Is there an opportunity for checkpoint inhibition in MYC amplified castration resistant prostate cancer?

393 Background: The paradigm of prostate cancer (PCa) management continues to expand with the advent of newer therapies. Unfortunately, checkpoint inhibitors (CPIs) have not been very effective in PCa. It has become increasingly apparent that leveraging the PCa tumor immune microenvironment (TiME) is crucial to maximizing therapeutic benefit. Methods: We examined a cohort of 77 radical prostatectomy specimens - 37 received neoadjuvant androgen deprivation therapy (nADT) +/- bicalutamide, and 40 did not receive any presurgical therapy (stage/grade matched as controls). Tumors were evaluated for tumor and immune cell abundance, somatic and germline genomic alterations. We conducted correlation analyses followed by simple and multiple linear regression analyses using the Data Analysis ToolPak on Microsoft excel to evaluate the relationship between genomic profiles and TiME while adjusting for co-occurring alterations and receipt of nADT. Results: These Genomic alterations included Tumor mutational burden (TMB - ranging from 3 to 189), PTEN alterations (PTEN = 8), MYC amplification (MYCa = 20) and Fusions (including ERG, ETV1, MRPS18C, NBEAL1). Among the 20 MYCa patients, 7 received nADT +/- bicalutamide and 13 were controls. We conducted a correlation analysis between PTEN, MYCa, TMB and fusions with the immune cells including M1 and M2 macrophages, T- helper cells (Th), T regulatory (Treg) cells and cytotoxic T lymphocytes (CTL) (Table). MYCa demonstrated correlation coefficients of -0.26 with M1 and -0.27 with M2. Simple linear regression analyses for MYCa with M1 and M2 revealed coefficients (c) -0.4 (p= 0.048), -1.9 (p=0.02) respectively. Multiple regression analyses including MYCa, PTEN, TMB, Fusions, nADT as independent variables were conducted to assess for effect on M1 and M2. For M1 abundance, the model fits well (p = 0.018) with adjusted R 2 0.11, likely due to the low sample size. The M1 coefficient (c) for MYCa was -0.29 (p value 0.14), and for nADT was 0.51 (p = 0.006). For M2 abundance, the model demonstrates a good fit (p value 0.0003) with adjusted R 2 0.23. The M2 c for MYCa was -1.45 with p value approaching statistical significance at 0.06, while nADT was 2.89 (p = 0.00007). Conclusions: MYCa in PCa is likely associated with M2 exclusion (p value 0.06) from the PCa TiME. In a hormone sensitive tumor, the effect of nADT is dominant and overcomes this effect. These data suggest castration resistant PCa may be more susceptible to M2 exclusion due to MYCa. Thus, MYCa tumors may have reduced immunosuppression, enabling successful application of CPIs. Further studies with larger samples and a comprehensive TiME evaluation may assist with therapeutic guidance. Correlation coefficient for genomic alterations and immunophenotype. Correlation coefficient PTEN aMYC TMB Fusions Treg 0.02 -0.14 -0.04 -0.09 M2 0.07 -0.27 -0.03 -0.11 CTL -0.13 -0.14 -0.09 0.06 M1 0.09 -0.26 0.15 -0.08 Th -0.07 -0.19 -0.06 -0.01