Blood-Based Test for Colorectal Cancer Deemed Comparable to Other Non-Invasive Options

Published on: 

The sensitivity for colorectal cancer detection of the blood-based test was 83%, meeting the co-primary sensitivity endpoint.

A blood-based test demonstrated an 83% sensitivity for the detection of colorectal cancer and 90% specificity among an average risk screening population, results of which were comparable to currently available non-invasive screening options, according to data presented at Digestive Disease Week (DDW) 2023.1 Investigators noted that adherence adjusted sensitivity for the blood-based testing strategy may be much higher than other colorectal cancer screening strategies, which could indicate a potential for this test to significantly improve adherence to these screenings.

“Colorectal cancer screening rates for average-risk individuals remain sub-optimal due to poor adherence with the currently available strategies,” wrote lead investigator Daniel Chung, MD, Massachusetts General Hospital and Harvard Medical School, and colleagues. “A blood-based test with performance comparable to available options may dramatically improve screening adherence. We evaluated the performance of a cell-free DNA (cfDNA) blood-based colorectal cancer screening test in an average-risk population.”

The ECLIPSE trial enrolled patients at an average risk for colorectal cancer who were aged ≥45 years, presenting for screening colonoscopy, from 265 clinical sites in the United States between October 2019 and September 2022. Patients were required to provide a whole blood sample prior to initiation of colonoscopy bowel preparations. Colonoscopy outcomes were categorized via the most advanced lesion: colorectal cancer, advanced precancerous lesion (APL), non-APL, negative or no precancerous lesion, or incomplete procedure. The primary endpoints were sensitivity for colorectal cancer detection and the specificity for the absence of colorectal cancer or APL, compared with the reference colonoscopy. The secondary endpoint was APL sensitivity.

A cohort of 10,258 participants were ultimately included, of which 7,861 patients met all eligibility criteria. The mean age of patients was 60 years, 54% were female, and most (79%) were White.

The sensitivity for colorectal cancer detection was 83% (n = 54/65, 95% confidence interval [CI] 72–90%) which met the co-primary sensitivity endpoint (lower bound of 95% CI > 65%). The sensitivity for cancers will full staging information available was 92% (n = 48/52). Sensitivity in Stage I – III colorectal cancer was 81%, with sensitivity by stage of 55% for Stage I, 100% for Stage II, 100% for Stage III, and 100% for Stage IV.

The specificity for the absence of colorectal cancer or APL was 90% (5,982/6,680; 95% CI 89–90%), which met the primary specificity endpoint (lower bound of 95% CI > 85%). Lastly, the specificity for negative or no precancerous lesions was 90% (4,057/4,514; 95% CI 89–91%) and sensitivity for detection of APL was 13% (147/1,116; 95% CI 11-15%).


  1. Chung D, Mason Gray M, Greenson J, Gupta S, et al. CLINICAL VALIDATION OF A CELL-FREE DNA BLOOD-BASED TEST FOR COLORECTAL CANCER SCREENING IN AN AVERAGE RISK POPULATION. Abstract presented at Digestive Disease Week (DDW) 2023 Annual Meeting. Chicago, IL. May 6-10, 2023.