Plasma cfDNA role in detecting cardiovascular, metabolic and other diseases in a FIT+ population enrolled for a colorectal cancer screening program.

Abstract

e15541 * Francesco Pepe made an equal contribution to the study Background: Cell-free DNA (cfDNA) has long been established as a useful tool in a variety of clinical settings, including infectious, auto-immune, cardiovascular and neoplastic diseases. In a study that is conducted at ASL NA 3 SUD (Naples, Italy) with the University of Naples Federico II, cfDNA was used to conduct mutational analysis using a novel biomarker for colorectal cancer. In this study, the observation of cfDNA value has suggested its potential use as a disease detector. This abstract reports findings of a secondary analysis in the study. Methods: 667 FIT+ subjects, aged 50-74, participating in a colorectal cancer screening program enrolled in a clinical trial focusing on the role of a novel biomarker in colorectal cancer diagnosis. The subjects' medical history included cardiovascular diseases (CVD) and diabetes primarily, collectively named as Concomitant Diseases (CD). 398 patients had at least one CD, while 269 had none. A whole blood sample was obtained from each study subjects and total plasma cfDNA was measured using QuantiDNA™ test, manufactured by DiaCarta Inc. ROC curve was run to assess diagnostic efficacy and determine cutoff points of cfDNA for detecting the CD and specifically cardiovascular ones. Logistic regression was used to determine the probability of disease in subjects with an elevated cfDNA concentration. Adjusted estimates were obtained to assess the potential confounding effect of CD vs colorectal neoplasia (CN). Results: The area under the curve (AUC) for cfDNA as a predictor of CD was 0.562 (95% C.I. 0.518, 0.607); the cutoff point was 13.44 ng/ml, with a sensitivity of 50% (95% C.I 45%, 55%) and a specificity of 63% (95% C.I. 57%, 69%). The logistic regression to assess the probability of CD showed an odds ratio (OR) for cfDNA of 1.6279 (95% C.I. 1.19, 2.234; p = 0.0023) that is, one unit increase in cfDNA will increase the odds of disease by ~63%. A specific assessment was made for cfDNA as a detector for CVD. The AUC was 0.567 (95% C.I. 0.523, 0.611), the cutoff point was 12.8 ng/ml, with a sensitivity of 55% (95% C.I. 50%, 60%) and a specificity of 57% (95% C.I. 51%, 63%). The logistic regression indicated an OR of 1.681 (95% C.I. 1.2378, 2.2876; p < 0.001) meaning that a one unit increase in cfDNA will increase the odds of disease by ~68.1%. Lastly, the logistic regression of cfDNA as a predictor of CN adjusted for CD showed a change in slope > 10% from the unadjusted form, suggesting a potential confounding effect. Conclusions: Although the study was not designed to assess the validity of cfDNA as a detector of diseases, these secondary findings suggest that cfDNA can be used as a disease detector for CD including CVD, which encourages further studies. Attention should be directed towards potential confounding effects, such as from CD or CN since cfDNA can be elevated by several different conditions.