Mutational and DNA fragment analysis may help in the triage of FIT+ patients enrolled in colorectal cancer screening programs.

Abstract

e15546 * F. Pepe made an equal contribution Background: Mutational and DNA fragment analysis have already been investigated for a possible role in the detection of colorectal cancer. This pilot study, conducted at ASL NA 3 SUD (Naples, Italy) with the University of Naples Federico II, attempts to understand their possible role in the detection of Advanced Adenomas+ (AA+) and their potential ability to triage subjects with a positive FIT test for referral to colonoscopy. Methods: 160 FIT+ subjects of both genders, aged 50-74, participating in a colorectal cancer screening program were enrolled in a clinical trial focusing on a novel biomarker for colorectal cancer diagnostics. Of these, 14 subjects had disease (12 AAs and 2 colorectal cancers, from now on “CRC”). After ethical committee’s approval and release of informed consents, a 20 ml whole-blood sample was collected from each subject. Samples were centrifuged following standard procedures and stored at -80°C. DNA extraction was later performed using a Minielute kit (Qiagen). Mutational and DNA fragment analysis were performed using a qPCR-based technology: the ColoScape™ 2.5 kit (DiaCarta Inc, Pleasanton, CA). A logistic regression was run to determine the probability of disease in subjects showing positivity for predictors. Estimates of sensitivity, specificity, PPV, NPV (with 95% C.I.), PLR, NLR and Youden’s Index were also obtained. Results: The logistic regression showed statistical significance, with a p-value < 0.0001, that is, the logistic regression model with provides a better fit than the model without the independent variable. ColoScape identified 12 subjects with disease out of 14, for a sensitivity of 86% (67%,100%); both CRCs were picked up by ColoScape together with 10 AAs. 2 AAs were missed. Out of 146 healthy subjects, 107 were successfully identified as negatives, for a specificity of 73% (66%, 80%). PPV was 24% (12%, 35%). NPV was 98% (96%, 100%). PLR, NLR and Youden’s Index were 3.21, 0.19 and 0.59 respectively. Conclusions: This pilot study was intended to provide performance indicators to design a larger population study that is currently underway. These preliminary findings suggest that liquid biopsy-based mutational and DNA fragment analysis can play a role in selecting subjects with a higher risk of returning a positive colonoscopy, thus potentially improving the quality of colorectal cancer screening programs and patients’ experience.