Genomic landscape and clinical characteristics of patients (pts) with neuroendocrine prostate cancer (NEPC) with or without aggressive variant prostate cancer characterized by molecular signature (AVPC-MS).

Abstract

5087 Background: NEPC is a rare, aggressive subtype that can arise de novo or after treatment with androgen receptor inhibitors. Recent NEPC clinical trials have also included pts who have AVPC-MS (Aggarwal et al , ASCO 2024). However, the prevalence and clinical characteristics of NEPC with or without AVPC-MS need further investigation. Methods: The de-identified Tempus Lens dataset was used to retrieve records of pts diagnosed with NEPC. Pts were classified as having AVPC-MS if they had ≥2 alterations in TP53 , RB1 , and/or PTEN . Demographic and clinical characteristics were summarized using descriptive statistics and compared between individuals with and without AVPC-MS using chi-square and Fisher’s exact tests. The prevalence of genomic alterations between pts with and without AVPC-MS was compared using a two-proportions Z-test with False Discovery Rate (FDR) correction to account for multiple comparisons. Results: A total of 308 pts diagnosed with NEPC were identified and included in this analysis. Among them, 124 (40.3%) had RB1 , 141 (45.8%) had TP53 , and 64 (20.8%) had PTEN alterations. A total of 109 (35.4%) pts met the criteria for AVPC-MS whereas 81 (26.2%) pts harbored only a single alteration in either TP53 , RB1 , or PTEN and thus did not meet the criteria. The proportion of individuals aged >60 years was similar between those with and without AVPC-MS (56.0% vs. 58.8%, p=0.70). Additionally, 48.6% of pts with AVPC-MS were white, compared to 60.8% of those without AVPC-MS (p=0.03). Mortality rates did not significantly differ between the groups (66.1% in AVPC-MS vs. 61.8% in non-AVPC-MS, p=0.40). Individuals with AVPC-MS had significantly higher alterations of TMPRSS2 (38.5% vs. 18.1%) and PIK3CA (8.3% vs. 2.5%) compared to those without AVPC-MS (FDR-adjusted p < 0.05). However, no significant difference was observed for alterations in FOXA1 and BRCA2 in the AVPC-MS group compared to non-AVPC-MS (10.1% vs. 10.1% and 8.3% vs. 11.1%, respectively). Among pts with available data, high tumor mutational burden (TMB) was observed in 7.8% (8 of 103) of pts with AVPC-MS compared to 2.1% (2 of 97) of those without AVPC-MS (p=0.10). Similarly, microsatellite instability-high (MSI-high) was detected in 6.6% (7 of 106) of pts with AVPC-MS and 1.9% (2 of 106) of those without AVPC-MS (p=0.17). Conclusions: Approximately one in three patients with NEPC also meet the criteria for AVPC-MS. Our findings suggest that the biological drivers for NEPC are mutually exclusive from AVPC-MS in most patients, and highlight the importance of utilizing pathologic confirmation of NEPC rather than relying on genomic surrogates. Further research is warranted to explore the clinical implications of these genomic differences and to develop personalized treatment approaches for pts with AVPC-MS.