Development and validation of a prognostic nomogram for gastric GIST patients under 65 years of age.

Abstract

Background: Gastric gastrointestinal stromal tumor (GIST) is the most common sarcoma of the digestive tract. Due to the fact that younger patients often present with more aggressive disease and exhibit different treatment responses compared to older patients, this study aimed to develop models to predict overall survival (OS) and cancer-specific survival (CSS) in postoperative gastric GIST patients under the age of 65, thereby aiding in the creation of optimal, individualized treatment strategies.

Methods: We first reviewed demographic and clinicopathological characteristics data from 1990 to 2021 of patients diagnosed with GIST in the Surveillance, Epidemiology, and End Results (SEER) database. Subsequently, we examined the data of the external validation cohort from Northern Jiangsu People's Hospital. Utilizing Lasso analysis and multivariate Cox regression analyses, we confirmed the independent risk factors and created nomograms for the prediction of OS and CSS of postoperative gastric GIST patients under age 65, followed by validation with the external validation cohort. To assess the predictive ability of these nomograms, we employed the concordance index (C-index), calibration curves, time-dependent receiver operating characteristic (ROC), and decision curve analysis (DCA).

Results: A total of 735 eligible gastric GIST patients from SEER were enrolled in the training cohort and 113 patients from Northern Jiangsu People's Hospital were enrolled in the validation cohort. 3 factors (grade, M stage, mitotic index) associated with OS and 4 factors (grade, T stage, M stage, mitotic index) associated with CSS were included in the model respectively. In the training cohort, the C-index was 0.706 (95% CI = 0.645-0.767) for OS and 0.880 (95% CI = 0.845-0.915) for CSS, while in the validation cohort, the C-index was 0.718 (95% CI = 0.618-0.818) for OS and 0.805 (95% CI = 0.715-0.895) for CSS. Calibration curves and ROCs for 3-, 5-, and 8-year OS and CSS showed high discrimination and calibration. DCA results showed that the nomograms had clinical value in predicting OS and CSS in gastric GIST patients.

Conclusion: Our nomograms satisfactorily predicted OS and CSS in postoperative gastric GIST patients under age 65, which could assist clinicians in evaluating postoperative status, guiding subsequent treatments, and improving patient prognosis.