Comparative efficacy of docetaxel versus novel hormonal agent in de novo metastatic hormone-sensitive prostate cancer: Real-world data curated by deidentified chart abstraction.

e17047 Background: The addition of docetaxel or a novel hormonal agent (NHA), such as abiraterone acetate or enzalutamide, has become standard of care for mHSPC, but prospective data for comparative efficacy remains limited. Clinical variables abstracted through natural language processing of free text from patient charts can provide real-world data to answer important clinical questions. Methods: Using an innovative data abstraction process, we retrospectively identified men with de novo mHSPC within deidentified charts from the Mount Sinai Health System treated with either docetaxel or a NHA between January 1, 2014 and April 30, 2019. Dates were chosen to reflect the timeline for which these therapeutic agents received regulatory approval for mHSPC, but also to allow sufficient clinical follow up after treatment initiation. Primary outcome of failure-free survival (FFS), defined as the time to next treatment, was assessed using the Kaplan-Meier method and multivariable Cox proportional hazards models. We additionally performed multivariable analysis to evaluate the prognostic significance of post-treatment PSA nadir on FFS. Results: A total of 94 de novo mHSPC patients that received either docetaxel or an NHA were identified using proprietary Sema4 data abstraction process: 52 received docetaxel, 42 received an NHA. Use of an upfront NHA in mHSPC was associated with a significantly longer FFS compared to docetaxel (20.7 vs. 10.1 months, p = 0.023). While NHAs were associated with a significantly longer FFS than docetaxel in patients with high metastatic burden of disease (25.12 vs. 9.63 months, p = 0.014), this was not observed in low-volume disease (20.71 vs. 26.5 months, p = 0.9). In multivariable model analysis adjusting for age, baseline PSA, and metastasis burden, docetaxel remains independently associated with worse FFS compared to NHA (HR 1.96, 95% CI 1.12−3.45, p = 0.019). Irrespective of docetaxel or NHA use, lower post-treatment PSA nadir levels were associated with improved FFS (58.95 vs. 11.57 vs. 9.4 months for PSA ≤0.2, 0.2-0.4, > 0.4ng/ml, respectively; p < 0.001). Conclusions: Clinical variables abstracted from deidentified clinical documentation can efficiently provide relevant and reliable data upon which clinical research can be based. Comparative analysis of real-world data demonstrates superior FFS in de novo mHSPC treated with a NHA as compared to docetaxel. The depth of PSA response holds prognostic value for mHSPC outcomes, regardless of treatment used.