SCREESCO Trial Compares Colonoscopy, FIT, and Usual Care

Key Takeaways

• In 278,280 adults aged 60 years, randomization assigned once-only colonoscopy, two rounds of low-cutoff two-stool FIT, or usual care, and participation reached 35% and 55% among invited groups.
• Overall colorectal cancer incidence was similar across groups, while stage I–II diagnoses were more frequent and stage III–IV disease was less frequent in the intervention arms, especially with FIT.
• Gastrointestinal and cardiovascular events were slightly more frequent during the first year in intervention groups, and colorectal cancer mortality remains planned for reporting through 31 December 2030.

After a median 4.8 years, the SCREESCO trial showed more stage I–II and fewer stage III–IV colorectal cancers after screening invitations than after usual care in Swedish adults aged 60 years. Overall colorectal cancer incidence rates were similar across the randomized comparisons in this early readout. The pattern was consistent with a stage shift in both intervention groups, with the clearest reduction in later-stage disease in the FIT comparison. The trial compared once-only colonoscopy, two rounds of low-cutoff two-stool FIT, and usual care. Colorectal cancer mortality was not assessed in this diagnostic-phase analysis.

Investigators conducted an individual randomized block trial in a screening-naive, population-based setting across 18 of 21 Swedish regions. Eligible adults were aged 60 years or turned 60 during randomization, with exclusions for prior colorectal cancer, anal cancer, or NordICC participation. The analyzed cohort included 278,051 unique individuals, with 31,113 in the primary colonoscopy arm, 60,267 in the FIT arm, and 186,671 controls, including 120,521 used for the FIT comparison. Invitees were offered once-only colonoscopy or two rounds of two-stool FIT at a 10 μg hemoglobin per g feces cutoff versus usual care, and analyses used intention to screen. Baseline demographic variables and comorbidity history were balanced between each intervention arm and its controls.

Maximum follow-up reached 6.9 years, and screening uptake was lower for colonoscopy invitations than for FIT invitations. Non-participants tended to have lower educational attainment, were more often born outside Sweden, and had slightly more comorbidity. Overall colorectal cancer incidence was 107.9 versus 99.9 per 100,000 person-years with colonoscopy, yielding an IRR of 1.08 with a 95% CI of 0.91 to 1.28. In the FIT comparison, rates were 96.0 versus 103.9 per 100,000 person-years, with an IRR of 0.92 and a 95% CI of 0.81 to 1.05. Among cancers diagnosed in the intervention arms, 32% in colonoscopy and 38% in FIT were screen detected, and these incidence patterns did not establish a mortality effect.

Stage I–II colorectal cancer was diagnosed more often than in controls, with colonoscopy IRR 1.38, 95% CI 1.09 to 1.74, and FIT IRR 1.19, 95% CI 0.99 to 1.43. Stage III–IV disease was less frequent, with colonoscopy IRR 0.86, 95% CI 0.67 to 1.11, and FIT IRR 0.71, 95% CI 0.58 to 0.86. In this diagnostic-phase analysis, gastrointestinal and cardiovascular events were slightly higher during the first year, then became more similar, although venous thromboembolism and some gastrointestinal bleeding outcomes remained higher with FIT. Serious adverse events directly linked to screening colonoscopies had been reported previously at 0.2%, including two bowel perforations and 15 major bleedings. Colorectal cancer mortality remains planned for later reporting through 31 December 2030.