The effect of therapeutic anticoagulation on overall survival (OS) in men receiving docetaxel chemotherapy for metastatic castration-resistant prostate cancer (mCRPC).

Abstract

28 Background: Anticoagulants have been postulated to possess antitumor activity, although clinical data supporting this claim are conflicting. We sought to examine the effect of therapeutic anticoagulation on OS in men with mCRPC receiving first-line docetaxel chemotherapy. Methods: We retrospectively reviewed the records of 247 consecutive mCRPC patients who received first-line docetaxel chemotherapy between 1/1/1998 and 1/1/2010. Information on anticoagulant use, type of anticoagulant administered, indication for anticoagulation, and duration of anticoagulation were captured. Univariate and multivariable Cox proportional hazards regression models were developed to investigate the effect of anticoagulant use on OS. Results: In all, 29/247 men (11.7%) received anticoagulation (LMW heparin: 17/247; warfarin: 12/247). The indication was DVT in 15/247, PE in 9/247, and both DVT and PE in 5/247 men. In univariate analysis, anticoagulant use was associated with improved OS (any anticoagulant, HR 0.61 [95%CI 0.40–0.94] P=0.024; LMW heparin, HR 0.58 [95%CI 0.34–0.99] P=0.048; warfarin, HR 0.82 [95%CI 0.55–1.28] P=0.23). Median OS was 20.9 mo (with any anticoagulant) versus 17.1 mo (with no anticoagulant). In multivariable analysis, anticoagulant use remained a significant predictor of OS after adjusting for other prognostic factors (Table). Conclusions: Anticoagulant use is an independent predictor of OS in men with mCRPC receiving docetaxel. This finding is surprising given that the occurrence of venous thrombosis might be expected to negatively influence OS. If validated, these data may provide the impetus to explore the antitumor potential of anticoagulants in prospective clinical trials. [Table: see text]