Real-World Study on the Safety of Sintilimab in Clinical Use and Establishment of a Risk Prediction Model

Abstract Objective To retrospectively analyze the clinical safety and adverse drug reactions (ADRs) of Sintilimab, evaluate risk factors for ADRs occurrence, and develop a predictive model to support individualized treatment strategies. Methods Medical records of patients who received Sintilimab treatment from January 2021 and December 2022 were reviewed. Clinical data, including demographic characteristics, medication details, and ADRs. Univariate and multivariate logistic regression analyses were conducted to identify independent risk factors for Sintilimab-induced ADRs, considering variables such as gender, age, comorbidities, and treatment regimens. Receiver operating characteristic (ROC) curve analysis was performed to assess the predictive accuracy of individual and combined risk factors. And the safety of Sintilimab also was evaluated. Results Among 337 included cases, 208 (61.72%) experienced ADRs. Multivariate analysis identified combination therapy (OR = 25.670, 95% CI: 11.319–58.218, P < 0.001) and baseline assessment (OR = 0.388, 95% CI: 0.191–0.789, P = 0.009) as two independent risk factors. The logistic model indicated that the combined prediction of these factors achieved an area under the curve (AUC) of 0.800 (P < 0.001) and a Youden index of 0.583, which outperformed single-factor predictions. Cross-validation using 115 cases demonstrated an accuracy rate of 80.87% for this model. Conclusion Combination therapy and baseline assessment are independent risk factors for ADRs in Sintilimab-treated patients. The combined predictive model exhibits high accuracy and may serve as a valuable clinical tool to anticipate risks and guide personalized treatment decisions.