Abstract
BACKGROUND: While nutritional factors have been suggested to influence cancer risk, the role of carnitine-a nutrient involved in fatty acid metabolism-remains controversial across cancer types. Emerging evidence indicates that carnitine's anti-inflammatory and metabolic properties might influence carcinogenesis, but its site-specific associations are yet to be fully elucidated. METHODS: We conducted a Mendelian randomization (MR) study to assess the causal relationship between genetically predicted serum carnitine levels and 12 site-specific cancers. Genetic variants associated with serum carnitine were used as instrumental variables to minimize confounding factors. Causal estimates were derived using inverse-variance weighted regression as the primary method, complemented by four regression methods. Furthermore, sensitivity analyses were applied to ensure the robustness and reliability of results. RESULTS: The MR analysis showed no significant causal associations between carnitine levels and 11 types of cancer. However, a modest positive causal relationship was identified between carnitine and pancreatic cancer (OR = 1.085, 95% CI = 1.027-1.14, P-value = 0.003). CONCLUSION: This study provides evidence for a causal association between elevated serum carnitine levels and pancreatic cancer risk, while no significant links were observed with 11 other cancers. Future research is needed to further explore carnitine's role in pancreatic carcinogenesis and its potential for cancer prevention.