AI-assessed sarcopenia as an independent predictor of neoadjuvant chemotherapy outcomes in muscle-invasive bladder cancer.
Authors
Affiliations (6)
Affiliations (6)
- Department of Radiological Sciences, Oncology and Pathology, Sapienza University, Sapienza/Policlinico Umberto I, Viale Regina Elena 324, 00161, Rome, Italy.
- Department of Mechanical and Aerospace Engineering, Sapienza University of Rome, Rome, Italy.
- Liver Failure Group, Institute for Liver and Digestive Health, UCL Medical School, Royal Free Hospital, London, UK.
- Department of Urology, IRCCS "Regina Elena" National Cancer Institute, Rome, Italy.
- Department of Translational and Precision Medicine, Sapienza University of Rome, 00161, Rome, Italy.
- Department of Radiological Sciences, Oncology and Pathology, Sapienza University, Sapienza/Policlinico Umberto I, Viale Regina Elena 324, 00161, Rome, Italy. [email protected].
Abstract
Sarcopenia has already been widely investigated as a potential indicator of negative outcomes in oncology patients. Our aim was to evaluate the potential predictive role of sarcopenia assessed using an Artificial Intelligence-powered software in response to neoadjuvant chemotherapy (NAC) in patients with muscle-invasive bladder cancer (MIBC). In this single-centre retrospective study, we enrolled patients diagnosed with non-metastatic MIBC who underwent NAC and had available pre-treatment mpMRI of the bladder and baseline CT scan. The follow-up MRI assessment was performed using the NacVI-RADS algorithm to evaluate response to systematic therapy. AI-based software automatically calculated the skeletal muscle index (SMI) from CT images at the L3 vertebral level. Multivariate logistic regression analysis was performed to assess independent predictors of response to NAC, and a receiver operating characteristic (ROC) analysis was subsequently conducted to provide an additional level of statistical validation. Fifty-five patients were included (mean age: 67.2 years). Sarcopenia was identified in 36.4% of patients. Multivariate logistic regression revealed sarcopenia (OR: 9.08; 95% CI 1.32-61.92; p = 0.024), comorbidities (OR: 14.63; 95% CI 2.12-100.71; p = 0.006), and high NacVI-RADS scores (4-5) (OR = 2.13 95% CI 1.03-4.42; p = 0.042) as independent predictors of poor response to NAC. Receiver operating characteristic (ROC) curve analysis confirmed the high discriminative ability of SMI for predicting treatment response (AUC = 0.952). Sarcopenia, assessed by AI-powered analysis, was negatively associated with tumor response following NAC in patients with MIBC. These findings support the integration of AI-driven sarcopenia evaluation into clinical staging workflows, enabling tailored nutritional interventions and improved patient stratification. Moreover, our study reinforces the prognostic value of the NacVI-RADS scoring system in predicting NAC outcomes.