Nomogram for predicting cardiovascular mortality in patients with gastrointestinal stromal tumor: A population-based study

• Published in: Medicine (Baltimore) 2024-09, Vol.103 (39), p.e39835
• Main Authors: Wang, Huimin, Zheng, Koulong, Tai, Chenhui, Sun, Yimei, Feng, Sujuan, Zhang, Yi, Gao, Ya-Dong
• Format: Article
• Language: English
• Subjects: Adult; Aged; Cardiovascular Diseases - epidemiology; Cardiovascular Diseases - mortality; Female; Gastrointestinal Neoplasms - complications; Gastrointestinal Neoplasms - mortality; Gastrointestinal Stromal Tumors - complications; Gastrointestinal Stromal Tumors - mortality; Humans; Male; Middle Aged; Nomograms; Observational Study; Proportional Hazards Models; Risk Assessment - methods; Risk Factors; SEER Program
• ISSN: 1536-5964; 0025-7974; 1536-5964
• DOI: 10.1097/MD.0000000000039835

This research aimed to develop and validate a clinical nomogram for predicting the probability of cardiovascular death (CVD) in patients with gastrointestinal stromal tumors (GIST). Information regarding patients diagnosed with GIST was extracted from the surveillance, epidemiology, and end results database. The multivariable competing risk model and multivariable Cox regression model were utilized to determine the independent predictive factors. A comparison was made between the results obtained from the 2 models. A nomogram was built to visualize the competing risk model. The nomogram's performance was assessed utilizing concordance index, calibrate curve, decision curve analysis, and risk stratification. A total of 9028 cases were enrolled for final analysis, with CVD accounting for 12.8% of all deaths since GIST diagnosis. The multivariate analysis of competing risks revealed that age, chemotherapy and marital status were identified as independent risk factors for CVD in GIST individuals. The nomogram model exhibited good calibration and strong discriminative ability, indicating its effectiveness in predicting outcomes, with a concordance index of 0.788 (95% confidence interval: 0.753-0.823) in the training set, and 0.744 (95% confidence interval: 0.673-0.815) in the validation set. Decision curve analysis indicated that the prediction model had good clinical practicability. Additionally, risk stratification analysis efficiently divided GIST individuals into high- and low-risk populations for CVD. This was the first research to construct and validate a predictive nomogram using a competing risk model to estimate the individual probabilities of CVD in GIST patients. The nomogram can assist clinicians in making personalized treatment and monitoring plans.