sustainably low nuclear GR localization. In contrast, the intermediate and

420 minimal responders demonstrated a slight decrease in intratumoral androgens and

421 sustained AR signaling associated with an increase in nuclear GR localization.

422

423 Discussion

424 AA is effective in a subset of patients, but responding tumors eventually develop

425 resistance. We used PDX models that recapitulated the diverse clinical responses of CRPC

426 to AA and identified heterogeneous response phenotypes, including ultraresponsive,

427 intermediate, and minimal. The ultraresponsive phenotype represents not only AA sensitivity

428 but also durability. We report for the first time that the AA ultraresponsive phenotype is

429 represented by a molecular signature of secreted proteins and biochemical features,

430 including low basal AR signaling and a low basal nuclear GR level, which is insensitive to

431 AA-induced upregulation.

432 Mechanisms underlying acquired resistance to AA are diverse and have not yet been

433 fully identified. GR was shown to compensate for reduced AR activity through activation of

434 overlapping target genes (35). High GR expression was associated with enzalutamide

16

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

435 insensitivity (16), and preliminary results of the COU-AA-203 study demonstrated that high

436 GR may predict low AA sensitivity (36). Our results provided novel information to highlight

437 the role of GR in response to AA: (a) a low level of nuclear GR, and sustainably low GR on

438 AA therapy, was predictive of durable AA inhibition; (b) low to intermediate levels of GR,

439 despite initial response, and increase in nuclear GR was associated with rapid, acquired

440 resistance to AA; and (c) a high basal level of GR was associated with de novo

441 resistance/minimal responsiveness. Notably, while we observed a concordant increase in

442 both GR transcript and protein expression levels in some models, discordance was present

443 in others. This result indicates that GR transcripts may not ideally reflect the protein level,

444 especially the nuclear protein level indicative of active GR signaling. Retrospective clinical

445 studies investigating response and resistance patterns have suggested cross-

446 response/resistance between enzalutamide and AA (37-44). However, whether a

447 sustainably low level of GR will lead to a durable response to either AA or enzalutamide in

448 patients, and whether an increase in GR is attributable to rapid AA resistance, requires

449 clinical confirmation.

450 Copy number gain of AR and CYP17A1 has been shown to predict shorter

451 progression-free survival with AA treatment (45). Our results supported, at a gene

452 expression level, that the intermediate responders LuCaP 77CR/LuCaP 96CR and the

453 minimal responder LuCaP 35CR demonstrated a higher AR level and enhanced androgen

454 signaling when compared with the ultraresponder LuCaP 136CR. On the other hand, other

455 preliminary studies on gene expression using pretreatment primary prostate cancer samples

456 reported a significant association between proliferation-associated genes, androgen-

457 regulated genes, and CYP17 cofactors with longer radiographic progression-free survival of

458 patients receiving AA (46).

459 In our studies, the ultraresponsive phenotype demonstrated reduced AR signaling

460 upon AA resistance, indicating an emergence of an AR-independent pathway to sustain

461 survival. Upon AA resistance, the ultraresponders presented an enrichment of genes

462 associated with EMT, prostate basal-type cells, and NF-κB activity. This is consistent with a

17

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

463 previous report showing an association between EMT induction and the emergence of

464 prostate cancer stem-like cells (CSC)-like phenotype following androgen deprivation (47).

465 Additionally, activation of the NF-κB pathway is involved in the induction and maintenance of

466 EMT (48, 49) and CSC-like characteristics in prostate cancer (50-52). These characteristics

467 are concordant with the results of a preclinical study identifying a progenitor-like cell

468 population with increased NF-κB activity upon resistance to androgen depletion (53) and

469 reduced AR signaling upon increased NF-κB activity in prostate cancer (54). A recent report

470 on NF-κB as a potential resistance mechanism for enzalutamide independent of ARv7 may

471 provide another cross-resistance mechanism for AA (55).

472 In view of the heterogeneity of patients’ responses to AA therapy, identification of

473 biomarkers of responses has important implications for treatment selection in the context of

474 precision oncology. The preclinical eight-gene molecular signature of secreted proteins

475 associated with AA durable response that we identified can potentially be developed into a

476 fast, noninvasive test to predict AA response. However, our results at this point are limited to

477 the preclinical setting and by the number of PDX models representing each response

478 phenotype. Validation in prospective clinical studies is needed to support translational value

479 of this signature.

480 Collectively, the diverse resistant phenotypes associated with differential AA

481 responses highlighted the need for a tailored next line of therapy. The resistance in the AA

482 ultraresponsive phenotype was represented by low intratumoral androgens and AR signaling

483 accompanied by a sustainably low nuclear GR localization, and alteration in gene

484 expression associated with NF-κB activity and a EMT/basal cell phenotype. In contrast,

485 resistance in the intermediate and minimally responding phenotypes demonstrated

486 sustained AR signaling and increased nuclear GR localization. Novel treatments may be

487 explored to target NF-κB activity with a rationale to prevent or revert an EMT basal cell

488 phenotype in the ultraresponders and to target sustained AR/GR signaling in the

489 intermediate or minimal responders upon AA resistance.

490

18

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

491 Acknowledgments 492 493 Editorial assistance was provided by Ira Mills, PhD, of PAREXEL and funded by Janssen

494 Global Services, LLC. We thank Bryce Lakely and Daniel Sondheim for their excellent

495 technical assistance.

19

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

496 References 497 498 1. Crawford ED. Understanding the epidemiology, natural history, and key pathways

499 involved in prostate cancer. Urology 2009;73:S4-10.

500 2. Attard G, Belldegrun AS, de Bono JS. Selective blockade of androgenic steroid

501 synthesis by novel lyase inhibitors as a therapeutic strategy for treating metastatic

502 prostate cancer. BJU Int 2005;96:1241-6.

503 3. Barrie SE, Potter GA, Goddard PM, Haynes BP, Dowsett M, Jarman M. Pharmacology

504 of novel steroidal inhibitors of cytochrome P450(17) alpha (17 alpha-hydroxylase/C17-

505 20 lyase). J Steroid Biochem Mol Biol 1994;50:267-73.

506 4. O'Donnell A, Judson I, Dowsett M, Raynaud F, Dearnaley D, Mason M, et al. Hormonal

507 impact of the 17alpha-hydroxylase/C(17,20)-lyase inhibitor abiraterone acetate

508 (CB7630) in patients with prostate cancer. Br J Cancer 2004;90:2317-25.

509 5. de Bono JS, Logothetis CJ, Molina A, Fizazi K, North S, Chu L, et al. Abiraterone and

510 increased survival in metastatic prostate cancer. N Engl J Med 2011;364:1995-2005.

511 6. Fizazi K, Scher HI, Molina A, Logothetis CJ, Chi KN, Jones RJ, et al. Abiraterone

512 acetate for treatment of metastatic castration-resistant prostate cancer: final overall

513 survival analysis of the COU-AA-301 randomised, double-blind, placebo-controlled

514 phase 3 study. Lancet Oncol 2012;13:983-92.

515 7. Ryan CJ, Smith MR, de Bono JS, Molina A, Logothetis CJ, de Souza P, et al.

516 Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J

517 Med 2013;368:138-48.

518 8. Rathkopf DE, Smith MR, de Bono JS, Logothetis CJ, Shore ND, de SP, et al. Updated

519 interim efficacy analysis and long-term safety of abiraterone acetate in metastatic

20

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

520 castration-resistant prostate cancer patients without prior chemotherapy (COU-AA-

521 302). Eur Urol 2014;66:815-25.

522 9. Ryan CJ, Smith MR, Fizazi K, Saad F, Mulders PF, Sternberg CN, et al. Abiraterone

523 acetate plus prednisone versus placebo plus prednisone in chemotherapy-naive men

524 with metastatic castration-resistant prostate cancer (COU-AA-302): final overall

525 survival analysis of a randomised, double-blind, placebo-controlled phase 3 study.

526 Lancet Oncol 2015;16:152-60.

527 10. Danila DC, Anand A, Sung CC, Heller G, Leversha MA, Cao L, et al. TMPRSS2-ERG

528 status in circulating tumor cells as a predictive biomarker of sensitivity in castration-

529 resistant prostate cancer patients treated with abiraterone acetate. Eur Urol

530 2011;60:897-904.

531 11. Antonarakis ES, Lu C, Wang H, Luber B, Nakazawa M, Roeser JC, et al. AR-V7 and

532 resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med

533 2014;371:1028-38.

534 12. Mostaghel EA, Marck BT, Plymate SR, Vessella RL, Balk S, Matsumoto AM, et al.

535 Resistance to CYP17A1 inhibition with abiraterone in castration-resistant prostate

536 cancer: induction of steroidogenesis and androgen receptor splice variants. Clin

537 Cancer Res 2011;17:5913-25.

538 13. Carreira S, Romanel A, Goodall J, Grist E, Ferraldeschi R, Miranda S, et al. Tumor

539 clone dynamics in lethal prostate cancer. Sci Transl Med 2014;6:254ra125.

540 14. Chen EJ, Sowalsky AG, Gao S, Cai C, Voznesensky O, Schaefer R, et al. Abiraterone

541 treatment in castration-resistant prostate cancer selects for progesterone responsive

542 mutant androgen receptors. Clin Cancer Res 2015;21:1273-80.

21

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

543 15. Romanel A, Gasi TD, Conteduca V, Jayaram A, Casiraghi N, Wetterskog D, et al.

544 Plasma AR and abiraterone-resistant prostate cancer. Sci Transl Med 2015;7:312re10.

545 16. Arora VK, Schenkein E, Murali R, Subudhi SK, Wongvipat J, Balbas MD, et al.

546 Glucocorticoid receptor confers resistance to by bypassing androgen

547 receptor blockade. Cell 2013;155:1309-22.

548 17. Montgomery B, Nelson PS, Vessella R, Kalhorn T, Hess D, Corey E.

549 suppresses tissue androgens and prostate cancer growth in castration resistant

550 prostate cancer. BMC Cancer 2010;10:244.

551 18. Montgomery RB, Mostaghel EA, Vessella R, Hess DL, Kalhorn TF, Higano CS, et al.

552 Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for

553 castration-resistant tumor growth. Cancer Res 2008;68:4447-54.

554 19. Nguyen HM, Ruppender N, Zhang X, Brown LG, Gross TS, Morrissey C, et al.

555 Cabozantinib inhibits growth of androgen-sensitive and castration-resistant prostate

556 cancer and affects bone remodeling. PLoS One 2013;8:e78881.

557 20. Morrissey C, Brown LG, Pitts TE, Vessella RL, Corey E. Bone morphogenetic protein 7

558 is expressed in prostate cancer metastases and its effects on prostate tumor cells

559 depend on cell phenotype and the tumor microenvironment. Neoplasia 2010;12:192-

560 205.

561 21. Pfitzenmaier J, Quinn JE, Odman AM, Zhang J, Keller ET, Vessella RL, et al.

562 Characterization of C4-2 prostate cancer bone metastases and their response to

563 castration. J Bone Miner Res 2003;18:1882-8.

564 22. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, et al.

565 Exploration, normalization, and summaries of high oligonucleotide array probe

566 level data. Biostatistics 2003;4:249-64.

22

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

567 23. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the

568 ionizing radiation response. Proc Natl Acad Sci U S A 2001;98:5116-21.

569 24. Hieronymus H, Lamb J, Ross KN, Peng XP, Clement C, Rodina A, et al. Gene

570 expression signature-based chemical genomic prediction identifies a novel class of

571 HSP90 pathway modulators. Cancer Cell 2006;10:321-30.

572 25. Lam HM, Ho SM, Chen J, Medvedovic M, Tam NN. A disrupts HNF4alpha-

573 regulated gene networks linking to prostate preneoplasia and immune disruption in

574 noble rats. Endocrinology 2016;157:207-19.

575 26. Ruppender N, Larson S, Lakely B, Kollath L, Brown L, Coleman I, et al. Cellular

576 adhesion promotes prostate cancer cells escape from dormancy. PLoS One

577 2015;10:e0130565.

578 27. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al.

579 Gene set enrichment analysis: a knowledge-based approach for interpreting genome-

580 wide expression profiles. Proc Natl Acad Sci U S A 2005;102:15545-50.

581 28. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids

582 Res 2000;28:27-30.

583 29. Chan QK, Lam HM, Ng CF, Lee AY, Chan ES, Ng HK, et al. Activation of GPR30

584 inhibits the growth of prostate cancer cells through sustained activation of Erk1/2, c-

585 jun/c-fos-dependent upregulation of p21, and induction of G(2) cell-cycle arrest. Cell

586 Death Differ 2010;17:1511-23.

587 30. Yu Z, Chen S, Sowalsky AG, Voznesensky OS, Mostaghel EA, Nelson PS, et al. Rapid

588 induction of androgen receptor splice variants by androgen deprivation in prostate

589 cancer. Clin Cancer Res 2014;20:1590-600.

23

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

590 31. Cai C, Balk SP. Intratumoral androgen biosynthesis in prostate cancer pathogenesis

591 and response to therapy. Endocr Relat Cancer 2011;18:R175-R182.

592 32. Zhao XY, Malloy PJ, Krishnan AV, Swami S, Navone NM, Peehl DM, et al.

593 Glucocorticoids can promote androgen-independent growth of prostate cancer cells

594 through a mutated androgen receptor. Nat Med 2000;6:703-6.

595 33. Chen EJ, Sowalsky AG, Gao S, Cai C, Voznesensky O, Schaefer R, et al. Abiraterone

596 treatment in castration-resistant prostate cancer selects for progesterone responsive

597 mutant androgen receptors. Clin Cancer Res 2015;21:1273-80.

598 34. Cai C, Chen S, Ng P, Bubley GJ, Nelson PS, Mostaghel EA, et al. Intratumoral de

599 novo steroid synthesis activates androgen receptor in castration-resistant prostate

600 cancer and is upregulated by treatment with CYP17A1 inhibitors. Cancer Res

601 2011;71:6503-13.

602 35. Sahu B, Laakso M, Pihlajamaa P, Ovaska K, Sinielnikov I, Hautaniemi S, et al. FoxA1

603 specifies unique androgen and glucocorticoid receptor binding events in prostate

604 cancer cells. Cancer Res 2013;73:1570-80.

605 36. Efstathiou E, Li W, Gormley M, McMullin R, Ricci DS, Davis JW, et al. Biological

606 heterogeneity in localized high-risk prostate cancer (LHRPC) from a study of

607 neoadjuvant abiraterone acetate plus leuprolide acetate (LHRHa) versus LHRHa. J

608 Clin Oncol 2015;33:(supp: abstract 5005)

609 37. Loriot Y, Bianchini D, Ileana E, Sandhu S, Patrikidou A, Pezaro C, et al. Antitumour

610 activity of abiraterone acetate against metastatic castration-resistant prostate cancer

611 progressing after docetaxel and enzalutamide (MDV3100). Ann Oncol 2013;24:1807-

612 12.

24

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

613 38. Noonan KL, North S, Bitting RL, Armstrong AJ, Ellard SL, Chi KN. Clinical activity of

614 abiraterone acetate in patients with metastatic castration-resistant prostate cancer

615 progressing after enzalutamide. Ann Oncol 2013;24:1802-7.

616 39. Schrader AJ, Boegemann M, Ohlmann CH, Schnoeller TJ, Krabbe LM, Hajili T, et al.

617 Enzalutamide in castration-resistant prostate cancer patients progressing after

618 docetaxel and abiraterone. Eur Urol 2014;65:30-6.

619 40. Badrising S, van dN, V, van Oort IM, van den Berg HP, Los M, Hamberg P, et al.

620 Clinical activity and tolerability of enzalutamide (MDV3100) in patients with metastatic,

621 castration-resistant prostate cancer who progress after docetaxel and abiraterone

622 treatment. Cancer 2014;120:968-75.

623 41. Schmid SC, Geith A, Boker A, Tauber R, Seitz AK, Kuczyk M, et al. Enzalutamide after

624 docetaxel and abiraterone therapy in metastatic castration-resistant prostate cancer.

625 Adv Ther 2014;31:234-41.

626 42. Brasso K, Thomsen FB, Schrader AJ, Schmid SC, Lorente D, Retz M, et al.

627 Enzalutamide antitumour activity against metastatic castration-resistant prostate

628 cancer previously treated with docetaxel and abiraterone: a multicentre analysis. Eur

629 Urol 2015;68:317-24.

630 43. Suzman DL, Luber B, Schweizer MT, Nadal R, Antonarakis ES. Clinical activity of

631 enzalutamide versus docetaxel in men with castration-resistant prostate cancer

632 progressing after abiraterone. Prostate 2014;74:1278-85.

633 44. Bianchini D, Lorente D, Rodriguez-Vida A, Omlin A, Pezaro C, Ferraldeschi R, et al.

634 Antitumour activity of enzalutamide (MDV3100) in patients with metastatic castration-

635 resistant prostate cancer (CRPC) pre-treated with docetaxel and abiraterone. Eur J

636 Cancer 2014;50:78-84.

25

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

637 45. Salvi S, Casadio V, Conteduca V, Burgio SL, Menna C, Bianchi E, et al. Circulating

638 cell-free AR and CYP17A1 copy number variations may associate with outcome of

639 metastatic castration-resistant prostate cancer patients treated with abiraterone. Br J

640 Cancer 2015;112:1717-24.

641 46. Ricci DS, Li W, Griffin TW, Gromley M, Henitz E, Ryan CJ, et al. Predicting response

642 to abiraterone acetate (AA): mRNA biomarker analysis of study COU-AA-302. J Clin

643 Oncol 2014;32(5s):(abstract 5058)

644 47. Sun Y, Wang BE, Leong KG, Yue P, Li L, Jhunjhunwala S, et al. Androgen deprivation

645 causes epithelial-mesenchymal transition in the prostate: implications for androgen-

646 deprivation therapy. Cancer Res 2012;72:527-36.

647 48. Huber MA, Azoitei N, Baumann B, Grunert S, Sommer A, Pehamberger H, et al. NF-

648 kappaB is essential for epithelial-mesenchymal transition and metastasis in a model of

649 breast cancer progression. J Clin Invest 2004;114:569-81.

650 49. Wang X, Belguise K, Kersual N, Kirsch KH, Mineva ND, Galtier F, et al. Oestrogen

651 signalling inhibits invasive phenotype by repressing RelB and its target BCL2. Nat Cell

652 Biol 2007;9:470-8.

653 50. Odero-Marah VA, Wang R, Chu G, Zayzafoon M, Xu J, Shi C, et al. Receptor activator

654 of NF-kappaB Ligand (RANKL) expression is associated with epithelial to

655 mesenchymal transition in human prostate cancer cells. Cell Res 2008;18:858-70.

656 51. Birnie R, Bryce SD, Roome C, Dussupt V, Droop A, Lang SH, et al. Gene expression

657 profiling of human prostate cancer stem cells reveals a pro-inflammatory phenotype

658 and the importance of extracellular matrix interactions. Genome Biol 2008;9:R83.

26

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

659 52. Kong D, Wang Z, Sarkar SH, Li Y, Banerjee S, Saliganan A, et al. Platelet-derived

660 growth factor-D overexpression contributes to epithelial-mesenchymal transition of

661 PC3 prostate cancer cells. Stem Cells 2008;26:1425-35.

662 53. Rajasekhar VK, Studer L, Gerald W, Socci ND, Scher HI. Tumour-initiating stem-like

663 cells in human prostate cancer exhibit increased NF-kappaB signalling. Nat Commun

664 2011;2:162.

665 54. Campa VM, Baltziskueta E, Bengoa-Vergniory N, Gorrono-Etxebarria I, Wesolowski R,

666 Waxman J, et al. A screen for transcription factor targets of glycogen synthase kinase-

667 3 highlights an inverse correlation of NFkappaB and androgen receptor signaling in

668 prostate cancer. Oncotarget 2014;5:8173-87.

669 55. Jeganathan S, Zoubeidi A, Gleave M, Wouters BG, Joshua AM. Using functional and

670 chemical genomics to identify mechanisms of Enzalutamide resistance in prostate

671 cancer. Cancer Res 2015;75(15):(supp: abstract 732)

672

673

27

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

674 Legends

675 Figure 1.

676 Treatment scheme for AA on CRPC PDXs. Castration-resistant tumors were developed and

677 mice were treated orally with either vehicle or AA (0.5 mmol/kg/day). Mice were sacrificed

678 and tumors were harvested on D7 or when tumors reached 1,000 mm3 (EOS).

679 Figure 2.

680 Ultraresponsiveness to AA in LuCaP 136CR PDX models. A, Kaplan-Meier curves showing

681 survival benefits of AA treatment in different LuCaP PDX models. B, Linear model analyses

682 of tumor volume. C, Serum PSA upon AA treatment. n = 9–14 per group.

683 Figure 3.

684 Gene expression associated with LuCaP 136CR ultraresponsiveness. A, Supervised

685 clustering analyses showing 531 differentially expressed genes between LuCaP 136CR and

686 LuCaP 35CR and LuCaP 96CR on D7. Yellow: high gene expression; blue: low gene

687 expression. B, Schematic diagram on gene shaving to identify an eight-gene signature

688 associated with the LuCaP 136CR AA ultraresponsive phenotype. C, qPCR confirmation on

689 the eight-gene signature associated with LuCaP 136CR AA ultraresponsive phenotype (D7

690 and EOS). D, Heat map showing the microarray gene expression of the eight-gene signature

691 in multiple LuCaP models. E, Correlation between the enrichment of the eight-gene

692 signature associated with AA ultraresponsive phenotype and percentage gained in survival

693 upon AA treatment. Percentage survival gained was calculated based on median survival in

694 AA-treated versus vehicle-treated mice in each xenograft model. Each data point or column

695 represented an individual animal. P < 0.05 was considered statistically significant.

696 Figure 4.

697 Biological mechanisms underlying the acquired resistance to AA. A. Venn diagrams showing

698 distinct gene alternations by AA upon treatment resistance at EOS among different LuCaP

699 PDXs. B. Ingenuity Pathway Analysis identified the molecular and cellular functions

700 associated with AA resistance in different LuCaP PDXs. C, Top regulator effect network in

701 AA-resistant tumors in the AA ultraresponder LuCaP 136CR PDXs.

28

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

702 Figure 5.

703 Reduction of androgen signaling upon treatment resistance in the AA ultraresponsive

704 phenotype. A, Levels of intratumoral androgens in control and AA-resistant CRPC PDXs

705 measured by mass spectrometry. n = 2–6 per group. B, Heat map showing the low AR

706 activity (top row, pink squares) and low expression of genes involved in androgen signaling

707 in LuCaP 136CR (n = 4–6 per group), and C, their respective AR signature score in LuCaP

708 PDXs (n = 4–6 per group). D, qPCR analysis in vehicle versus AA-resistant tumors at EOS

709 (n = 4–6 per group). E, Representative IHC pictures of AR and GR, and F, their respective

710 H-score in control and AA-resistant PDXs (n = 6–12 per group). Scale bar: 50 μm.

711 Magnification: 200×. Each data point or column (heat map) represented an individual animal.

712 P < 0.05 was considered statistically significant.

29

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 1

LuCaP 136CR and LuCaP 77CR

Castrate 3 Androgen- Tumor ≥100 mm sensitive tumor bits implanted Start treatment Tumor ≥1.5× volume D7 EOS Vehicle or sacrifice sacrifice AA 0.5 mmol/kg/day

LuCaP 96CR and LuCaP 35CR

CR tumor bits implanted Castrate Start treatment 3 2 weeks Tumor ≥100 mm D7 EOS Downloaded from clincancerres.aacrjournals.orgVehicle or sacrifice on September 23, 2021. © 2016 Americansacrifice Association for Cancer AA 0.5 mmol/kg/day Research. Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 2

A Treatment CX AA LuCaP 136CR LuCaP 77CR LuCaP 96CR LuCaP 35CR HR (95% CI), 18.40 (4.79–70.62) HR (95% CI), 4.11 (1.22–13.78) HR (95%CI), 1.88 (0.63–5.60) HR (95% CI), 0.90 (0.27–2.97) P = 0.0001 P = 0.05 P = 0.02 P = 0.52 100 100 100 100 80 80 80 80 60 21.8 wks 60 9.5 wks 60 10 wks 60 9.5 wks 40 40 40 40 6.8 7.0 wks 5.75 wks 8.5 wks 20 20 20 20 Survival (%) Survival (%) Survival (%) wks Survival (%) 0 0 0 0 0 20 40 60 0 5 10 15 0 5 10 15 0 5 10 15 Time Time Time Time (weeks post enrollment) (weeks post enrollment) (weeks post enrollment) (weeks post enrollment)

B Treatment CX AA LuCaP 136CR LuCaP 77CR LuCaP 96CR LuCaP 35CR P = 0.0003 P = 0.013 P = 0.032 P = 0.22 1,500 1,500 1,500 1,500 ) ) ) ) 3 1,000 3 1,000 3 1,000 3 1,000

500 500 500 500 TV (mm TV (mm TV (mm TV (mm

0 0 0 0 0 20 40 60 0 4 8 12 16 0 10 20 30 40 0 5 10 15 20 Time Time Time Time (weeks post enrollment) (weeks post enrollment) (weeks post enrollment) (weeks post enrollment)

C Treatment CX AA LuCaP 77CR LuCaP 96CR LuCaP 35CR P = 0.001 P = 0.051 P = NS 1,500 1,500 1,500

1,000 1,000 1,000

500 500 500 PSA (ng/mL) PSA (ng/mL) PSA (ng/mL) 0 0 0 0 1 2 3 0 20 40 60 0 5 10 15 20 25 Time Time Time (weeks post enrollment) (weeks post enrollment) (weeks post enrollment)

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. A B -1 -2 -3 3 2 1 0 (8 upregulated and10downregulated) Heatmap 136CRvs 35CRand96CR Downloaded from P <0.0001,fold change≥3 Author ManuscriptPublishedOnlineFirstonDecember19,2016;DOI:10.1158/1078-0432.CCR-16-2054 Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. 531 genes PDX lines 18 genes 20 genes 68 genes 8 genes 136CR qPCR conrmed forOptimal primers qPCR down-regulated genes Top and 10mostup- Secreted family: Gene Validation 35CR clincancerres.aacrjournals.org 96CR C D E Relative mRNA expresssion Relative mRNA expresssion Relative mRNA expresssion 0 1 2 3 4 0 1 2 3 0 1 2 3 136CR 136CR 136CR

Enrichment score (0–8) responder 0 2 4 6 8 Ultra- 50 0 P <0.0001 P <0.0001 P <0.0001 77CR 77CR 77CR on September 23, 2021. © 2016American Association for Cancer (w-ald 0.0032 0.9537 P (two-tailed) 95% CI Pearson r SNTB1 BMP7 Research. CEL 6R 35CR 96CR 6R 35CR 96CR 6R 35CR 96CR % Survival gaineduponAAtreatment % Survival Intermediate responders 0 150 100 0.62.91–0.9951 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 0 1 2 3 4 5 136CR 136CR 136CR P <0.0001 P <0.0001 P <0.0001 77CR 77CR 77CR ARMCX1 FBN2 IER3 6R 35CR 96CR 6R 35CR 96CR 35CR 96CR 0 250 200 responders Minimal 0.0 0.5 1.0 1.5 2.0 2.5 0 1 2 3 4 5 136CR 136CR P <0.0001 P <0.0001 77CR 77CR FSTL5 TNC 6R 35CR 96CR 6R 35CR 96CR Figure 3 Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Figure 4

A B 136CR 96CR 35CR 136CR 77CR P = 1.92 -4 to 1.13 -21 P = 4.46-2 to 1.81-4 Cellular growth and proliferation Cell morphology P = 1.90 -4 to 2.25 -20 P = 4.92-2 to 5.98-4 Cellular development Small molecular biochemistry P = 2.02 -4 to 1.26 -19 P = 4.68-2 to 5.98-4 315 714 59 212 Cellular death and survival Cellular function and maintenance P = 1.70 -4 to 6.02 -19 P = 4.90-2 to 1.36-3 Cellular movement Molecular transport

-4 -16 -3 9 0 P = 2.02 to 1.78 P = 1.59 2 Cell cycle Antigen presentation 0 10 20 30 40 50 0 10 20 30 40 42 % Genes % Genes

77CR 35CR

P = 1.23-2 to 1.26-10 P = 6.58-3 to 8.24-4 Cellular growth and proliferation Cell-to-cell signaling and interaction 96CR P = 1.23-2 to 1.26-10 P = 1.96-2 to 8.24-4 Cellular movement Cell death and survival P = 1.23-2 to 3.56-6 P = 4.28-2 to 8.24-4 Cellular development Cell morphology P = 1.23-2 to 6.15-5 P = 1.15-2 to 8.24-4 Cellular function and maintenance Cellular assembly and organization P = 1.23-2 to 6.15-5 P = 6.58-3 to 8.24-4 Cellular assembly and organization Cellular function and maintenance

0 10 20 30 40 50 0 5 10 15 20 25 % Genes % Genes

C

Prediction Legend

more extreme less Upregulated Downregulated

more con dence less Predicted activation Predicted inhibition

Predicted Relationships Leads to activation Leads to inhibition Findings inconsistent with state of downstream molecule Eect not predicted

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research. GR AR E D A Testosterone ua 3C LCP7C LCP9C LuCaP 35CR LuCaP 96CR LuCaP 77CR LuCaP 136CR LuCaP 35CR LuCaP 96CR LuCaP 77CR LuCaP 136CR NR3C1 V567 O ua 3C ua 7RLCP9C LuCaP35CR LuCaP96CR LuCaP77CR LuCaP136CR EOS AR V7 DHT AED Downloaded from

change pg/mg

Fold pg/mg .408 .001 .405 .90.10 2.59 0.52 1.14 0.12 1.20 0.83 1.04 .408 .60.65 0.86 0.87 0.94 .50.10 3.65 pg/mg DN DN .6N .70.54 1.77 ND 2.66 ND ND ND ND Author ManuscriptPublishedOnlineFirstonDecember19,2016;DOI:10.1158/1078-0432.CCR-16-2054 Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. 0.0 2.5 0.5 1.0 1.5 2.0 0.00 0.08 0.02 0.04 0.06 15 10 0 5 P =0.0381 P =0.009 136CR LuCaP 136CR LuCaP P =0.0333 136CR LuCaP P change P =0.0317 Fold .500 .10.02 2.41 0.03 1.45 LuCaP clincancerres.aacrjournals.org LuCaP 77CR 77CR LuCaP 77CR P =NS CX P =NS P =0.0159 P =0.111 P LuCaP LuCaP 96CR 96CR LuCaP 96CR P =NS AA change Fold .50.00 4.45 LuCaP LuCaP 35CR 35CR P =NS LuCaP P =NS P =NS 35CR P change Fold .60.22 1.66 0.80 on September 23, 2021. © 2016American Association for Cancer Research. 0.40 P C B nGR IHC nAR IHC F

IHC score100 150 200 IHC score100 150 200 50 50 0 0 LuCaP 136CR LuCaP 136CR control

XAA CX AA CX AR signature EOS P =0.5664 P =0.1998 CX sum Z-score 136CR – D7 –15 –10 10 15 20 25 –5 20

Control 0 5 EOS AA P

D7 =NS D7

100 150 200 100 150 200 AA Control 50 50 control 0 0 EOS CX LuCaP 77CR LuCaP 77CR P 77CR =NS XAA CX XAA CX D7

P =0.0302 P =0.0013 D7 AA Control EOS P AA =0.004 EOS ControlD7 AA control EOS 100 150 200 100 150 200 50 50 P 96CR 0 0 EOS AA EOS Control =0.02 LuCaP 96CR LuCaP 96CR XAA CX AA CX EOS P =0.0007 P =0.7669 AA

EOS AA P

EOS Control =NS control EOS 35CR P 200 200 100 150 100 150 EOS AA =NS

50 50 EOS Control EOS AA 0 0 LuCaP 35CR LuCaP 35CR XAA CX AA CX P =0.0132 P =0.8467

EOS AA P =NS 35CR 77CR 96CR 136CR Figure 5 Author Manuscript Published OnlineFirst on December 19, 2016; DOI: 10.1158/1078-0432.CCR-16-2054 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Characterization of an abiraterone ultraresponsive phenotype in castration-resistant prostate cancer patient-derived xenografts

Hung-Ming Lam, Ryan McMullin, Holly Nguyen, et al.

Clin Cancer Res Published OnlineFirst December 19, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-16-2054

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2016/12/17/1078-0432.CCR-16-2054.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2016/12/17/1078-0432.CCR-16-2054. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2016 American Association for Cancer Research.