MP45-19: Comparison of predictive markers of effectiveness of PARP inhibitors vs immunotherapy for treatment of clinically advanced prostate cancer (CAPC) based on TMPRSS2:ERG fusion status

Aleksandar Popovic*, Anastasija Useva, Alex Wang, Michael Basin, Rebecca Sager, Jeffrey Ross, Gennady Bratslavsky, Syracuse, NY

Introduction: In recent years, immuno-oncology (IO) has emerged as a novel and significant treatment modality for a variety of metastatic cancers. However, CAPC often has failed to show significant response to IO in the majority of cases, while dramatic responses are noted in selected subgroups. This may be due to CAPC’s locally immunosuppressive microenvironment, referred to as “cold tumors”. Poly(ADP-ribose) polymerase inhibitors (PARPi), meanwhile, may provide promising avenues for systemic therapy in patients with castrate resistant advanced prostate cancer, particularly when considering the high treatment resistance rate of these cancers to other modalities. Furthermore, the TMPRSS2:ERG gene fusion is common in prostate cancer, occurring in 40% of cases. However, studies have reported conflicting results regarding its predictiveness of tumor progression. With recent increases in IO popularity, we wanted to compare the frequency of predictors to PARPi versus IO response, stratified by TMPRSS2 gene fusion status.

Methods: DNA was extracted from 40 µm of formalin-fixed, paraffin-embedded sections of 2,424 samples of CAPC. Comprehensive genomic profiling (CGP) was performed on hybridization-captured, adaptor ligation-based libraries for 315 cancer-related genes. Frequency of mutations predictive of PARP inhibition (BRCA2 and ATM genes) were compared to predictors of IO response tumor mutational burden (TMB) and microsatellite instability (MSI)). TMB was determined on 1.1 Mbp of sequenced DNA and microsatellite instability (MSI) was determined on 114 loci.

Results: We examined samples from 755 TMPRSS2 fusion positive (TMP+) CAPC and 1,669 TMPRSS2 fusion negative (TMP-) CAPC patients with median ages of 66 and 65 years, respectively.  TMP+ CAPC and TMP- CAPC tumors featured mutations in BRCA2 in 8% and 10% (p > 0.05) of samples as well as in ATM in 4% and 7% (p = 0.0032) of samples, respectively. In contrast, predictors of immunotherapy were much less frequent in both TMP+ and TMP- CAPC patients with MSI-high seen in only 1% and 3% (p = 0.01), TMB median of 1.7 and 2.7 (p > 0.05), and TMB > 20 mut/Mb in 2% and 4% (p = 0.01), respectively.

Conclusions: CGP reveals that regardless of TMP fusion status the likelihood of response to PARP inhibitors is much greater as compared to potential immunotherapy responsiveness. Use of CGP may allow identification of actionable cellular and molecular targets to help identify appropriate candidates for systemic therapy in cases of CAPC.

Source of Funding: None