270 Development of a Predictive Nomogram for Circumferential Resection Margin in Rectal Cancer Surgery

Abstract

OBJECTIVES/GOALS: A negative circumferential resection margin (CRM) after surgical resection of rectal cancer decreases local recurrence and increases overall survival. While MRI is used to predict this risk, there is no predictive model that incorporates clinical factors to predict the risk of CRM positivity. METHODS/STUDY POPULATION: Utilizing the National Cancer Database from 2010-2014, we performed a retrospective study evaluating factors predictive for positive CRM after surgical resection of rectal cancer. The primary outcome was positive CRM (tumor≤1 mm from the surgical margin). Our population included patients with clinical stage I-III rectal cancer who underwent total mesorectal excision. For the primary outcome, multivariable logistic models were used to estimate the probability of a positive CRM. Model performance was evaluated by using the area under the receiver operating characteristic curve (AUC). Model calibration was assessed by examining the calibration plot. Bootstrapping method (300-iteration) was used to internally validate and estimate optimism-adjusted measures of discrimination and overall model fit. RESULTS/ANTICIPATED RESULTS: There were 28,790 patients included. 2,245 (7.8%) had positive CRM. Older age, race, larger tumor size, higher tumor grade, mucinous and signet tumor histology, APR, open operative approach, facility location, higher T stage, lymphovascular invasion, lack of neoadjuvant chemotherapy/radiation, and perineural invasion were all significantly associated with positive CRM (p DISCUSSION/SIGNIFICANCE: An objective model that predicts positive CRM and associated poor clinical outcomes is possible to be used in conjunction with MRI. Positive CRM is associated with specific patient demographics, tumor characteristics, and operative approach. These factors can be used to predict CRM positivity in the preoperative period and plan accordingly.