Beschreibung:
<jats:p> 223 </jats:p><jats:p> Background: The impact of number of lines of therapy on outcomes following docetaxel (D) in metastatic castration-resistant prostate cancer (mCRPC) is unclear. We examined outcomes with cabazitaxel (C) and/or abiraterone acetate (A), following D during a period when all three therapies were available. We previously reported that most patients received only two of these three therapies. Among patients who received all three, DCA was administered more commonly and exhibited better overall survival (OS) than DAC after controlling for prognostic factors. Here, we report the impact of number of lines of therapy following D. Methods: A retrospective analysis of the U.S. Oncology network electronic health records (EHR) was conducted of post-D patients with mCRPC who received C and/or A from April 2011 to May 2012. Median OS was analyzed by Kaplan-Meier method. Cox proportional hazard models were used to evaluate impact on OS of number of therapies administered, age, Prostate Cancer Working Group (PCWG2) subtype, Charlson comorbidity index, prostate-specific antigen (PSA), alkaline phosphatase, hemoglobin, narcotic use, and treatment duration. Results: Multivariate analysis showed significantly lower mortality in the three-drug group compared to the two-drug group (HR 0.209 95% CI: 0.092-0.476, p<0.05). 113 patients received three drugs (DCA=77, DAC=36) and 237 received two drugs (DA=183, DC=54). The three-drug cohort was significantly younger than the two-drug group (median age 69 vs. 73). Other significant covariates (p<0.05) for mortality were narcotic use (HR 2.010 [1.240-3.259]), PSA (HR 1.014 [1.001-1.027] and alkaline phosphatase (HR 1.001 [1.000-1.001]. Conclusions: In men with mCRPC receiving C and/or A post-D, patients receiving all three therapies were younger and exhibited significantly better OS after controlling for clinical factors. In those receiving only two therapies, there appeared to be no difference in outcomes for second-line C versus A. Given the favorable impact of receiving all three therapies, more frequent administration of DCA in the three-drug group and better OS for DCA compared to DAC, we hypothesize that DCA may be a more optimal sequence. These results are exploratory and prospective validation is necessary. </jats:p>