Abstract 2510: Combining genetic and histopathologic features to predict response to anti-androgen therapy in aggressive prostate cancer

Abstract

Abstract Background: Genetic alterations in lethal, metastatic prostate cancer include the loss of PTEN, translocation of the TMPRSS2 and ERG genes, upregulation of the androgen receptor (AR), and disruption of the DNA homologous repair pathway driven by mutations to BRCA1, BRCA2, and ATM. Many of these mutations can be observed while the cancer is still localized in the prostate. Here, we seek to identify genetic features indicative of therapeutic response and novel drivers of cancer progression to inform clinical practice. Methods: Using tissue from 33 patients in an intense neoadjuvant anti-androgen clinical trial at the NCI (NCT02430480), we examined genetic features that would predict exceptional or poor response to anti-androgen enzalutamide therapy. Each patient on trial presented with multiple tumor foci, allowing us to investigate the intratumoral heterogeneity across foci within individual patients, as well as investigating the tumor profiles across multiple patients. Several patients exhibited exceptional response with residual tumor burdens less than 0.5cc, while others had substantial treatment-resistant cancers. We developed a panel of 12 immunohistopathological stains and used this panel to guide laser capture microdissection on pre-treatment biopsies and spatially matched post-treatment radical prostatectomy specimens to isolate ultrapure tumor foci from each patient. DNA from these foci was used for whole exome sequencing, as somatic copy number alterations and mutations also confirm their evolutionary relationship. This enabled us to classify baseline specimens as responder or nonresponder, while examining variations in genetic features between the two cohorts. Results: To date, focal PTEN loss was observed in all nonresponders, while focal ERG staining was absent in 100% of responders and present in 60% of nonresponders. Synaptophysin positivity was rare at baseline but predicted resistance to treatment with 100% sensitivity. Intriguingly, baseline copy number profiles highlight a 6q deletion in 100% of exceptional responders, but not in the non-responders. Together, these data suggest an immunostain panel to assess oncogene and tumor suppressor alterations can predict response to anti-androgen therapy, while also suggesting a novel role of 6q in resistance to anti-androgen hormone therapy. Current studies are examining the role of 6q in response to anti-androgen therapies, which remains an area of active interest in our laboratory. Conclusions: These findings demonstrate the feasibility in identifying intratumoral heterogeneity based on prostate biomarker status both in pre-treatment and post-treatment specimens. Using these data, comprehensive molecular analysis of prostate cancer at diagnosis may better-enable physicians to predict response to anti-androgen therapy and provide tailored treatment based on gene expression status. Citation Format: Scott Wilkinson, Huihui Ye, Nicole Carrabba, Rayann Atway, Shana Y. Trostel, Thomas Hennigan, Ross Lake, Stephanie Harmon, Guinevere Chun, Baris Turkbey, Peter A. Pinto, Peter L. Choyke, Fatima Karzai, David J. VanderWeele, Kathleen Kelly, William L. Dahut, Adam G. Sowalsky. Combining genetic and histopathologic features to predict response to anti-androgen therapy in aggressive prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2510.