Prognostic Significance of Bone Marrow Computed Tomography Attenuation in Patients With Non-small Cell Lung Cancer Without Curative Surgery

Introduction Bone marrow fluorodeoxyglucose (FDG) uptake and bone marrow computed tomography (CT) attenuation are imaging biomarkers reflecting bone marrow activity. This study aimed to investigate whether bone marrow CT attenuation is associated with the prognosis of patients with non-small cell lung cancer (NSCLC). Methods Overall, 160 patients with NSCLC who underwent fluorodeoxyglucose positron emission tomography (FDG-PET)/CT staging between January 2018 and December 2019 were retrospectively reviewed. Using an elliptical region of interest in an axial image of the fused PET and CT scans, the mean standardized uptake value of 18F-FDG was measured for the vertebral bone marrow and liver as reference areas, and the mean CT attenuation value of the femoral bone marrow. Survival curves were constructed using the Kaplan-Meier method to calculate overall survival (OS) rates. The predictive value of the variables for OS was evaluated using the log-rank test in univariate analysis and the Cox proportional hazards regression model in multivariate analysis. Age, sex, stage, serum albumin level, C-reactive protein (CRP) level, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), bone marrow-to-liver ratio (BLR), and bone marrow CT number were evaluated in the OS analysis. All continuous variables in the survival analysis were dichotomized according to the cut-off values determined by receiver operating characteristic curve analysis. Results Of 160 patients, 105 underwent surgery; 24 (22.9%) experienced recurrence, and 22 (21.0%) died during the 60-month clinical follow-up. Of the 55 patients who did not undergo surgery, 37 (67.2%) died during follow-up. Bone marrow CT number showed a weak correlation with BLR (γ=0.257) and BMI (γ=0.323). In the analysis of patients who received curative surgery for NSCLC, the following were significantly associated with prognosis in the univariate analysis: age (P