Disease characteristics, survival outcomes, and tumor molecular landscape in patients (pts) with metastatic hormone-sensitive prostate cancer (mHSPC) harboring CDK12 alterations (CDK12a) receiving intensified androgen deprivation therapy (ADTi).

Abstract

189 Background: CDK12 is one of the homologous recombination repair ( HRR) genes involved in genomic stability. The presence of CDK12a has been associated with worse survival outcomes in pts with metastatic prostate cancer [PMID: 32988971; 32462107]. However, the impact of CDK12a on outcomes in pts with mHSPC undergoing ADTi with doublets has never been reported. Herein, we sought to assess this impact while evaluating disease characteristics and tumor molecular landscape. Methods: This was a retrospective IRB-approved study. Eligibility: pts with mHSPC undergoing ADTi with either docetaxel or an androgen receptor pathway inhibitor (ARPI) (i.e., abiraterone, enzalutamide, apalutamide) harboring pathogenic CDK12a on comprehensive genomic profiling (CGP) of primary prostate tissue, metastatic lesion, or on cell-free DNA. Descriptive statistics were used for clinical features and survival outcomes. Optimal PSA response (PSA-R) was defined as PSA ≤ 0.2 ng/ml. Progression-free survival (PFS) was defined from the start of therapy for mHSPC to progression (per PCWG-3) or death from any cause. Overall survival (OS) was defined from treatment initiation for mHSPC to death from any reason or censored at the last follow-up. Concurrent genomic alterations with a frequency ≥ 10% were reported. Results: Of 323 pts with mHSPC receiving ADTi doublet with CGP testing, 19 pts harbored CDK12a and were included. The median age at diagnosis was 66.1 (56 – 77), with 63.2% of pts receiving ADTi with an ARPI. Baseline characteristics are summarized in the Table. Overall, 7 pts (36.8%) achieved an optimal PSA-R by 6 months (mo). The median PFS was 18.8 mo (95% CI 4.2 – 33.4 mo), and the median OS was 33.1 mo (95% CI 9.5 – 56.7 mo). CGP was performed on primary tissue (84.2%), followed by metastatic lesion (10.5%) and blood (5.3%). PIK3CA and TP53 were the most frequent associated genomic alterations (15.8%), followed by ARID1A, CTNNB1, RB1, and TMPRSS2 (10.5%). Conclusions: This is the first real-world study to report outcomes of pts with mHSPC and tumor CDK12a undergoing ADTi with doublets. We show these pts to have inferior outcomes than those enrolled in the registration trials of mHSPC with ADTi. Upon external validation in larger cohorts, these hypothesis-generating data may guide future clinical trial design for these pts, their preferential selection for trials, and counseling and prognostication in the clinic. [Table: see text]