Development of Nomogram for Predicting Postoperative Progression Risk in High-Risk Thymoma and Thymic Carcinoma Utilizing Clinical and Preoperative CT Features.

Abstract

Background: Computed tomography (CT) features and clinical characteristics have been shown in recent studies to be effective predictive indicators for risk stratification of thymic epithelial tumors. High-risk thymoma and thymic carcinoma (HRT-TC) are highly aggressive and are associated with poor prognoses. The aim of this study is to evaluate the predictive value of CT features and clinical characteristics to assess postoperative progression in patients with HRT-TC.

Methods: Clinical and enhanced CT data were retrospectively collected from patients who underwent HRT-TC surgery between June 1, 2012, and June 1, 2022. A univariate Cox regression analysis was conducted to identify the risk factors associated with postoperative progression. A multivariate Cox regression analysis was then used to determine the independent risk factors. Three-year and 5-year single-factor models as well as multifactorial combined models were then constructed based on the results of these analyses to assess their efficacy, accuracy, and net benefit. The best-performing model was selected to create a nomogram for a consistency assessment.

Results: A total of 215 patients were included in the study. The multivariate Cox regression analysis revealed that independent prognostic factors that influenced postoperative progression were the tumor length (hazard ratio [HR] = 1.027; 95% confidence interval [CI] = 1.004-1.049, P = .018), tumor resection (HR = 4.122; 95% CI = 2.054-8.274, P < .001), and the mediastinal vascular invasion (MVI; HR = 2.779; 95% CI = 1.140-6.775, P = .025). The 3-year and 5-year combined models demonstrated superior predictive efficacy, accuracy, and net benefits. The nomogram and calibration curves showed that the predicted risk probabilities from the nomogram aligned well with actual observations.

Conclusions: A nomogram based on clinical and CT features provided effective predictions of progression following HRT-TC. This prognostic tool holds significant value for clinicians to guide therapeutic decisions and personalize survival assessments.