Abstract Background and aims Patients with active cancer are at increased risk of ischaemic stroke. Smaller studies of endovascular thrombectomy (EVT) in such stroke patients indicate worse clinical outcomes despite similar recanalization rates as patients without active cancer. However, evidence from larger multicentre cohorts is lacking. Methods We used data from the EVA-TRISP registry comprising stroke patients treated with EVT (11 centres, 8 countries, 2014−2023). Using multivariable regressions adjusting for prognostic variables and propensity score weighting we compared patients with active cancer to patients without cancer regarding early neurological improvement (ENI), defined as NIHSS improvement of -4 points or score 0−1 at 24h after EVT, successful recanalization (mTICI 2b−3), symptomatic intracranial haemorrhage (sICH), functional independence (modified Rankin Scale mRS 0−2) and death at 3 months. Results We included 526 EVT patients with active cancer and 6,894 without cancer. ENI (47.5 vs 45.3%, weighted OR wOR 1.05, 95%CI 0.87−1.26) and successful recanalization (78.8 vs 70.4%, wOR 0.86, 95%CI 0.68-1.08) did not differ significantly. However, patients with active cancer showed higher rates of sICH (7.2 vs 4.0%, wOR 1.88, 95%CI 1.30−2.70), less functional independence at 3 months (30.1 vs 44.0%, wOR 0.56, 95%CI 0.45−0.69) and higher death rates (42.8 vs 22.3%, wOR 2.60, 95%CI 1.14−3.17). Conclusions In this large observational multicentre study, recanalization rates and ENI after EVT were comparable in stroke patients with and without active cancer. Still, worse 3−month outcomes in patients with active cancer may have to be considered in EVT-decision making in such patients. Conflict of interest Malin Woock: nothing to disclose, Katarina Jood: nothing to disclose, Turgut Tatlisumak: nothing to disclose, Gianluca Costamagna: nothing to disclose, Christian Nolde: nothing to disclose, Mirjam Heldner: nothing to disclose, Nicolas Martinez-Majander: nothing to disclose, Ronen Leker: nothing to disclose, Anton Schmick: nothing to disclose and Joâo Pedro Marto: nothing to disclose.
Woock et al. (2026) studied this question.