Does pre-checkpoint inhibitor sarcopenia, visceral, or subcutaneous fat predict survival in non-small cell lung...

Sarcopenia and altered body composition have been associated with poor outcomes in cancer patients. We investigated whether skeletal muscle index (SMI), visceral fat area (VFA), and subcutaneous fat area (SFA) measured on CT scans predict survival in non-small cell lung cancer (NSCLC) patients receiving immune checkpoint inhibitor therapy.

This retrospective pilot study analyzed 46 NSCLC patients who had CT scans within 60 days of initiating checkpoint inhibitor therapy (93.5% pembrolizumab, 6.5% nivolumab) at a single tertiary care center. Body composition metrics were measured at the L3 vertebral level using TeraRecon software by a trained reader. Sarcopenia was defined using validated cutoffs (males < 45.4 cm²/m², females < 34.4 cm²/m²). Cox proportional hazards models evaluated associations between body composition and survival, with analyses stratified by sex.

The cohort included 25 males (54.3%) and 21 females (45.7%), with median time from CT scan to checkpoint inhibitor initiation of 14 days (IQR: -47 to 18). One-year survival was 72.0% for males and 59.3% for females; two-year survival was 33.7% and 27.3%, respectively. Those who met the clinical criteria for sarcopenia did not statistically differ in survival compared to those who did not meet the clinical cut off (HR 1.21 (95% CI 0.63, 2.43). This finding remained consistent when males and females were examined separately and when the model was adjusted for timing to checkpoint inhibitor.

In this small pilot study, pre-treatment body composition metrics did not significantly predict survival in NSCLC patients receiving checkpoint inhibitors. However, the study was likely underpowered with wide confidence intervals, limiting definitive conclusions. These findings provide a methodological framework for larger multicenter studies to validate the prognostic utility of CT-derived body composition in this population.