Prognostic nomograph of patients with locally advanced (stage II-III) rectal adenocarcinoma after surgery - A Study based on the SEER Database and a Chinese Cohort

Background: Rectal cancer has a very high global incidence rate, among which rectal adenocarcinoma is the most common. The treatment of rectal adenocarcinoma involves surgery, chemotherapy, radiotherapy, and biological immunity. However, radical resection and total mesenteric resection are standard surgical methods for locally advanced rectal adenocarcinoma (stage II-III). The current study aimed to investigate the factors of overall survival (OS) after local advanced rectal cancer (LARC) surgery. Moreover, a prognostic nomograph was constructed to predict OS. Methods: We collected the clinical data of stage II-III rectal adenocarcinoma patients undergoing radical surgery from 2010 to 2015 from the monitoring, epidemiology, and final results (SEER) database. They were divided into training and validation cohort. Simultaneously, 393 patients were retrospectively obtained from Yunnan Cancer Hospital as the external validation queue. Depending on the risk factors affecting the prognosis, the Cox proportional hazard regression model was utilized to construct a nomograph. The identification and calibration of the nomogram were evaluated using the C index, receiver operating characteristic (ROC), and calibration curve. Result: A total of 1317 patients with stage II-III rectal adenocarcinoma were enrolled. This included 924 in the training group, 393 in the internal validation group, and 393 in the external validation group. Multivariate COX analysis revealed that age, chemotherapy history, carcinoembryonic antigen (CEA), and neural invasion (PNI) were associated with postoperative OS. Nomograms were developed after independent prognostic factors were identified through univariate and multivariate analyses.The C index of nomographs in the training and internal validation cohorts was 0.682(95%CI0.610-0.754) and 0.672(95%CI0.510-0.734),respectively.In the training cohort,the AUC(ROC curve) at 3 years was 0.7(95%CI:0.667-0.733) and 0.657(95%CI:0.631-0.683) at 5 years.In the internal validation cohorts,the AUC (ROC curve) at 3 years was 0.656(95%CI:0.618-0.694) and 0.633(95%CI:0.598-0.668) at 5 years.In the external validation cohort, the C index was 0.701(95%CI0.602-0.780), and the AUC (ROC curve) at 3 years was 0.669(95%CI:0.634-0.704) and 0.71(95%CI:0.68-0.74) at 5 years,indicating good discrimination for the model. The calibration curve of 3-year and 5-year OS probability had good consistency between the predicted and actual survival rates. In addition, patients were divided into three different OS risk groups based on the nomograph: high-risk, medium-risk, and high-risk. The Kaplan-Meier curve indicated that the nomograph had a strong predictive effect on the prognosis of the three groups (PConclusion: In this study, a nomogram was established for patients with locally advanced rectal adenocarcinoma after surgery, which has been an effective clinical tool and can predict the prognosis of patients after surgery.