Investigating Causal Associations of Circulating Micronutrients Concentrations with the Risk of Lung Cancer: A Mendelian Randomization Study

Abstract

Previous observational studies have suggested that the effect of diet-derived circulating micronutrient concentrations on lung cancer (LC) risk is controversial. We conducted a two-sample Mendelian randomization (MR) analysis to investigate the causal relationship between circulating micronutrient concentrations and the overall risk of LC and three LC subtypes (namely lung adenocarcinoma (LA), squamous cell lung cancer (SqCLC), and small cell lung cancer (SCLC)). The instrumental variables (IVs) of 11 micronutrients (beta-carotene, calcium, copper, folate, lycopene, magnesium, phosphorus, retinol, selenium, zinc, and vitamin B6) were screened from the published genome-wide association studies (GWAS). Summary statistics related to LC and its subtypes came from the largest meta-analysis, including 29,266 cases and 56,450 controls. Inverse-variance weighted (IVW) method is used as the main MR analysis, and the sensitivity analysis is carried out to ensure the MR assumptions. This MR study found suggestive evidence that genetically predicted 6 circulating micronutrient concentrations was correlated with the risk of overall LC (odds ratio (OR): 1.394, 95% confidence interval (CI): 1.041–1.868, p = 0.026, phosphorus), LA (OR: 0.794, 95% CI: 0.634–0.995, p = 0.045, beta-carotene; OR: 0.687, 95%CI: 0.494–0.957, p = 0.026, calcium), SqCLC (OR: 0.354, 95% CI: 0.145–0.865, p = 0.023, retinol), and SCLC (OR: 1.267, 95% CI: 1.040–1.543, p = 0.019, copper; OR: 0.801, 95% CI: 0.679–0.944, p = 0.008, zinc). We found no evidence that other micronutrients are associated with the risk of overall LC or its subtypes. Our study suggested that the increase in circulating beta-carotene, calcium, retinol, and zinc concentration may reduce the risk of LC; the increase in circulating copper and phosphorus concentration may be related to the increased risk of LC. In the future, larger replication samples of LC genetic data and larger micronutrient-related GWAS will be needed to verify our findings.