Interval-Based Estimation of Colorectal Neoplastic Polyp Growth and Its Value in Polyp Screening: A Retrospective...

BackgroundCurrent post-polypectomy surveillance guidelines rely primarily on baseline polyp pathology, lacking integration of metabolic factors and quantitative growth estimation for individualized interval stratification. This study sought to translate cancer growth modeling concepts into clinically actionable insights for surgical practice.MethodsThis retrospective cohort study enrolled 795 patients who underwent polypectomy and at least one follow-up colonoscopy. Independent prognostic factors were identified and a nomogram was constructed using Cox proportional hazards regression and LASSO Cox regression with 10-fold cross-validation. Interval-based polyp progression was quantified via multivariable linear regression (ordinary least squares, robust regression, and log-transformed models). Model performance was evaluated by receiver operating characteristic (ROC), calibration and decision curve analyses with internal validation.ResultsOver a median follow-up of 13 months, the crude recurrence rate was 80.8%. Abnormal HDL-C, baseline polyp count, and baseline polyp volume were independently associated with shorter time to recurrence in the multivariable Cox model. The model showed moderate and consistent discrimination, with good calibration and positive net benefit across moderate-to-high-risk thresholds. In exploratory progression analyses, baseline polyp count and surveillance interval were robustly positively associated with recurrent polyp number and maximum diameter across all model specifications, while associations of abnormal HDL-C were inconsistent and not significant for diameter outcomes in log-transformed models.ConclusionsIntegrating metabolic biomarkers with endoscopic data may modestly improve recurrence risk stratification, though this added value is limited by inconsistent associations and potential confounding. Prospective studies with adjustment for medication use and metabolic syndrome components are needed for validation.