The immune landscape of prostate cancer Shuang G. Zhao MD1, Jonathan Lehrer2, S. Laura Chang PhD3, Rajeep Das MD, PhD3, Nicholas Erho MS2, Martin Sjostrom PhD4, Robert B. Den MD5, Stephen J. Freedland MD6, Eric A. Klein MD7, R. Jeffrey Karnes MD8, Edward M. Schaeffer MD, PhD9, Melody Xu MD3, Corey Speers MD, PhD1, Paul L. Nguyen MD10, Ashley E. Ross MD, PhD11, June M. Chan ScD13,14, Matthew R. Cooperberg, MD13, Peter R. Carroll MD13, Elai Davicioni PhD2, Lawrence Fong MD12, Daniel E. Spratt MD1, Felix Y Feng MD3,12,13 1
Department of Radiation Oncology, University of Michigan, Ann Arbor, MI GenomeDx Biosciences Inc., Vancouver, British Columbia, Canada Departments of 3Radiation Oncology, 12Medicine, 13Urology, and 14Epidemiology & Biostatistics, Helen Diller Comprehensive Cancer Center, University of California at San Francisco, San Francisco, CA 4 Lund University, Lund, Sweden 5 Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA 6 Samuel Oschin Comprehensive Cancer Institute, Cedars-Sinai Medical Center, Los Angeles, CA 7 Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, OH 8 Department of Urology, Mayo Clinic, Rochester, MN 9 Department of Urology, Northwestern University, IL 10 Dana-Farber/Brigham and Women’s Cancer Center, Department of Radiation Oncology, Harvard Medical School, Boston, MA 11 James Buchanan Brady Urological Institute, Johns Hopkins Medical Institutions, Baltimore, MD 2
Abstract Purpose: Immunotherapy has been less successful in treating prostate cancer than other solid tumors. We sought to better understand the immune landscape in prostate cancer. Experimental Design: We analyzed gene expression data from 7826 prospectively collected prostatectomy samples collected between 2013-2016, and 1567 retrospective samples with long-term clinical outcomes, all profiled on a commercial clinical platform. Results: Unsupervised hierarchical clustering of the hallmark pathways of all 9393 samples demonstrated an immune-related cluster of tumors. Increased estimated immune content scores based on immunespecific genes from the literature were associated with worse biochemical recurrence free survival (bRFS, p=0.0002, HR=1.26 [1.12−1.42]), distant metastasis-free survival (DMFS, p=0.0006, HR=1.34 [1.13−1.58]), prostate cancer-specific survival (PCSS, p=0.0003, HR=1.53 [1.21−1.92]), and overall survival (p=0.006, HR=1.27 [1.07−1.50]). Additionally, de-convolution using Cibersort revealed that mast cells, NK cells, and dendritic cells conferred improved DMFS, and macrophages and T-cells conferred worse DMFS, all consistent with the literature. Interestingly, while PD-L1 was not prognostic, consistent with its low expression in prostate cancer, PD-L2 was expressed at significantly higher levels (p<0.0001) and was associated with worse bRFS (p=0.013, HR=1.17 [1.03−1.33]), DMFS (p=0.014, HR=1.25 [1.05−1.49]), and PCSS (p=0.0033, HR=1.45 [1.13−1.86]). PD-L2 may also predict response to post-operative radiation therapy (PORT) on a multivariate interaction analysis (p=0.029) with PD-L2 conferring worse DMFS only in patients who did not receive PORT. Conclusions: These results illustrate the complex relationship between the tumor-immune interaction, prognosis, and response to radiotherapy, and suggest PD-L2 as a therapeutic target in prostate cancer which merits further investigation, potentially in combination with radiotherapy. Disclosures: SGZ reports travel expenses from GenomeDx. SGZ, SLC, and FYF have patents pending with GenomeDx not related to the work in this manuscript. JL, NE, and ED are employees of GenomeDx. SJF
reports receiving research funding from GenomeDx; and serving as a consultant for GenomeDx, Astellas, Medivation, Bayer, Sanofi, Janssen, Dendreon, Armune, Parallel 6, Singulex, Boston Scientific, and Churchill. EAK reports serving as a consultant for Berg and GenomeDx, serving on the speakers bureau for Genomic Health, and receiving research funding from GenomeDx and Genomic Health. RJK reports receiving research funding and travel expenses from GenomeDx. EMS reports serving as a consultant for GenomeDx and Myriad Genetics. PLN reports serving as a consultant for Medivation, GenomeDx, Ferring, Nanobiotix, and Dendreon, and receiving research funding from Astellas. AER reports serving as a consultant for GenomeDx, having an ownership interest in GenomeDx, and receiving research funding from Merck, Novartis, and Metamark. PRC reports receiving honoraria from Takeda Pharmaceutical and Genomic Health and serving as a consultant for Janssen Pharmaceutical and Medivation. FYF and CS report having ownership in PFS Genomics. SLC is an employee of PFS Genomics. FYF reports consulting for Medivation/Astellas, GenomeDx, Celgene, Sanofi, Dendreon, and Merck, and receiving research funding from Varian Medical Systems.