Screening for atrial fibrillation in cancer: a randomised controlled clinical trial (SARIC)

Abstract

Patients with cancer are at increased risk of atrial fibrillation (AF); however, to date, no randomised controlled trials have implemented and evaluated a systematic AF screening strategy in ambulatory patients with cancer, representing an important evidence gap in cardio-oncology. To implement and evaluate a single-timepoint AF screening strategy using a mobile electrocardiogram (ECG) device in ambulatory patients with cancer, and to assess its impact on AF detection, anticoagulation initiation, and AF prevalence. We conducted an open-label, randomised controlled trial among ambulatory patients aged ≥65 years with a current or prior cancer diagnosis and no history of AF. Participants were randomised 1:1 to either a screening strategy consisting of a single 30-second handheld mobile ECG recording or to usual care without systematic screening. The primary outcome was newly diagnosed AF during follow-up. Secondary outcomes included initiation of oral anticoagulation following AF detection, and the prevalence of pre-existing AF among all eligible patients. Group comparisons were performed using Fisher’s exact test. Among 1,012 patients screened for eligibility, 480 were randomised (screening: n=242; usual care: n=238). The median age was 72 years (interquartile range 68.0–76.3), and 73.5% were female. During follow-up, AF was newly diagnosed in 5 patients (2.1%) in the screening group and 4 patients (1.7%) in the usual care group (p=1.00). Initiation of oral anticoagulation occurred in 1.7% of patients in both groups (p=1.00). Pre-existing AF was identified in 19% of the overall population. In this first randomised clinical trial to implement and evaluate AF screening in ambulatory patients with cancer, a single-timepoint mobile ECG strategy was feasible but did not increase AF detection or anticoagulation initiation compared with usual care. Despite a high underlying prevalence of AF, brief single-timepoint screening was insufficient to identify new AF events, defining an important limitation of this approach in cardio-oncology. These findings provide prospective evidence to inform the design of future screening strategies, including longer-duration or repeated rhythm monitoring, tailored to ambulatory cancer populations.