Quantifying Suicide Risk in Prostate Cancer: A SEER-Based Predictive Model.

Abstract

Background: Prostate cancer patients have a significantly higher risk of suicide compared to the general population. This study aimed to develop a nomogram for identifying high-risk patients and providing empirical evidence to guide effective intervention strategies.

Methods: We analyzed data from 176,730 prostate cancer patients diagnosed between 2004 and 2021, sourced from the Surveillance, Epidemiology, and End Results (SEER) database. Patients were randomly allocated to training (n = 123,711) and validation (n = 53,019) cohorts in a 7:3 ratio. Feature selection was conducted using the Least Absolute Shrinkage and Selection Operator (LASSO), followed by model construction with Cox proportional hazards regression. The results were visualized using nomogram. Model performance was evaluated with time-dependent receiver operating characteristic (ROC) curves, concordance index (C-index), and internal validation.

Results: Multivariate analysis identified seven independent predictors of suicide. The nomogram demonstrated favorable discriminative capability in both cohorts, with C-index of 0.746 and 0.703 for the training and bootstrapped validation cohorts. Time-dependent ROC analysis indicated strong accuracy in predicting suicide risk. Calibration plots displayed high concordance between predicted probabilities and actual outcomes, Kaplan-Meier analysis confirmed the model's significant discriminative ability among risk groups.

Limitations: This retrospective study, based on SEER data, lacks detailed clinical and mental health information. Additionally, potential coding errors and reporting biases may affect the accuracy of the results.

Conclusion: We developed a applicable nomogram for the individualized quantification of suicide risk in prostate cancer patients. This model provides clinicians with a robust tool for identifying high-risk patients and implementing timely interventions.