Exploring the potential causal relationship between lipid metabolism and oral carcinoma in situ: A 2-sample Mendelian randomization study

Oral carcinoma in situ (OCIS) is a Stage 0 precancerous condition with a significant risk of developing into invasive oral squamous cell carcinoma. Despite this, the metabolic processes driving its onset and progression remain largely unknown. Lipid metabolism, a crucial factor in the development and behavior of various cancers, has been extensively studied. However, its exact role in OCIS has not been clearly defined, highlighting the need for further research in this area. The present study is designed to investigate the potential causal associations between specific lipid metabolic markers and OCIS through the application of Mendelian randomization (MR) analysis. Our statistical analyses indicated nominal positive correlations between 2 lipid metabolic markers and the risk of OCIS. Specifically, polyunsaturated fatty acids and total cholesterol levels in very large high-density lipoprotein were associated with increased OCIS risk, with odds ratios of 2.32 (nominal P value = .0463) and 2.43 (nominal P value = .0439) per 1 standard deviation in this marker, respectively. Conversely, the phospholipid-to-total lipid ratio in small very low-density lipoprotein showed a nominal inverse association with OCIS risk, with an odds ratio of 0.31 (nominal P value = .0408) per 1 standard deviation in this marker. False discovery rate correction was applied for multiple testing, and no associations reached study-wide significance. Subsequent sensitivity analyses confirmed the robustness of these nominal findings, and no substantial evidence of horizontal pleiotropy was detected for 3 of the lipid metabolic marker-OCIS associations. This exploratory study identifies 3 lipid metabolic markers (polyunsaturated fatty acids, total cholesterol levels in very large high-density lipoprotein, the phospholipid-to-total lipid ratio in small very low-density lipoprotein) that are consistent with a possible nominal effect on OCIS risk. Summary-level genome-wide association study data were used for exposures (233 lipid metabolic markers, 136,016 European-ancestry participants, genome-wide association study Catalog accession numbers GCST90301941-GCST90302173) and outcome (OCIS, 71 cases and 378,725 controls of European-ancestry participants from the FinnGen project, release version R12, data accessed in November 2024). Instrumental variables were selected based on genome-wide significance (P < 5 × 10-8), linkage disequilibrium (R2 < 0.001, 10,000 kb window), and F-statistic > 10. The primary MR analysis used the inverse-variance weighted method. Sensitivity analyses included MR-Egger regression, weighted median, MR-Pleiotropy RESidual Sum and Outlier, and leave-one-out tests.