In The Lancet<\/em>, Francis Couturaud and colleagues7<\/sup><\/a> report the findings of the multicentre, open-label, randomised RENOVE trial, in which they aimed to show non-inferiority of reduced-dose versus full-dose direct oral anticoagulants on efficacy, and, using hierarchical sequential testing, superiority of the reduced dose for safety and net clinical benefit. Patients at high risk of recurrent venous thromboembolism after initial treatment with full-dose anticoagulants for 6\u201324 months were randomly assigned to either reduced-dose or full-dose apixaban or rivaroxaban. During the study, the investigators noted a lower than anticipated rate of recurrent venous thromboembolism and decided to increase the sample size from 2060 to 2678 and follow-up length from 24 months to a median of 36 months.<\/p>\n\n\n\n Between November, 2017, and July, 2022, 1383 patients were assigned to the reduced-dose group and 1385 to the full-dose group in 50 French hospitals. The mean age of participants was around 62 years; ethnicity was not reported. The indications for extended anticoagulation were mainly a first episode of unprovoked venous thromboembolism (60\u00b77%) and multiple episodes of venous thromboembolism (33\u00b71%), explaining the under-representation of female participants in the trial (35\u00b70%). Furthermore, most included patients (85\u00b76%) had pulmonary embolism with or without deep vein thrombosis as the index event. The primary outcome (adjudicated symptomatic recurrent pulmonary embolism or proximal deep vein thrombosis) occurred in 19 of 1383 patients (three fatal) in the reduced-dose group (5-year cumulative incidence 2\u00b72% [95% CI 1\u00b71\u20133\u00b73]) and in 15 of 1385 patients (three fatal) in the full-dose group (5-year cumulative incidence 1\u00b78% [0\u00b78\u20132\u00b77]). The adjusted hazard ratio (HR) for reduced-dose anticoagulation was 1\u00b732 (95% CI 0\u00b767\u20132\u00b760; absolute difference 0\u00b740% [95% CI \u20131\u00b705 to 1\u00b785]; p=0\u00b723 for non-inferiority). With regard to the key secondary outcome on safety, major or clinically relevant non-major bleeding occurred in 96 patients in the reduced-dose group (5-year cumulative incidence 9\u00b79% [95% CI 7\u00b77\u201312\u00b71]) and in 154 patients in the full-dose group (5-year cumulative incidence 15\u00b72% [12\u00b78\u201317\u00b76]), resulting in an adjusted HR for reduced-dose anticoagulation of 0\u00b761 (95% CI 0\u00b748\u20130\u00b779). Major bleeding occurred in 15 patients in the reduced-dose group (two fatal; 5-year cumulative incidence 2\u00b71% [95% CI 0\u00b77\u20133\u00b75]) and in 38 patients in the full-dose group (three fatal; 5-year cumulative incidence 4\u00b70% [2\u00b77\u20135\u00b73]). There were no differences in all-cause mortality or arterial events between the treatment groups.<\/p>\n\n\n\n Strengths of the study are the rigorous design and execution of the trial, with a long median duration of follow-up, low rate of loss to follow-up, and good adherence to medication. Limitations are the open-label study design, which was mitigated by adjudication by assessors who were masked to treatment allocation, and lack of power for relevant subgroup analyses.<\/p>\n\n\n\n What are the implications of these findings? In our view, although the hypothesis of non-inferiority of reduced-dose versus full-dose direct oral anticoagulants was not met statistically, the 5-year cumulative incidence of recurrent venous thromboembolism of around 2% in either treatment group is reassuringly low, and a potential inferiority of the reduced dose lacks clinical relevance in the context of these low absolute risks. Therefore, we are interested in the benefits in terms of safety and net clinical benefit. Although Couturaud and colleagues7<\/sup><\/a> designed the RENOVE trial to allow testing for superiority of safety only if non-inferiority for efficacy was met, we believe that the data on bleeding speak for themselves. The RENOVE trial showed a risk reduction in major and clinically relevant non-major bleeding of 39% with the reduced dose, with clinically relevant differences in 5-year cumulative incidences between the full dose and the reduced dose. We hypothesise that the RENOVE trial was able to show this difference due to the median duration of follow-up of 36 months, in contrast to the previously executed trials that followed up patients for 12 months.4<\/a>, 5<\/a><\/p>\n\n\n\n
![\"\"\/](\"https:\/\/ars.els-cdn.com\/content\/image\/1-s2.0-S0140673625000996-fx1_lrg.jpg\")
<\/figure>\n","protected":false},"excerpt":{"rendered":"