Impact Of Treatment Sequence In Metastatic Castration-Resistant Prostate Cancer (Mcrpc) On Outcome In A Prospective Cohort Study.

264 Background: Abiraterone (Abi), enzalutamide (Enza) and docetaxel (Doc) are all valid first-line (1L) mCRPC treatment options. Evidence suggests a degree of cross-resistance between agents, which may impact the efficacy of subsequent lines of therapy. Evidence on the optimal treatment sequence is lacking. Methods: We evaluated the outcome of patients (pts) treated with 1L Doc, Abi or Enza in the prospective PROREPAIR-B cohort study. We assessed the impact of 1L treatment option (Doc vs Abi/Enza) on overall survival (OS), progression-free survival (PFS) to 1L-therapy (PCWG2) and PFS2 (time from initiation of 1L-therapy to progression on second-line [2L] therapy). Uni- (UV) and multivariable (MV) cox-regression models were used. MV model covariates included age (≥75 years), local therapy, Gleason Score, metastases at diagnosis, visceral metastases, ALP (≥ULN), LDH (≥ULN), haemoglobin (≤LNL), albumin (≤LNL) and ECOG PS. Results: 406 pts received 1L-Doce (N=188) or Abi/Enza (N=218). Pts receiving Doc were younger (p=0.002), had higher rates of visceral metastases (17.6 vs 8.7%; p=0.008), ALP (52.1% vs 40.4%; p=0.018), LDH (48.1% vs 31.2%; p<0.001) and lower Hb (7.4 vs 2.8%; p=0.029) and albumin (11.3 vs 4.6%). PFS was higher in pts receiving 1L-Abi/Enza (10.8 vs 8.3 months; HR:0.5; p<0.001). Pts receiving 1L Abi/Enza had higher rates of radiographic progression (88.4 vs 80%; p=0.032). 123/188 pts treated with 1L-Doc received 2L-Abi/Enza: 30 received other 2L and 35 had not started 2L. 111/216 pts treated with 1L-Abi/Enza received 2L-Doc, 26 were started on other 2L and 79 had not initiated 2L. A significant difference between pts treated with initial Abi/Enza vs Doc was observed in PFS2 (20.6 vs 16.6m; HR:0.78; p=0.006) but not OS (31.3 vs 29.9 m; HR:1.05; p=0.725). Choice of first-line agent was not associated with OS in the MV model. Conclusions: Despite longer PFS to 1L and PFS2 in pts treated with initial Abi/Enza, no differences in OS were observed between treatment sequences starting with Doc vs Abi/Enza. Pts treated with 1L-Doc had worse baseline prognostic features. Molecular stratification may enable biomarker-driven patient selection to optimize benefit in pts. (NCT03075735).