Venous and arterial thromboembolic events with immune checkpoint inhibitors: A systematic review

Background: Venous (VTEs) and arterial thromboembolic events (ATEs) are causes of morbidity, disability, mortality, and increase in treatment costs in cancer patients. The risk associated with immune checkpoint inhibitors (ICIs) has not yet been clarified. The primary objective of this systematic review was to evaluate the incidence of VTEs and ATEs in patients treated with ICIs as single agents or in combination with other treatments. Material and methods: Data from retrospective and prospective studies were selected from PubMed, EMBASE, SCOPUS, and The Cochrane Library from inception up to May up to 21st May 2020. All studies had to be in English and use human study participants. The studies were eligible if they provided a number (or rate) of VTEs and ATEs and the size of the population included. The PRISMA guidelines were followed. The data on the incidence of VTEs and ATEs were extracted for each arm, analyzed using random-effects models, and reported as weighted measures. Results: A total of 20,273 patients from 68 studies were included (median follow-up ranged from a few months up to three years). Overall, there were 390 VTEs and 59 ATEs, with incidence rates of 2.7% (95%CI 1.8%–4%) and 1.1% (95%CI 0.5%–2.1%), respectively. The rate of pulmonary embolism was 1.6% (95%CI 0.7%–3.2%) and deep venous thrombosis was 2.7% (95%CI 1.4%–5.4%). In studies where ICIs were administered with chemotherapy, rates of VTEs were similar to ICI alone arms (2.8% vs 2.5%). The rate of stroke and myocardial infarction were 1.1% (95%CI 0.65%–1.45%) and 0.7% (95%CI 0.15%–1.15%), respectively. In randomized trials, compared with non-ICIs containing arms (e.g. chemotherapy), the relative risk (RR) of VTEs due to ICIs was similar (RR 1.08, 95%CI 0.6–1.9; P =.79). Conclusions: Thromboembolic events associated with ICIs are relatively rare in cancer patients with an advanced stage of the disease. However, in randomized studies, their incidence is similar to control arms, suggesting that the contributory role of ICIs to the thromboembolic risk in many cancer settings is small