Development and validation of a prognostic nomogram for gastrointestinal stromal tumors in the postimatinib era: A study based on the SEER database and a Chinese cohort

Abstract

BackgroundAfter the standardization, recording and follow‐up of imatinib use that significantly prolongs survival of gastrointestinal stromal tumors (GISTs), a comprehensive reassessment of the prognosis of GISTs is necessary and more conductive to treatment options.MethodsA total of 2185 GISTs between 2013 and 2016 were obtained from the Surveillance, Epidemiology, and End Results database and comprised our training (n = 1456) and internal validation cohorts (n = 729). The risk factors extracted from univariate and multivariate analyses were used to establish a predictive nomogram. The model was evaluated and tested in the validation cohort internally and in 159 patients with GIST diagnosed between January 2015 and June 2017 in Xijing Hospital externally.ResultsThe median OS was 49 months (range, 0–83 months) in the training cohort and 51 months (0–83 months) in the validation cohort. The concordance index (C‐index) of the nomogram was 0.777 (95% CI, 0.752–0.802) and 0.7787 (0.7785, bootstrap corrected) in training and internal validation cohorts, respectively, and 0.7613 (0.7579, bootstrap corrected) in the external validation cohort. Receiver operating characteristic curves and calibration curves for 1‐, 3‐, and 5‐year overall survival (OS) showed a high degree of discrimination and calibration. The area under the curve showed that the new model performed better than the TNM staging system. In addition, the model could be dynamically visualized on a webpage.ConclusionWe developed a comprehensive survival prediction model for assessing the 1‐, 3‐ and 5‐year OS of patients with GIST in the postimatinib era. This predictive model outperforms the traditional TNM staging system and sheds light on the improvement of the prognostic prediction and the selection of treatment strategies for GISTs.

Evaluating nomogram models for predicting survival outcomes in gastric gastrointestinal stromal tumors with SEER database analysis

“…Nevertheless, previous research has indicated that the Mitotic rate did not display any significant variations in survival outcomes when undergoing multifactor feature selection. The observed disparity is incongruous with the clinical guidelines indicated earlier 7,8,10,11 . The veracity is found within the survival Cox nomograms, which incorporate the independent clinical factor Mitotic rate and the comprehensive index AJCC stage as simultaneous variables in the nomogram.…”

“…However, the existing literature oversimplifies the treatment of missing data classified as 'unknown' in the development of nomograms. Several research 7 , 8 , 10 , 11 consider the classification of 'unknown' as a distinct category. The validity of this assumption is frequently disregarded, as it is contingent upon the proportions of categories inside the 'unknown' group being similar to the overall proportions.…”

“…Due to the rarity of GISTs, big datasets such as the Surveillance, Epidemiology, and End Results (SEER) Program can serve as a global real-world cohort database for researching GISTs 7 – 11 . Since data missingness is a typical occurrence in real-world data gathering, some studies 7 , 8 , 10 , 11 have categorized missing data as 'unknown' as one of the values for multi-category variables in subsequent study. However, such a basic method requires that the missingness of the 'unknown' is balanced across different categories, which is difficult to satisfy.…”

“…Directly adding them may lead to model non-convergence and instability. The traditional design of the cox nomogram normally involves a two-stage process of single-factor and multi-factor selection before nomogram construction 7 – 11 . However, with the widespread usage of Least Absolute Shrinkage and Selection Operator (LASSO) in clinical research as a feature selection and standalone model, Penalized Models have steadily developed as a unique clinical modeling strategy to replace the old two-stage method 14 – 17 .…”