Work

Clinical, endoscopic, and pathological features of synchronous multiple early-stage colorectal cancer: a two-center retrospective case–control study

Year: 2026

Type: article

Abstract: Objectives: To characterize the clinical, endoscopic, and pathological features of synchronous multiple early-stage colorectal cancer (sMECRC) and develop a prediction model to identify patients at risk of synchronous lesions at index colonoscopy. Methods: We retrospectively analyzed 1,081 consecutive patients with early-stage colorectal cancer who underwent endoscopic treatment at Peking University Third Hospital (2013-2024; development cohort) and 161 patients treated at Peking University International Hospital (2016-2024; external validation cohort). A prediction model was developed using Firth penalized logistic regression, internally validated using 1,000 bootstrap resamples, and externally validated. Sensitivity analyses were also performed. Results: Of the 1,081 patients in the development cohort, 47 (4.3%) had sMECRC. Compared with patients with a single lesion, those with sMECRC had more concomitant adenomatous polyps (median, 5 vs. 2), larger index lesions (2.50 vs. 2.00 cm), a higher rate of FOB positivity (72.2% vs. 46.5%), and a higher frequency of index lesions in the descending colon (17.0% vs. 4.7%). Independent predictors of sMECRC were positive FOB [odds ratio [OR], 2.441; 95% confidence interval [CI], 1.157-5.150], number of adenomatous polyps (OR, 1.166 per polyp; 95% CI, 1.094-1.244), index lesion size (OR, 1.247 per cm; 95% CI, 1.099-1.416), and descending colon location (OR, 4.411; 95% CI, 1.627-11.959). The model showed moderate-to-good discrimination in the development cohort [apparent area under the curve (AUC),0.767;bootstrap optimism-corrected AUC, 0.748] and had a Brier score of 0.040, although calibration was suboptimal. External validation in an independent cohort of 161 patients, including 16 with sMECRC, confirmed good discrimination (AUC, 0.864; 95% CI, 0.778-0.950), with improved calibration after intercept-and-slope recalibration. Significant correlations were found between index and second lesions for diameter, Paris classification, depressed morphology, depth of invasion, and histological differentiation. Conclusion: FOB positivity, a greater adenomatous polyp burden, larger index lesion size, and an index lesion located in the descending colon independently predicted sMECRC.Using routine clinical variables, the model exhibited moderate to good discrimination in the development cohort and preserved good discrimination on external validation, but required recalibration for reliable absolute risk estimates. Concordance between the characteristics of index and second lesions may help guide more targeted inspection of the remaining colon.

Source: Frontiers in Medicine

DOI: 10.3389/fmed.2026.1929117