AI-Assistance Body Composition CT at T12 and T4 in Lung Cancer: Diagnosing Sarcopenia, and Its Correlation with...

Sarcopenia and low muscle mass are prevalent and prognostically relevant in patients with lung cancer, yet their diagnosis remains challenging in routine clinical practice. Opportunistic assessment using computed tomography (CT) has emerged as a valuable tool for body composition evaluation. We aimed to assess the utility of thoracic CT at T12 and T4 levels in identifying sarcopenia and low muscle mass and explore their correlation with morphofunctional tools such as bioelectrical impedance vector analysis (BIVA), nutritional ultrasound (NU), and functional performance tests.

In this prospective observational study, 80 patients with lung cancer were evaluated at diagnosis. Body composition was assessed using BIVA-, NU-, and CT-derived parameters at T12 and T4 levels. Functional status was measured using the Timed Up and Go (TUG) and 30-Second Chair Stand Test. Sarcopenia was defined according to EWGSOP2 criteria.

Sarcopenia was identified in 20% of patients. CT-derived indices at T12CT demonstrated better diagnostic performance than T4CT. For detecting low muscle mass, the optimal SMI cut-off values were SMI_T12CT < 31.98 cm²/m² and SMI_T4CT < 59.05 cm²/m² in men and SMI_T12CT < 28.23 cm²/m² and SMI_T4CT < 41.69 cm²/m² in women. For sarcopenia diagnosis, the values were SMI_T12CT < 24.78 cm²/m² and SMI_T4CT < 57.23 cm²/m² in men and SMI_T12CT < 21.24 cm²/m² and SMI_T4CT < 49.35 cm²/m² in women. A combined model including SMI_T12CT, RF_CSA, and the 30 s squat test showed high diagnostic accuracy (AUC = 0.826). In multivariable analysis, lower SMA_T12CT was independently associated with risk of sarcopenia (OR = 0.96, 95% CI: 0.92-0.99, p = 0.022), as were older age (OR = 1.23, 95% CI: 1.07-1.47, p = 0.010) and fewer repetitions in the 30 s squat test (OR = 0.78, 95% CI: 0.63-0.91, p = 0.007).

CT-derived body composition assessment, particularly at the T12 level, shows good correlation with morphofunctional tools and may offer a reliable and timely alternative for identifying sarcopenia and low muscle mass in patients with lung cancer.