Prospective Comprehensive Genomic Profiling of Primary and Metastatic Prostate Tumors

JCO Precis Oncol. 2019:3:PO.18.00283. doi: 10.1200/PO.18.00283. Epub 2019 May 10.

Abstract

Purpose: Comprehensive genomic profiling (CGP) is increasingly used for routine clinical management of prostate cancer. To inform targeted treatment strategies, 3,476 clinically advanced prostate tumors were analyzed by CGP for genomic alterations (GAs) and signatures of genomic instability.

Methods: Prostate cancer samples (1,660 primary site and 1,816 metastatic site tumors from unmatched patients) were prospectively analyzed by CGP (FoundationOne Assay; Foundation Medicine, Cambridge, MA) for GAs and genomic signatures (genome-wide loss of heterozygosity [gLOH], microsatellite instability [MSI] status, tumor mutational burden [TMB]).

Results: Frequently altered genes were TP53 (44%), PTEN (32%), TMPRSS2-ERG (31%), and AR (23%). Potentially targetable GAs were frequently identified in DNA repair, phosphatidylinositol 3-kinase, and RAS/RAF/MEK pathways. DNA repair pathway GAs included homologous recombination repair (23%), Fanconi anemia (5%), CDK12 (6%), and mismatch repair (4%) GAs. BRCA1/2, ATR, and FANCA GAs were associated with high gLOH, whereas CDK12-altered tumors were infrequently gLOH high. Median TMB was low (2.6 mutations/Mb). A subset of cases (3%) had high TMB, of which 71% also had high MSI. Metastatic site tumors were enriched for the 11q13 amplicon (CCND1/FGF19/FGF4/FGF3) and GAs in AR, LYN, MYC, NCOR1, PIK3CB, and RB1 compared with primary tumors.

Conclusion: Routine clinical CGP in the real-world setting identified GAs that are investigational biomarkers for targeted therapies in 57% of cases. gLOH and MSI/TMB signatures could further inform selection of poly (ADP-ribose) polymerase inhibitors and immunotherapies, respectively. Correlation of DNA repair GAs with gLOH identified genes associated with homologous recombination repair deficiency. GAs enriched in metastatic site tumors suggest therapeutic strategies for metastatic prostate cancer. Lack of clinical outcome correlation was a limitation of this study.