Clinical Application of Artificial Intelligence-Based Computed Tomography Analysis of Myosteatosis in Localized Renal Cell Carcinoma

Abstract

Purpose Myosteatosis, defined as fat infiltration in muscle tissue, has been linked to poor outcomes in various cancers. However, the prognostic impact of myosteatosis on renal cell carcinoma (RCC) remains poorly understood. This study evaluated the predictive value of myosteatosis based on an artificial intelligence (AI)-driven computed tomography (CT) analysis in patients with localized RCC who underwent partial nephrectomy. Materials and Methods This retrospective study included 170 patients with localized RCC who underwent partial nephrectomy at a single institution between 2011 and 2017. Myosteatosis was assessed on CT scans using an AI-based tool. The patients were categorized into 2 groups according to the presence or absence of myosteatosis. The clinical outcomes, including disease-free survival (DFS), were compared to determine the prognostic significance of myosteatosis. Results Of 170 patients, 36 (21.2%) were diagnosed with myosteatosis. These patients were older and had a higher body mass index. The myosteatosis group had a higher proportion of females than the no myosteatosis group. Lymphovascular invasion and tumor necrosis were prevalent pathological features in patients with myosteatosis. Kaplan-Meier analysis demonstrated that myosteatosis was associated with significantly shorter DFS (p<0.05). Multivariate analysis confirmed that myosteatosis independently predicted adverse outcomes in patients with localized RCC. Conclusion AI-based CT analysis of myosteatosis is a reliable method for improving the risk stratification of patients with localized RCC. Patients with myosteatosis demonstrate poor pathological features and shorter DFS. These findings highlight the potential of AI-driven body composition analysis to refine prognostic models and personalized treatment strategies.

Keywords

• renal cell carcinoma
• myosteatosis
• artificial intelligence