Body composition analysis in metastatic non-small-cell lung cancer: depicting sarcopenia in Portuguese tertiary care

Background/Objectives: Sarcopenia is an emergent prognostic biomarker in clinical oncology. Albeit increasingly defined through skeletal muscle index (SMI) thresholding, literature cut-offs fail to discern heterogeneous baseline muscularity across populations. This study assesses the prognostic impact of using cohort-specific SMI thresholds in a Portuguese metastatic non-small-cell lung cancer (mNSCLC) cohort. Methods: Retrospective study including mNSCLC patients treated between January 2017 and December 2022. ImageJ v1.54g was used to assess cross-sectional CT imaging at the third lumbar vertebra (L3) and calculate L3SMI. Sarcopenia was defined both according to Prado et al. and L3SMI thresholds derived from receiver operating characteristic analysis. Overall survival (OS) was the primary endpoint. Secondary endpoints included first-line (1L) progression-free survival (PFS) and sarcopenia subgroup analysis regarding body mass index impact on OS. Results: The initial cohort included 197 patients. Mean age was 65 years (± 11.31). Most tumors were adenocarcinomas (n = 165) and presented with metastasis (n = 154). SMI was evaluable in 184 patients: cohort-specific thresholds (< 49.96 cm2/m2 for men; < 34.02 cm2/m2 for women) yielded 46.74% sarcopenic patients (n = 86) versus 66.30% (n = 122) per the literature definition. Cohort-specific thresholds predicted both OS (12.75 versus 21.13 months, hazard ratio [HR] 1.654, p = 0.002) and PFS (7.92 versus 9.56 months, HR 1.503, p = 0.01). Conversely, Prado et al. definition lacked prognostic value. Among sarcopenic patients, overweight (HR 0.417, p = 0.01) and obesity (HR 2.723, p = 0.039) had contrasting impacts on OS. Conclusions: Amid reclassification of nearly one-fifth of the cohort, cohort-specific thresholds improved sarcopenia prognostication in mNSCLC. Homogeneity regarding both cancer treatment setting and ethnicity could be key to defining sarcopenia based on SMI.