Direct oral anticoagulants for cancer‐associated venous thromboembolisms: a systematic review and network meta‐analysis

Abstract

Background: Several recent randomised controlled trials (RCT) have investigated the use of direct oral anticoagulants (DOAC) in the treatment of malignancy-associated venous thromboembolism (VTE).Aims: This meta-analysis combines all RCT data to determine the risks of recurrent VTE and bleeding with DOAC in patients with malignancy-associated VTE compared with low-molecular-weight heparin (LMWH). Methods:The study followed PRISMA guidelines. MEDLINE, EMBASE and CENTRAL were systematically searched from inception to 1 April 2020. References of reviews and relevant conference proceedings were searched by hand. Two authors independently evaluated study eligibility, extracted data and assessed risk of bias. Direct and indirect meta-analyses were performed.Results: In four RCT with low risk of bias (2907 patients), high certainty evidence suggested that DOAC had a 37% reduction in risk of recurrent VTE compared with LMWH (direct pooled risk ratio (RR) 0.63; 95% confidence interval (CI) 0.44-0.91; I 2 = 28%). No significant difference was observed in the risk of major bleeding with DOAC compared with LMWH (RR 1.31; 95% CI 0.83-2.07; I 2 = 22%; moderate certainty evidence), including in patients in gastrointestinal and genitourinary malignancy. An increased risk of combined major or clinically relevant non-major bleeding was seen with DOAC (RR 1.52; 95% CI 1.09-2.12; I 2 = 51%; low certainty evidence). Apixaban had the highest probability of being ranked the most effective and least bleeding risk among the DOAC.Conclusion: DOAC are effective in treating malignancy associated VTE; however, caution is required in patients with high risk of bleeding. Apixaban had lower risk of bleeding compared to other DOAC in this population.

An epidemic of redundant meta‐analyses

“…At the time the search was ended (October 6, 2020), 46 records were identified, of which 26 were discarded on title and abstract ( n = 24) or full‐text ( n = 2). The 20 remaining meta‐analyses were published in nonpredatory peer‐reviewed journals: five studies pooled the data of two RCTs, 6‐10 four studies pooled the data of three RCTs, 11‐14 and 11 studies pooled the data of four RCTs 15‐25 . The main characteristics of the 20 meta‐analyses are described in Table 1 and the results of the 20 meta‐analyses for efficacy and safety endpoint are shown Figure 1.…”

“…The 20 remaining meta-analyses were published in nonpredatory peer-reviewed journals: five studies pooled the data of two RCTs, 6-10 four studies pooled the data of three RCTs, [11][12][13][14] and 11 studies pooled the data of four RCTs. [15][16][17][18][19][20][21][22][23][24][25] The main characteristics of the 20 meta-analyses are described in Table 1 and the results of the 20 meta-analyses for efficacy and safety endpoint are shown Figure 1. Within each category of meta-analysis (two, three, or 4 RCTs), the results were similar regarding the risks of VTE recurrence (differences in pooled relative…”

Anticoagulation Practice in Patients with Cancer-Associated Thrombosis: Insights from GeCAT, a German Prospective Registry Study

“…These trials indicated that oral treatment with a factor Xa inhibitor may be an alternative to subcutaneous LMWH for selected CAT patients, and this statement was eventually adopted in the current guidelines. However, DOACs may be associated with an increased risk of bleeding in patients with gastrointestinal and potentially genitourinary cancer [14]. In fact, there seems to be a higher risk for mucosal bleeding.…”

“…In fact, there seems to be a higher risk for mucosal bleeding. Moreover, potential interactions need to be checked before use of DOACs [14]. These and similar precautions for the use of DOACs are included in several guidelines including those of the ASCO [15] and the National Comprehensive Cancer Network [16].…”

Direct oral anticoagulants in the treatment of venous thromboembolic disease associated with cancer. Evidence and recommendations

“…Respecto de las hemorragias graves, los AOD presentaron menor incidencia en dos ensayos (ambos con apixabán) y mayor incidencia en los otros dos, si bien globalmente las diferencias no fueron significativas en 3 de los 4 ensayos (rivaroxabán en un ensayo y apixabán, en dos) (Tabla III). Son aspectos particulares de estas RS/M que las hemorragias son más frecuentes en los cánceres gastrointestinales (50) y que apixabán presenta el mejor perfil de seguridad (52,54).…”