A nomogram based on the log odds of positive lymph nodes for predicting the prognosis of T1 stage rectal cancer.

ABSTRACT

Early detection and timely treatment is the key to improving the prognosis of rectal cancer. Lymph node metastasis is one of the reasons for the poor prognosis of rectal cancer, especially early-stage rectal cancer. In this study, we developed a nomogram based on log odds of positive lymph nodes (LODDS) to predict cancer-specific survival (CSS) in patients with T1 rectal cancer. We included 1934 patients from the Surveillance, Epidemiology, and End Results (SEER) database and divided them into a training cohort and an in-validation cohort. 140 patients from our hospital formed the ex-validation cohort. Multivariate Cox regression analysis indicated that age, sex, grade, and M stage were independent prognostic factors for CSS. LODDS showed better predictive ability than the N stage and PLNs (positive lymph nodes) and was further selected as an independent prognostic factor for the construction of the nomogram. The C-index of the nomogram was 0.743, 0.756, and 0.876 in the training, in-validation, and ex-validation cohorts, respectively. The AUC values of the three cohorts were 0.750, 0.703, and 0.958 at 3 years and 0.731, 0.678, and 0.783 at 5 years. The calibration curves and DCA demonstrated the nomogram's excellent performance. In conclusion, we developed and validated a new nomogram based on LODDS that can effectively predict CSS at 3 and 5 years for patients with T1 rectal cancer.