Introduction: The prognostic value of myopenia following radical gastrectomy (RG) remains controversial.
Materials and methods: Patients who underwent RG between January 2015 and December 2018 at two East Asian centers were retrospectively included. Myopenia was defined using sex-specific computed tomography-derived L3 skeletal muscle index, with cut-off values determined by X-tile software based on overall survival (OS). Cox regression models identified independent prognostic factors for OS and disease-free survival (DFS), whereas logistic regression identified risk factors for textbook outcomes (TO). A LASSO-based prognostic nomogram was developed, and its performance was evaluated using calibration curves, concordance index, time-dependent receiver operating characteristic, and decision curve analysis.
Results: Overall, 943 patients were included (211 and 732 with and without myopenia, respectively). The myopenia group had a significantly lower TO rate (P = 0.003), and myopenia was confirmed as an independent risk factor for TO failure (P < 0.05). Moreover, the myopenia group had significantly lower 5-year OS (45.5 % vs. 69.3 %, P < 0.001) and DFS (45.0 % vs. 66.8 %, P < 0.001), with myopenia independently associated with poor prognosis. The nomogram developed in this study outperformed the tumor-node-metastasis staging system in predicting 3- and 5-year OS (both P < 0.05), demonstrating superior net clinical benefits across a threshold range of 10 %-80 %. Risk stratification effectively discriminated 5-year OS (94.3 % vs. 61.2 % vs. 24.2 %, P < 0.001). These results were reproduced in the validation cohort.
Conclusions: Myopenia independently predicts both short-term outcomes and long-term prognosis after RG. The myopenia-integrated nomogram showed strong predictive accuracy and clinical utility for individualized prognostic assessment of patients with gastric cancer.
Keywords: Gastric cancer; LASSO; Multimodal treatment; Myopenia; Nomogram.
Copyright © 2025. Published by Elsevier Ltd.