Untargeted lipidomics of non-small cell lung carcinoma shows differentially abundant lipid classes in cancer vs non-cancer tissue
Lung cancer is the leading cause of cancer death worldwide and non-small cell lung carcinoma (NSCLC) represents 85% of newly diagnosed lung cancers. The high mortality rate of lung cancer is due in part to the lack of effective treatment options for advanced disease. A major limitation in the development of effective treatment options is our incomplete understanding of NSCLC metabolism at a molecular level, especially lipids. Improvements in mass spectrometry combined with our untargeted assignment tool SMIRFE enable the systematic and less biased examination of NSCLC lipid metabolism. Lipids were extracted from paired tumorous and non-tumorous lung tissue samples from 86 patients with suspected resectable stage I or IIa primary NSCLC and were analyzed using ultra-high resolution Fourier transform mass spectrometry. Pathological examination of the samples revealed that the majority of the samples were primary NSCLC; however, the disease group does include examples of metastases of other cancers and several granulomas. Information-content-informed (ICI) Kendall tau correlation analysis revealed correlation and co-occurrence patterns consistent with significant changes in lipid profiles between disease and non-disease samples. Lipids assigned using the program SMIRFE (Mitchell et al., 2019) were analyzed for differential abundance, followed by machine learning to classify the SMIRFE formula assignments into lipid categories. At the lipid category level, sterol abundances were consistently higher in diseased versus non-diseased lung tissues at statistically significant levels. The statistically significant increase in sterol abundances in primary NSCLC compared with non-cancerous lung tissue suggests for treatment of primary NSCLC a possible therapeutic role for statins and nitrogenous bisphosphonates, pharmaceuticals that inhibit endogenous sterol biosynthesis. This hypothesis is consistent with previous epidemiological studies that have identified a therapeutic role for statins in the treatment of NSCLC but were unable to identify a molecular mechanism for this effect. Additionally, the majority of the consistently increased sterol abundances belong to the sterol ester subcategory, suggesting increased SCD1 and ACAT1 activity. SCD1 expression is a known negative prognostic indicator for survival in NSCLC. In our study, a large fraction of the NSCLC samples displayed this phenotype; however, SCD1 mutants would be unexpected in all of these samples. This suggests that this metabolic phenotype may be shared across multiple genetic subtypes of NSCLC. Thus, inhibitors of SCD1 and other enzymes involved in the production of this metabolic phenotype could have utility in the treatment of many genetic subtypes of NSCLC.