Robert B. Montgomery

Abiraterone Treatment in Castration-Resistant Prostate Cancer Selects for Progesterone Responsive Mutant Androgen Receptors

Purpose: The CYP17A1 inhibitor abiraterone markedly reduces androgen precursors and is thereby effective in castration-resistant prostate cancer (CRPC). However, abiraterone increases progesterone, which can activate certain mutant androgen receptors (AR) identified previously in flutamide-resistant tumors. Therefore, we sought to determine if CYP17A1 inhibitor treatment selects for progesterone-activated mutant ARs. Experimental Design: AR was examined by targeted sequencing in metastatic tumor biopsies from 18 patients with CRPC who were progressing on a CYP17A1 inhibitor (17 on abiraterone, 1 on ketoconazole), alone or in combination with dutasteride, and by whole-exome sequencing in residual tumor in one patient treated with neoadjuvant leuprolide plus abiraterone. Results: The progesterone-activated T878A-mutant AR was present at high allele frequency in 3 of the 18 CRPC cases. It was also present in one focus of resistant tumor in the neoadjuvant-treated patient, but not in a second clonally related resistant focus that instead had lost one copy of PTEN and both copies of CHD1. The T878A mutation appeared to be less common in the subset of patients with CRPC treated with abiraterone plus dutasteride, and transfection studies showed that dutasteride was a more potent direct antagonist of the T878A versus the wild-type AR. Conclusions: These findings indicate that selection for tumor cells expressing progesterone-activated mutant ARs is a mechanism of resistance to CYP17A1 inhibition. Clin Cancer Res; 21(6); 1273–80. ©2014 AACR. See related commentary by Sharifi, p. 1240

C11-acetate and F-18 FDG PET for men with prostate cancer bone metastases: Relative findings and response to therapy

Purpose of the Report: This study tested the feasibility of C11-acetate (acetate) positron emission tomography (PET) imaging to assess response to therapy in men with bone metastatic prostate cancer and compared results for disease detection and response evaluation with F-18 fluorodeoxyglucose (FDG) PET. Materials and Methods: Men with ≥3 prostate cancer bone metastases identified by Tc-99m methylene diphosphonate (MDP) bone scintigraphy and/or computed tomography were enrolled in a prospective study of serial acetate and FDG PET imaging. Patients were imaged before and 6 to 12 weeks after initial androgen deprivation therapy for new metastatic prostate cancer or first-line chemotherapy with docetaxel for castration-resistant prostate cancer. Qualitative assessment and changes in the tumor:normal uptake ratio were used to assess response by both acetate and FDG PET. In addition, the detection of bone metastases pretherapy was compared for acetate and FDG PET. Results: A total of 8 patients with documented bone metastases were imaged, of which 6 were imaged both pre- and post-therapy. Acetate PET detected bone metastases in all 8 patients, whereas FDG PET detected lesions in 6 of the 7 imaged patients. Acetate PET generally detected more metastases with a higher tumor:normal uptake ratio. Qualitative and quantitative assessments of post-treatment response correlated with composite clinical designations of response, stable disease, or progression in 6 of 6 and 5 of 6 by acetate and 4 of 5 and 3 of 5 by FDG PET, respectively. Conclusions: In this pilot study, results indicate that acetate PET holds promise for response assessment of prostate cancer bone metastases and is complementary to FDG PET in bone metastasis detection.

The basic biochemistry and molecular events of hormone therapy

Data regarding the molecular response of prostate cancer to hormone therapy continue to emerge, identifying a complex network of autocrine and paracrine signaling events mediating the tumor response to androgen suppression. Emerging data provide insight into cellular pathways important in the apoptotic response to therapy, including the transforming growth factor-β, insulin-like growth factor-1, and vascular endothelial growth factor signaling axes. They also reveal mechanisms of direct antitumor cytotoxicity mediated by various hormonal agents and highlight the importance of developing antiandrogens capable of irreversibly inhibiting the androgen receptor. Accumulated data emphasize the presence of residual androgens and persistent activation of androgen receptor signaling in advanced prostate tumors despite castration. These factors suggest that a multitargeted treatment approach designed to ablate all contributions to the androgen receptor signaling axis within the prostate tumor microenvironment will be required in order for hormonal therapy to achieve optimal antitumor efficacy.

Effects of luteinizing hormone receptor signaling in prostate cancer cells

The importance of androgen signaling in prostate cancer (PC) is well described and prostate cancer cells retain the ability to directly synthesize androgens. Luteinizing hormone (LH) can induce expression of steroidogenic enzymes and trigger androgen production, but the regulation of this process is not well‐described. Here, we explored the impact of silencing LH receptor (LHR) silencing on androgen synthesis and on several relevant signaling pathways in PC.

Influence of ischemia on tissue androgens in eugonadal and androgen-deprived men treated with prostatectomy.

118 Background: Optimizing tissue androgen suppression may provide a higher rate of complete pathologic response in patients with localized prostate cancer (PCa). The ability to accurately assess tissue androgens may be influenced by ischemic time and catabolism after devascularization. This study was performed to determine if ischemia influences tissue androgen in eugonadal men and in men treated with various types of neoadjuvant androgen deprivation (ADT). METHODS Eugonadal men (n=12) or men treated with neoadjuvant ADT using LHRH agonist and Casodex or LHRH agonist with Dutasteride, Casodex or all three with Ketoconazole (n=34) underwent prostatectomy for localized PCa. Tissue was acquired by needle biopsy at the time of ligation of the vascular pedicles, at prostate removal and at 30 minutes after prostate removal. Prostate tissue was flash-frozen and tissue dihydrotestosterone (DHT) and testosterone (T) were determined by LC/MS/MS. Change in tissue androgens from clamp was assessed at each time point. Trends in androgen levels were assessed by ANOVA for eugonadal men and within each treatment group Results: Average time from clamp to removal was 23 minutes (± 8) and from removal to subsequent biopsy 32 minutes (± 2). There was no evidence for a significant change in T or DHT over the time intervals tested (one way ANOVA p>0.05 for all measures) except in patients treated with LHRH agonist and Casodex alone. In this group of patients, T declined by 34%by the time of prostate removal, and by 83% at 30 minutes after removal when compared to levels at clamp (one way ANOVA p=0.017, test for linear trend =0.007). In all other treatment groups, both T and DHT were stable for 60 minutes. CONCLUSIONS Assessment of DHT levels in prostate tissue at prostatectomy is not compromised by ischemia. Tissue sampling from 30-60 minutes after vascular clamping yields a stable measure of tissue DHT in the majority of patients. In the absence of SRD5A blockade, testosterone sampling in castrate men is sensitive to ischemia, and delayed sampling underestimates tissue androgen levels due to ischemia-related metabolism. These results lend important insight into clinical trial design and translational endpoints.

Abstract 305: The biological and molecular characterization of clinically relevant prostate cancer xenograft lines (LuCaP series), including responses to therapy.

Proceedings: AACR 104th Annual Meeting 2013; Apr 6-10, 2013; Washington, DC Introduction: Prostate cancer (PCa) is a heterogeneous disease, which results in an unpredictable and varied response to therapy. A limitation in unraveling the complexities of PCa and developing / evaluating novel therapeutic strategies has been the lack of pre-clinical models that closely replicate this heterogeneity. To overcome this limitation we have established over 3 dozen PCa xenograft lines (LuCaP series). Methods: Characterization of the xenograft lines derived from PCa primaries and metastases includes: (a) growth properties, (b) expression of 45 biomarkers by immunohistochemistry (IHC), (c) gene expression, (d) copy number gains and losses, (e) expression of the androgen receptor (AR) and its splice variants, (f) bone response (i.e. osteoblastic, osteolytic or mixed), and (g) response to therapy, i.e. androgen ablation, docetaxel and anti-IGF-1R. Results: Forty distinct xenograft lines comprise the current LuCaP panel. Four are neuroendocine, 12 are castration resistant (CR) sublines and 7 are abiraterone or MDV-3100 resistant sublines. Comprehensive characterization studies have been done on 24 lines. All lines histologically resemble the originating clinical specimen. Unsupervised gene expression array clustering analyses revealed (a) association between the xenograft and the originating clinical specimen, (b) pairing of androgen-sensitive lines with their CR sublines, (c) a distinction between adenocarcinoma and neuroendocrine phenotypes and (d) insignificant drift over a 2-5 year period of serial passage. Biomarker expression is quite heterogeneous and in most cases, protein expression correlated well with gene expression. Importantly, 7 LuCaP models elicit an osteoblastic reaction in the bone, 5 models are PTEN negative, and 8 lines have the TMPRSS2:ERG fusion. The xenograft lines express different levels of AR with some expressing AR splice variants. Heterogeneity was also observed in responses to therapy; prolonged survival (PS) following androgen ablation or docetaxel treatment ranged from 1 - 7 fold. Interestingly, LuCaP 86.2, expressing predominantly ARv567es, was among the least responsive to androgen ablation (PS 1.1) whereas it is one of the most responsive to docetaxel (PS >4). Several novel anti-androgen therapies are currently under investigation as individual agents and in combination; heterogeneous responses are being observed. To explore mechanisms of resistance, we are also maintaining sublines that developed resistance to abiraterone and MDV-3100. Conclusions: These LuCaP PCa xenograft lines are highly diverse and clinically relevant models to study PCa biology and to evaluate new treatment modalities. The diversity of phenotypes and responses to therapy most importantly suggests that misleading conclusions can be drawn from the use of only one or two models in preclinical evaluations. Citation Format: Holly Nguyen, Eva Corey, Colm Morrissey, Peter Nelson, Xiaotun Zhang, Martine Roudier, Stephen Plymate, Lawrence True, Celestia Higano, Robert Montgomery, Paul Lange, Robert Vessella. The biological and molecular characterization of clinically relevant prostate cancer xenograft lines (LuCaP series), including responses to therapy. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 305. doi:10.1158/1538-7445.AM2013-305

AR-V7 Protein in Circulating Tumor Cells—The Decider for Therapy?

In this issue of JAMA Oncology, Scher et al 1 present additional evidence that the presence of the constitutively active androgen receptor variant 7 (AR-V7) is a marker for resistance to standard androgen deprivation or second line androgen receptor (AR) targeting agents in men with metastatic prostate cancer. Blood samples were drawn prior to therapy and nucleated cells analyzed for protein expression by immunofluorescence for AR-V7 protein. Results were not used for clinical decision making. Remarkably, no patients with circulating tumor cells (CTCs) positive for AR-V7 protein responded to AR-directed therapy (as defined by a >50% prostate-specific antigen decline), and they had significantly worse progression-free survival and overall survival (OS) compared with patients with AR-V7– negative CTCs. These findings are in agreement with those previously reported by Antonorakis et al 2 using a different messenger RNA (mRNA)-based CTC assay. 2 These 2 studies provide compelling evidence that AR-V7–positive CTCs are a predictive biomarker for failure to respond to AR-directed therapy (AR-DT). The studies by Scher et al 1 and Antonorakis et al 2 differ in several respects. First, there was a lower proportion of patients with AR-V7–positive CTCs at each line of therapy using the protein-based assay compared with the mRNAbased assay. Although the patient groups were not entirely identical, this difference in detection may reflect greater sensitivity provided by polymerase chain reaction (PCR) vs antibody detection systems. Second, the issues of specificity of antibodies vs PCR are clearly different. Work has been presented showing that AR-V7 antibodies can react with nonprostate tissue and cell lines such as PC-3, in which no AR-V7 message or N-terminal AR protein c an be identified. 3 In the study by Scher et al, AR-V7 immunofluorescence was identified in CTCs that were cytokeratin negative or where other presumed CTCs in the sample were negative for AR N-terminal staining that should detect AR-V7. Although relevant in only a small fraction of the patient samples, appropriate prospective and interinstitutional correlative studies should be done to confirm sensitivity and specificity of the assay before the current test can be applied broadly.

Abstract 1718: The importance of clinically relevant rapid autopsy specimens and LuCaP patient-derived xenograft models to interrogate the heterogeneous and evolving treatment resistance of castration-resistant prostate cancer

Recent advances in our understanding of metastatic castration-resistant prostate cancer (mCRPC) biology have led to the approval and use of multiple novel drugs, including abiraterone, enzalutamide, cabazitaxel, sipuleucel-T and radium-223. Although these agents all provide improvement in survival, resistance always develops after an initial response. To understand the biology of responses and resistance, clinically relevant specimens and preclinical models are needed. To address this critical need, our objectives are: 1) to collect samples of metastases prior to and after treatment for evaluation of potential biomarkers of responses and resistance and 2) to generate patient-derived xenografts (PDXs) that are representative of these novel phenotypes of advanced mCRPC. A Rapid Autopsy Program (RAP) was established at the University of Washington to study the lethal phenotype of mCRPC. Through the RAP we have collected more than 3000 tissue specimens, including bone and visceral metastases from over 110 patients who died of mCRPC. The samples were processed for paraffin embedding and frozen for future histological and molecular analyses. All samples are tracked in a relational database and cross-linked to clinical and pathologic data. To provide clinically relevant preclinical models for studies of late stage advanced prostate cancer, we also implant tissues acquired during the RAP into immunocompromised mice in order to establish novel CRPC PDXs. At present we have 28 primary PDXs (the LuCaP series), and 10 castration-resistant sublines of the primary lines. The LuCaP models have been extensively characterized by light microscopy, immunohistochemistry, expression arrays, RNASeq, IHC, exome sequencing and other methodologies. We have also determined responses of these PDXs to castration and docetaxel. To address the need for models representing the current tumor phenotypes that exist in patients who develop resistance to secondary androgen-targeting therapy, we have also evaluated responses to abiraterone and enzalutamide, and are experimentally developing abiraterone- and enzalutamide-resistant LuCaP PDXs. In addition, we are continuing to generate novel PDXs using abiraterone- and enzalutamide-resistant patient samples, and have four models that were established from abiraterone-resistant tumors. In summary, we have established a large biorepository of mCRPC specimens and PDXs. Moreover, due to the constant evolution of the advanced PCa phenotype we continue to acquire mCRPC specimens and establish novel PDXs that are refractory to the most recent treatments which allow us to investigate biology of late stage mCRPC and to provide valuable clinical specimens and LuCaP PDXs to the PCa research community. Citation Format: Colm Morrissey, Ming H. Lam, Tia S. Higano, Lawrence D. True, Martine Roudier, Robert B. Montgomery, Peter S. Nelson, Paul H. Lange, Evan Y. Yu, Robert L. Vessella, Eva Corey. The importance of clinically relevant rapid autopsy specimens and LuCaP patient-derived xenograft models to interrogate the heterogeneous and evolving treatment resistance of castration-resistant prostate cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1718. doi:10.1158/1538-7445.AM2015-1718

665 EFFECT OF NEOADJUVANT ABIRATERONE ACETATE PLUS LEUPROLIDE ACETATE ON PSA, PATHOLOGIC RESPONSE, AND INTRAPROSTATIC/SERUM ANDROGEN LEVELS IN LOCALIZED HIGH-RISK PROSTATE CANCER: RESULTS OF A RANDOMIZED PHASE 2 STUDY

Jerome P. Richie*, Boston, MA; Robert B. Montgomery, Seattle, WA; Christopher J. Logothetis, Houston, TX; Glenn J. Bubley, Boston, MA; Bruce L. Dalkin, Seattle, WA; Martin G. Sanda, Massimo F. Loda, Rosina T. Lis, Boston, MA; Lawrence D. True, Seattle, WA; Patricia Troncoso, Houston, TX; Elizabeth M. Genega, Steven P. Balk, Boston, MA; Elahe A. Mostaghel, Seattle, WA; Trevor M. Penning, Philadelphia, PA; Peter S. Nelson, Seattle, WA; Wanling Xie, Boston, MA; Christopher M. Haqq, NamPhuong Tran, Weimin Peng, Los Angeles, CA; Daniel Tamae, Philadelphia, PA; Thian Kheoh, Arturo Molina, Los Angeles, CA; Philip W. Kantoff, Mary-Ellen Taplin, Boston, MA

Abstract 5603: Response of human prostate cancer xenografts to IGF-IR inhibition correlates with PTEN and androgen receptor splice variant status

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Clinical studies with agents targeting the insulin-like growth factor receptor (IGF-IR) in prostate cancer have shown promise but as with most single agent trials only a portion of patients that enter into a targeted agent trial respond, e.g. activating mutations in the EGF-receptor in patients with lung cancer. We have shown in human tumor xenografts that the combination of IGF-IR inhibition with the human mab IMC-A12 in combination with androgen deprivation (ADT) was more efficacious than ADT alone. In a subsequent neoadjuvant study with IMC-A12, responses were at least as efficacious in hi-grade disease (Gleason 7 (4+3), 8, and 9) when compared to ADT alone. In order to determine predictors of response to IMC-A12 plus ADT (surgical castration) in human prostate cancer, we examined the response of 21 human prostate cancer xenografts (LuCaP series) grown in SCID mice for responses to castration, IMC-A12 40mg/kg ip 3x a week for four weeks, or castration followed by four weeks of IMC-A12. At the end of the IMC-A12 treatment animals were followed until tumor recurrence. When all animals were completed tumor growth was analyzed. The main differences in growth rates, as expected based on previous studies, occurred in mice receiving IMC-A12 plus castration compared to castration alone. Responses were segregated into three groups: Major Response - slope of tumor growth was negative in the IMC-A12 + Castration group compared to castration alone; Moderate Response - slope of combined treatment is flat and castration alone is positive; No-response - no differences in slope between castration and castration plus IMC-A12. We then examined the association of response to IMC-A12 to potential effectors of IGF-IR expression and signaling in the prostate including AR and AR splice variants and PTEN status. Tumor transcript profiles were generated on all xenografts prior to treatment and gene sets predictive of response determined. None of the PTEN null tumors showed a strong response to IMC-A12 and all PTEN null tumors expressed the AR-splice variant ARv567es. Of the 3 xenografts with one copy of PTEN, non- or partial responders either expressed no AR (neuroendocrine tumor) or expressed ARv567es. Of xenografts with 2 or more copies of PTEN, non- or moderate responders expressed AR-variants. Finally, transcript profiles provided a gene signature that was highly predictive of response to IMC-A12. These data indicate that PTEN insufficiency or the presence of an AR-splice variant that cannot increase IGF-IR via non-genomic activity are associated with resistance to IMC-A12. Further, a gene signature has been developed from the 21 xenografts that is predictive of response to IMC-A12 and castration. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5603. doi:1538-7445.AM2012-5603