Direct versus conventional anticoagulants for treatment of cancer associated thrombosis: a pooled and interaction analysis between observational studies and randomized clinical trials

Background: There are emerging observational studies (OSs) to assess real-world comparative effectiveness and safety of direct oral anticoagulants (DOACs) in cancer associated thrombosis (CAT). We conducted a pooled and interaction analysis to compare the treatment effect estimates of DOACs between OSs and randomized controlled trials (RCTs). Methods: We systematically searched PUBMED, EMBASE and Cochrane Library for OSs and RCTs that reported recurrent venous thromboembolism (VTE) and/or major bleeding events in CAT patients receiving DOACs and conventional anticoagulants [warfarin or low molecular-weight heparins (LMWHs)]. Relative risks (RRs) for OSs and RCTs were calculated using random-effects models separately, and interaction analyses were afterward applied to assess the comparability between OSs and RCTs. Results: Baseline characteristic was comparable between identified 10 OSs (35,142 patients) and 8 RCTs (2,602 patients). Overall, no significant difference of treatment effect estimates between OSs and RCTs was detected (P interaction : 0.42 for recurrent VTE; P interaction : 0.38 for major bleeding). DOACs significantly decreased the risk of recurrent VTE compared with conventional anticoagulants in CAT patients (RR: 0.74, 95% CI: 0.63–0.86, I 2 : 0% for OSs; RR: 0.65, 95% CI: 0.49–0.86; I 2 : 0% for RCTs), without increasing major bleeding risk (RR: 0.90, 95% CI: 0.76–1.07, I 2 : 24.0% for OSs; RR: 1.17, 95% CI: 0.72–1.88, I 2 : 26.2% for RCTs). Whereas, increased risk of gastrointestinal bleeding (GIB) was found with DOACs versus conventional anticoagulants in CAT patients (RR: 2.77, 95% CI: 1.35–5.68, I 2 : 0% for RCTs). Analyses of subgroups, based on comparators and follow-up duration, did not significantly affect results. Conclusions: In this study, effectiveness and safety of DOACs versus conventional anticoagulants in CAT from OSs are in agreement with those from RCTs, confirming a low risk of recurrent VTE and similar risk of major bleeding in CAT patients receiving DOACs.