Genomic profiles and clinical outcomes to distinguish primary from secondary metastatic hormone-sensitive prostate cancer (mHSPC).

Abstract

e17558 Background: Clinical outcomes may differ among pts with primary (de novo) mHSPC vs. secondary (relapsed) mHSPC occurring after previous local therapy. However, it is unknown what molecular features distinguish these potentially distinct presentations. Methods: A single-center retrospective study was performed using pts with mHSPC and available somatic genomic data. Clinicopathologic characteristics were collected, and pts were classified as primary mHSPC (n = 118) or secondary mHSPC (n = 104). A targeted set of genes suspected to influence disease severity and response to systemic therapy was analyzed: BRCA2, PTEN, RB1, TP53, SPOP, CDK12, combined PTEN/RB1/TP53 alterations (2 out of 3), and a composite of any homologous recombination deficiency (HRD) mutation. Hazard ratios for PFS and OS were compared between groups using the log-rank test. Differences in genomic features and baseline characteristics between groups were compared using Fischer exact and chi-squared tests. Results: Median PFS from initiation of first-line ADT was 16.0 mo and 28.9 mo for men with primary and secondary mHSPC, respectively (HR 0.59, 95% CI 0.43–0.83; P= 0.002). OS did not differ between groups (HR 0.92, 95% CI 0.56–1.52; P= 0.7) There were more men with Gleason 8-10 disease in the primary vs. secondary mHSPC groups (91% vs. 67%, P= < 0.01). In terms of molecular alterations, there were more PTEN and CDK12 mutations in the primary mHSPC group ( PTEN: 27% vs 19%, CDK12: 7% vs 3%), but more BRCA2 mutations in the secondary mHSPC group (5% vs 10%, respectively). In univariate analysis, PTEN mutations (HR 1.39, P= 0.08) and combined PTEN/RB1/TP53 mutations (HR 1.46, P= 0.09) predicted worse PFS; PTEN mutations (HR 1.63, P= 0.09) and combined PTEN/RB1/TP53 mutations (HR 2.38, P= 0.01) also predicted worse OS. In multivariable models adjusting for age at first metastasis and Gleason sum, only combined PTEN/RB1/TP53 mutations were predictive of worse OS (adjusted HR 2.32, 95% CI 1.14–4.71; P= 0.02). Conclusions: Compared to secondary mHSPC, primary mHSPC pts had higher Gleason scores, more frequent PTEN and CDK12 mutations, and less frequent BRCA2 mutations. In our study, such pts demonstrated worse PFS but not OS. While the strongest negative predictor of OS was the presence of combined PTEN/RB1/TP53 mutations, other contributors such as CDK12 and BRCA2 may also partially explain differences in outcomes. Thus, primary mHSPC appears to be clinically and genomically distinct from secondary mHSPC, and may require unique treatment paradigms.