Abstract
BACKGROUNDS AND AIMS: Esophageal cancer (EC) causes approximately 508,000 deaths annually, making it a significant cause of cancer-related mortality. While previous studies have suggested an association between lipoprotein levels and EC risk, the causal relationship remains unexplored. This study aims to investigate the causal link between lipoproteins and EC using Mendelian randomization (MR). METHODS AND FINDINGS: This study employed MR to determine the causal effect between lipoproteins and EC risk, with body mass index (BMI) used as a confounder in multivariable MR (MVMR) analysis. Sensitivity analyses were conducted to assess the reliability of the results. Univariable MR (UVMR) analysis indicated that low-density lipoprotein (LDL) had a significant inverse association with EC risk (p = 0.03; OR = 0.89; 95%CI, 0.73-0.98), while high-density lipoprotein (HDL) and triglycerides showed no significant association. In the synthesis of findings across diverse datasets, LDL maintained a notable inverse association with the likelihood of EC (p < 0.001; OR = 0.89; 95%CI, 0.84-0.94). Triglyceride levels indicated a potential trend toward an adverse correlation with EC susceptibility (p = 0.03; OR = -0.94; 95%CI, 0.89-0.99), whereas HDL levels did not establish a definitive causal link with the occurrence of EC. MVMR analysis, adjusting for BMI, confirmed these findings. CONCLUSION: LDL exhibits a clear inverse causal relationship with EC risk, regardless of BMI adjustment. No causal effects were observed for HDL in relation to EC risk. Meanwhile, there is a small but statistically significant causal relationship between triglycerides and EC risk.