Assessing body composition, particularly muscle mass and quality, is crucial in managing advanced non-small-cell lung cancer (NSCLC) patients. While the third lumbar vertebra (L3) computed tomography (CT) scans are the reference standard, thoracic CTs are more commonly performed in clinical practice.This study evaluated T12 CT-derived muscle measurements as an alternative to L3, correlating them with dual-energy X-ray absorptiometry (DXA) and handgrip strength (HGS).We conducted a cross-sectional study of patients with stage IIIB or IV NSCLC. Skeletal muscle area (SMA), skeletal muscle index (SMI), skeletal muscle radiodensity (SMD), and intermuscular adipose tissue (IMAT) were measured at T12 and L3 using CT. DXA measured total lean soft tissue (LST) mass, trunk LST, and body fat mass. HGS was assessed via a dynamometer. Correlations, linear regression, and Bland-Altman analyses were performed.We included 332 participants (69.3% men; mean age, 57.3 ± 8.5 y). Strong correlations were found between T12 and L3 SMA (men: r = 0.81; women: r = 0.87) and SMI (men: r = 0.81; women: r = 0.89) (all P < 0.001). T12 SMA and SMI exhibited significantly stronger correlations with trunk LST than corresponding L3 measures (SMA: men, r = 0.82 vs. 0.67; women, r = 0.82 vs. 0.72; SMI: men, r = 0.70 vs. 0.53; women, r = 0.73 vs. 0.64; all P < 0.001). SMD correlated well between T12 and L3 (men: r = 0.80, P < 0.001; women: r = 0.82), but IMAT showed poor correlation (men: r = 0.67, P < 0.001; women: r = 0.52) (all P < 0.001). Linear regression showed T12 SMA/SMI predicted L3 SMA/SMI, with better R² in women (all P < 0.001). Bland-Altman analysis showed good agreement for SMI and SMD, but poor agreement for IMAT. Multiple linear regression showed T12 SMA (β = 0.108, R² = 0.751) and SMI (β = 0.245, R² = 0.610) were stronger predictors of trunk LST than L3 SMA (β = 0.069, R² = 0.583) and SMI (β = 0.144, R² = 0.459) (all P < 0.001). Similarly, T12 SMA (β = 0.238, R² = 0.474) and SMI (β = 0.694, R² = 0.513) better predicted HGS than L3 parameters (all P < 0.001).T12 CT-derived SMA and SMI are reliable alternatives to L3 for assessing muscle mass in advanced NSCLC, showing strong agreement with L3 and DXA trunk LST. T12 SMD is a suitable substitute for L3 SMD, but T12 IMAT is not. These findings support incorporating T12-level muscle assessment into clinical research as a viable alternative, leveraging the widespread use of thoracic CT scans.Copyright © 2025 Elsevier Inc. All rights reserved.