Utility of ctDNA burden as a prognostic biomarker for efficacy in TALAPRO-2: A phase 3 study of talazoparib (TALA) + enzalutamide (ENZA) vs placebo (PBO) + ENZA as first-line (1L) treatment in patients (pts) with metastatic castration-resistant prostate cancer (mCRPC).

5020 Background: TALAPRO-2 (NCT03395197) demonstrated that 1L TALA + ENZA significantly improved radiographic progression-free survival (rPFS) vs PBO + ENZA for pts with mCRPC. ctDNA burden is a candidate prognostic biomarker with potentially broad utility across treatments and tumor types. We assessed the potential prognostic utility of baseline (BL) ctDNA burden and changes in ctDNA burden at Week 9 (WK 9) in TALAPRO-2 pts. Methods: Retrospectively analyzed serial ctDNA samples from BL and WK 9 were assessed using FoundationOneLiquid CDx. Plasma tumor fraction was calculated based on aneuploidy (Husain et al. JCO Precis Oncol. 2022. PMID: 36265119). We categorized ctDNA burden as high (ctDNA burden quantifiable) vs low (unknown ctDNA burden). Data cutoff date was August 16, 2022. Results: In the all-comers intent-to-treat population, 678 pts were evaluable for ctDNA burden at BL: 26% (89/337) of TALA + ENZA pts were ctDNA-high and 74% (248/337) were ctDNA-low; 29% (98/341) of PBO + ENZA pts were ctDNA-high and 71% (243/341) were ctDNA-low. High ctDNA burden at BL was prognostic of inferior rPFS in the TALA + ENZA and PBO + ENZA arms (Table). A relatively favorable median rPFS was observed for pts with low ctDNA at BL and WK 9 for TALA + ENZA (n=206) and PBO + ENZA (n=207), as reported in the Table. At WK 9, 72 pts in the TALA + ENZA and 77 pts in the PBO + ENZA arms were evaluable for ctDNA conversion from high to low. In both treatment arms, conversion from high to low ctDNA was prognostic of improved rPFS vs pts who remained ctDNA-high (Table). Pts who remained ctDNA-low had a more favorable rPFS vs conversion from high to low ctDNA: TALA + ENZA, hazard ratio (HR) 95% confidence interval (CI), 0.45 (0.29–0.71), P=0.0003; PBO + ENZA, 0.34 (0.23–0.52), P<0.0001). Conclusions: High ctDNA burden at BL was negatively prognostic, and ctDNA conversion from high to low at WK 9 was prognostic of improved rPFS in TALAPRO-2. Limitations included not all clinical trial sites were able to perform ctDNA collection and most samples were below the limit of quantification. These results support the broad prognostic utility of ctDNA burden in mCRPC. Clinical trial information: NCT03395197 . Table: see text