Abstract
Esophageal cancer (EC) is a major malignancy with poor prognosis and a 5-year survival rate below 30%. Metabolites are key biomarkers for cancer diagnosis, prognosis, and treatment. Their changes and metabolic reprogramming are crucial in EC progression. However, traditional studies struggled to determine causality. A two-sample Mendelian randomization (MR) analysis was performed to determine the causal association between blood metabolites and EC. Using publicly available genetic data, causal associations between 1400 blood metabolites and EC risk were explored. Multiple MR estimation techniques were incorporated, including inverse variance weighting (IVW), MR Egger regression, weighted median, weighted mode, and simple mode. Additionally, sensitivity analyses were conducted to assess the reliability of the results. Using strict inclusion criteria and sensitivity analyses, 9 blood metabolites were identified as significantly associated with EC risk. However, reverse MR analysis indicated potential reverse causality for tyrosine levels, narrowing the focus to 8 metabolites. The identified metabolites and their associations with EC risk were: maltotriose levels (IVW odds ratio [OR] = 1.2396, 95% CI = 1.0658-1.4418, P = .0053); 3-hydroxy-2-ethylpropionate levels (IVW OR = 0.7760, 95% CI = 0.6631-0.9081, P = .0016); 5-dodecenoate (12:1n7) levels (IVW OR = 1.4176, 95% CI = 1.0908-1.8421, P = .0090); 3-methyladipate levels (IVW OR = 0.7469, 95% CI = 0.6125-0.9108, P = .0039); 8-methoxykynurenate levels (IVW OR = 1.2246, 95% CI = 1.0654-1.4076, P = .0043); spermidine to 5-methylthioadenosine ratio (IVW OR = 1.2558, 95% CI = 1.0610-1.4864, P = .0081); adenosine 5'-diphosphate to sulfate ratio (IVW OR = 1.1923, 95% CI = 1.0435-1.3622, P = .0097); glucose-to-mannose ratio (IVW OR = 0.8153, 95% CI = 0.7091-0.9375, P = .0042). Our study has demonstrated the close connection between blood metabolites and EC by genetic means, thus providing guidance for future clinical research.