Background The obesity paradox has been shown in many cancers, with improved overall survival in higher BMI groups. Lower incidence, better prognosis, and improved treatment response of lung cancer have been associated with increased BMI. However, these differences have varied based on age, gender, and race. This variability can be explained by the heterogeneity of fat deposition across gender and race, which is poorly captured by BMI. Other measures of central adiposity better portray the effects of obesity on lung cancer risk and prognosis. We report the results of another measure of central adiposity, liver fat, on survival and treatment tolerance in NSCLC. Methods: This retrospective cohort study enrolled 106 patients with stage IIIA-IIIC NSCLC treated to 60 Gy at Fox Chase Cancer Center. Liver fat percentage was assessed via CT in Hounsfield units (HU). Patients were separated into two groups, high HU and low HU, correlating to low and high liver fat percentages, respectively. The primary objective assessed mortality while secondary objectives assessed treatment response, disease progression, and acute/chronic treatment toxicities. Results Initial staging was the only categorical variable with significant difference in HU groups. BMI and primary tumor size varied significantly between groups. Overall survival (OS) and progression-free survival (PFS) appeared to be better in the higher HU group, but not statistically significant. No significant difference in acute or chronic treatment-related adverse events between HU groups. Conclusion Liver fat percentage has the potential to be used as a proxy for central adiposity to risk stratify NSCLC patients. Excess adiposity may provide an energy reserve to endure biological stress from disease progression and treatment, leading to a survival advantage. This study showed no statistical significance in OS or PFS between HU groups. One limitation is the small sample size leading to reduced statistical power and difficulty in detecting subgroup effects. Future studies should focus on involving large and diverse cohorts, considering confounding factors like BMI, smoking history, and tumor characteristics. If statistical significance can be measured, liver fat may have the potential to be used as a prognostic marker for outcomes to stratify patients into different risk categories, guiding treatment decisions and improving personalized medicine approaches.