Nomogram for predicting suicide risk in pancreatic cancer patients: a retrospective study based on SEER database

Bi JinzheYU YAQUN^*

• Department of Hepatobiliary and Pancreatic Surgery, Guilin Medical University, Guilin, Guangxi Zhuang Region, China

The incidence of suicide among individuals diagnosed with pancreatic cancer surpasses that observed in both the general population and patients with other types of cancer. This study seeks to identify independent risk factors associated with suicide post-pancreatic cancer diagnosis and to develop a predictive model that may contribute to reducing suicide rates in this patient population.We screened and identified patients diagnosed with pancreatic cancer from the Surveillance, Epidemiology, and End Results (SEER) database between 2000 and 2020, and randomly divided them into training and validation sets. A combination of least absolute shrinkage and selection operator (LASSO) and Cox proportional hazards regression was employed to identify relevant predictive factors. Based on these analyses, we developed a novel nomogram to visualize suicide risk following a pancreatic cancer diagnosis. The performance of the nomogram was evaluated utilizing the concordance index (C-index), receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).A cohort of 129,301 eligible pancreatic cancer patients was incorporated into the study. Age at diagnosis, sex, marital status, surgery, and chemotherapy are independent predictors of suicide in pancreatic cancer patients. The accuracy of the nomogram was evaluated using the C-index and ROC curves, demonstrating acceptable performance in both the training and validation sets. Additionally, the calibration plot exhibited a high degree of concordance between the predicted and observed values, while the clinical decision curves indicated a significant clinical benefit.The proposed nomogram is a reliable tool for predicting suicide risk following a pancreatic cancer diagnosis, providing valuable support for personalized and precise risk quantification in clinical practice.