Abstract
BACKGROUND: Metabolomics studies to date have described widespread metabolic reprogramming events during the development of non-squamous non-small cell lung cancer (NSCLC). Extending far beyond the Warburg effect, not only is carbohydrate metabolism affected, but also metabolism of amino acids, cofactors, lipids, and nucleotides. METHODS: We evaluated the clinical impact of metabolic reprogramming. We performed comparative analysis of publicly available data on non-squamous NSCLC, to identify concensus altered metabolic pathways. We investigated whether alterations of metabolic genes controlling those consensus metabolic pathways impacted clinical outcome. Using the clinically annotated lung adenocarcinoma (LUAD) cohort from The Cancer Genome Atlas, we surveyed the distribution and frequency of function-altering mutations in metabolic genes and their impact on overall survival (OS). RESULTS: We identified 42 metabolic genes of clinical significance, the majority of which (37 of 42) clustered across three metabolic superpathways (carbohydrates, amino acids, and nucleotides) and most functions (40 of 42) were associated with shorter OS. Multivariate analyses showed that dysfunction of carbohydrate metabolism had the most profound impact on OS [hazard ratio (HR) =5.208; 95% confidence interval (CI): 3.272 to 8.291], false discovery rate (FDR)-P≤0.0001, followed by amino acid metabolism (HR =3.346; 95% CI: 2.129 to 5.258), FDR-P≤0.0001 and nucleotide metabolism (HR =2.578; 95% CI: 1.598 to 4.159), FDR-P=0.0001. The deleterious effect of metabolic reprogramming on non-squamous NSCLC was observed independently of disease stage and across treatments groups. CONCLUSIONS: By providing a detailed landscape of metabolic alterations in non-squamous NSCLC, our findings offer new insights in the biology of the disease and metabolic adaptation mechanisms of clinical significance.