Image-Based Measures of Obesity are Associated with Alterations in Metabolic Pathways in Non-small Cell Lung Cancer

Background: Studies suggest that the obesity paradox in non-small cell lung cancer (NSCLC) results from the use of body mass index (BMI) as a measure of obesity. However, the mechanistic basis linking body fat and lung cancer behavior remains unclear. We examined the association of image-based measures of obesity with tumor gene expression to identify transcriptional signatures concordant with adiposity and their underlying biology.

Patients and methods: RNA-sequencing data for 143 NSCLC tumor samples generated by the ORIEN consortium was compiled with image-based measurements of total fat. Total fat area (TFA) was quantified at the third lumbar vertebra level using computed tomography images and the SliceOmatic software. Differential gene expression analysis was conducted between patients in the highest and lowest TFA tertiles. Utilizing a validated metabolic analysis pipeline, these differences in gene expression were used to enrich dysregulated metabolic pathways crucial in carcinogenesis.

Results: We identified 1154 gene transcripts as differentially expressed (p ≤ 0.05 and log fold change ≥ 0.58) in metabolic pathways of normal physiology as well as cancer growth. Utilizing the metabolic pipeline, we found 58/114 metabolic pathways were significantly enriched (p ≤ 0.05) in the high TFA individuals, some of which are expected in obese individuals (lipids metabolism), and some were novel. Gene set enrichment analysis (GSEA) identified transcriptional alterations to inflammatory mediation, cell-signaling, and cellular respiration pathways based on TFA.

Conclusions: Image-based measures of adiposity correlate with significant gene expression changes in NSCLC tumors. We have identified altered biological processes associated with obesity, including metabolic vulnerabilities, that can be leveraged in developing new treatment strategies.