Body Composition and Inflammatory-Metabolic Indices in Immunotherapy-Treated Metastatic Non-Small Cell Lung Cancer: The SARC-CTI Study

Immune checkpoint inhibitors (ICIs) have improved outcomes in metastatic non-small cell lung cancer (mNSCLC), but substantial heterogeneity in clinical outcomes remains. Host-related factors reflecting systemic inflammation, metabolic status, nutritional reserve, and skeletal muscle composition may provide prognostic information beyond conventional tumor- and treatment-related characteristics. This retrospective, single-center observational cohort study includes consecutive adult patients with histologically confirmed mNSCLC who initiated ICI-based systemic therapy at Ankara Etlik City Hospital between December 2022 and March 2025. A total of 155 patients meeting the eligibility criteria constitute the overall cohort. The survival data cutoff is June 11, 2026. Baseline CT-derived body composition is assessed at the third lumbar vertebral level using an eligible abdominal CT examination performed within 30 days before ICI initiation. Skeletal muscle area (SMA), skeletal muscle index (SMI), and skeletal muscle attenuation are evaluated using a standardized segmentation protocol. SMI is calculated as skeletal muscle area divided by height squared (cm²/m²). Sarcopenia is defined using sex-specific L3 SMI thresholds of \<52.4 cm²/m² for men and \<38.5 cm²/m² for women. Two composite baseline indices are evaluated. The C-reactive protein-triglyceride glucose index (CTI) is calculated as: TyG = ln\[(fasting triglyceride \[mg/dL\] × fasting glucose \[mg/dL\]) / 2\] CTI = 0.412 × ln\[CRP (mg/L)\] + TyG. The cachexia index (CXI) is calculated as: CXI = (SMI × serum albumin \[g/dL\]) / neutrophil-to-lymphocyte ratio. CTI and CXI are analyzed as continuous variables and are evaluated in separate multivariable models because they capture partly overlapping inflammatory, metabolic, nutritional, and body-composition domains. The principal survival outcomes are progression-free survival (PFS) and overall survival (OS), calculated from ICI initiation. Associations of baseline CTI and CXI with PFS and OS are evaluated using multivariable Cox proportional hazards regression models incorporating prespecified clinically relevant covariates. Model performance is explored using Harrell's concordance index and time-dependent receiver operating characteristic analyses for 12-month OS and 6-month PFS. Calibration and decision-curve analyses are also performed. These analyses describe apparent performance within the study cohort and are not externally validated. Early longitudinal changes in body composition are evaluated in a landmark cohort of 140 patients who have both an eligible baseline CT and an early follow-up CT performed 8-12 weeks after ICI initiation and who are evaluable at the landmark assessment. Changes in SMI, skeletal muscle attenuation, sarcopenia status, and body weight are assessed. Landmark PFS and OS are calculated from the date of the early follow-up CT to reduce immortal-time bias. Radiologic response at the first response assessment is additionally described using RECIST version 1.1 whenever adequate measurable disease and anatomic imaging are available. Sensitivity analyses are performed in patients treated with ICI monotherapy, and exploratory analyses assess relationships among CTI, CXI, sarcopenia, body composition, and other clinical characteristics. The study is observational and retrospective; all treatments were administered as part of routine clinical care, and no study-specific therapeutic intervention was performed.