Distal Resection Margin Collagen Score to Predict Major Low Anterior Resection Syndrome After Neoadjuvant Chemoradiotherapy in Rectal Cancer.

Abstract

Low anterior resection syndrome (LARS) is common after neoadjuvant chemoradiotherapy (nCRT) and sphincter-preserving surgery for rectal cancer and is associated with poor quality of life. However, reliable tools to identify patients at high risk remain limited. This study aims to develop and validate a prediction model for major LARS in patients with rectal cancer after nCRT and sphincter-preserving surgery. A total of 315 consecutive patients between 2019 and 2021 were retrospectively enrolled (training cohort: 213; independent validation cohort: 102). A distal resection margin collagen score (CS_DRM) was derived from multiphoton imaging using least absolute shrinkage and selection operator (LASSO) logistic regression. A prediction nomogram incorporating CS_DRM and clinicopathologic factors was developed and evaluated for discrimination, calibration, and clinical utility. The CS_DRM was developed on the basis of eight features. Multivariable analysis revealed that the CS_DRM (odds ratio [OR] 3.57, 95% confidence interval [CI] 2.56-5.37), tumor distance from the anal verge, and time to stoma closure were independent predictors of major LARS. The CS_DRM-integrated nomogram showed good discrimination in the training cohort (area under the receiver operating characteristic curve [AUROC], 0.914, 95% CI 0.863-0.957) and validation cohort (AUROC 0.922, 95% CI 0.851-0.976). Compared with the traditional model, incorporating CS_DRM significantly improved discrimination in both the training cohort (AUROC 0.914 versus 0.631; p < 0.001) and validation cohort (AUROC, 0.922 versus 0.619; p < 0.001). The CS_DRM was associated with major LARS after nCRT and sphincter-preserving surgery. The CS_DRM-integrated model may support postoperative risk stratification for major LARS in patients with rectal cancer.