The following is a summary of “A nomogram based on pretreatment radiomics and dosiomics features for predicting overall survival associated with esophageal squamous cell cancer,” published in the June 2024 issue of Surgery by Kawahara et al.
The researchers propose a nomogram-based survival prediction model for patients with esophageal squamous cell carcinoma (ESCC) undergoing definitive chemoradiotherapy. This model incorporates pretreatment computed tomography (CT), positron emission tomography (PET) radiomics, dosiomics features, and common clinical factors.
Radiomics and dosiomics features were extracted from CT and PET images and dose distribution data from two institutions. The least absolute shrinkage and selection operator (LASSO) with logistic regression was utilized to select significant radiomics and dosiomics features, leading to the calculation of radiomics and dosiomics scores (Rad-score and Dos-score). The model was trained using data from 81 patients and validated in 35 patients at Center 1 through 10-fold cross-validation. An external test was conducted on 26 patients from Center 2. Predictive clinical factors, Rad-score, and Dos-score, were integrated to develop the final nomogram model.
Through LASSO Cox regression, 13 CT-based, 11 PET-based radiomics, and 19 dosiomics features were selected. Univariate Cox regression identified T-stage, N-stage, and clinical stage as significant prognostic clinical factors. In the external validation cohort, the combined model’s C-index values for CT-based radiomics, PET-based radiomics, and dosiomics features with clinical factors were 0.74, 0.82, and 0.92, respectively. Significant overall survival (OS) differences were observed between high- and low-risk groups in the combined model (P = 0.019 for CT-based radiomics, P = 0.038 for PET-based radiomics, and P = 0.014 for dosiomics features).
Dosiomics features demonstrated superior predictive capability for overall survival compared to CT- and PET-based radiomics features in patients with ESCC treated with radiotherapy. This nomogram provides a valuable tool for personalized prognostication and treatment planning, offering significant insights for clinical decision-making in ESCC management.
Source: sciencedirect.com/science/article/abs/pii/S074879832400502X
The post Predicting OS in ESCC Using a Pretreatment Radiomics and Dosiomics-Based Nomogram first appeared on Physician's Weekly.