Nomograms for Predicting the Overall and Cause-Special Survival of Elderly Patients with Primary Ocular Adnexal Lymphoma

Abstract

Abstract Background: The goal of our study was to construct and validate nomograms for the prognosis of elderly primary ocular adnexal lymphoma (POAL) patients from the Surveillance, Epidemiology, and End Results Program (SEER) database. Methods: We screened the data of POAL patients aged 60 years or older from the SEER database from 2010 to 2015. For the prediction of the cause-special survival (CSS) and the overall survival (OS) at 1, 3, and 5 years in elderly POAL patients, we constructed nomograms. The nomograms were validated by the decision curve analysis (DCA), the area under the curve (AUC), and the calibration curve. Results: Among 821 enrolled POAL patients from the SEER database, 547 were assigned to the training group and 247 to the validation group. The C-index of the two groups was 0.744 and 0.755 in the OS model, 0.745 and 0.855 in the CSS model. The AUC values of the nomograms for OS were 0.819, 0.785, and 0.767 in the training cohorts and 0.872, 0.789, and 0.755in the validation cohorts at 1, 3, and 5 years, respectively. The AUC values of predicted CSS in the training and validation cohorts at 1, 3, and 5 years were 0.754, 0.772, 0.765 and 0.840, 0.864, 0.791, respectively. The calibration and DCA curves also demonstrated the predictive performances. Conclusions: Herein, for predicting the CSS and OS in elderly POAL patients, we constructed predictive nomograms using independent risk factors. These nomograms can help clinicians in predicting the prognosis of elderly POAL patients.