NHS-Galleri: Primary results from a randomised controlled trial to assess the clinical utility of a multi-cancer early detection (MCED) test in population screening.

Abstract

LBA100 Background: MCED blood tests can detect a shared cancer signal from circulating cell-free DNA. NHS-Galleri (NCT05611632) is a randomised controlled trial in England of 142,924 enrolled participants evaluating the clinical utility of an MCED test (Galleri) for annual screening of asymptomatic individuals aged 50–79 yrs. Methods: Peripheral blood samples were taken at up to 3 annual visits (Y0, Y1, Y2). After the Y0 blood draw, participants were randomised 1:1 to the intervention (I; blood tested by MCED test) or control (C; blood not tested) arms. I arm participants with a cancer signal detected MCED test result were referred into the NHS for diagnostic workup via appropriate urgent suspected cancer pathways. As agreed with NHS England, clinical utility was assessed by reduction in late stage (Stage III/IV [primary objective] and Stage IV [key secondary objective]) cancer incidence in I vs C arm ~3 yrs after the last participant’s randomisation. Results: Trial arms were well balanced. The primary endpoint of statistically significant Stage III/IV reduction in 12 prespecified cancers was not met (706 vs 688; incidence rate ratio [IRR] 1.03; Table). In these 12 cancers, Stage III/IV decreased from the prevalence screening round (IRR 1.19) to incidence rounds (0.95; 0.88). A 14% reduction in Stage IV cancers (342 vs 397) was observed after 3 yrs of screening (Y0: 9%; Y1: 22%; Y2: 26%). Stage I/II cancers increased by 16% (647 vs 559; relative risk [RR] 1.16 [1.03, 1.30]) after 3 yrs of screening. Similarly, Stage I-III cancers increased by 19% (1007 vs 846; RR 1.19 [1.09, 1.30]). Across all cancer types, 3637 and 3400 were diagnosed in I and C arms, respectively, over 3 yrs of screening. MCED quadrupled the number of screen-detected cancers (1173 vs 290). MCED reduced clinically-detected cancers by 21% (2464 vs 3110) and emergency presentations by 21% (225 vs 286). Aggregate test performance (99.55% specificity, 52.0% positive predictive value, 92.5% top-two and 87.0% top-one cancer signal origin accuracy) was consistent with prior studies and real-world evidence. There were 381 and 333 related adverse events (AEs) in I and C arms, respectively, and no related serious AEs. Conclusions: Although the primary endpoint was not met, adding annual MCED testing to standard-of-care cancer screening substantially increased screen-detected cancers and reduced Stage IV cancers and emergency presentations. This evidence combined with the favourable safety profile and robust performance suggests that integrating MCED into population screening programs may help reduce late stage cancer burden. Clinical trial information: NCT05611632 . Late-stage IRR (I/C) of 12 prespecified cancers. Stage III/IV Stage IV Y0 1.19 (95% CI: 0.98, 1.43) 0.91 (0.71, 1.18) Y1 0.95 (0.77, 1.17) 0.78 (0.57, 1.06) Y2 0.88 (0.73, 1.07) 0.74 (0.57, 0.95) Overall 1.03 (0.92, 1.14)p=0.6324 0.86 (0.744, 0.998)