Abstract: Objective To explore the causal association between circulating leptin levels and the risk of colorectal adenoma and colorectal cancer. Methods We collected demographic and clinical data and serum samples from 497 patients with colorectal adenoma, 955 patients with colorectal cancer, and 911 healthy individuals from the First Affiliated Hospital of Zhejiang University School of Medicine, Zhejiang Cancer Hospital, Zhuji People's Hospital, and Lin'an District First People's Hospital. Instrumental variables of leptin were selected and genotyping tests were performed. A logistic regression model and stratified analysis were used to evaluate the association of serum leptin levels with colorectal adenoma, colorectal cancer, and the progression of colorectal adenoma to colorectal cancer. Genetic risk score (GRS) and single nucleotide polymorphisms (SNPs) were further used as instrumental variables in one-sample and two-sample Mendelian randomization analyses leveraging two-stage least squares and inverse-variance weighted methods to estimate the causal association of leptin levels with the risk of colorectal adenoma, colorectal cancer, and progression of colorectal adenoma to colorectal cancer. Results High levels of leptin, compared with its lowest quartile, were positively correlated with colorectal adenoma (P=0.005) and negatively with colorectal cancer (P<0.001) and the risk of progression of colorectal adenoma to colorectal cancer (P<0.001). Mendelian randomization analysis showed that GRS of leptin, either weighted or not, was not significantly correlated with the risk of colorectal adenoma, colorectal cancer, or the progression of colorectal adenoma to colorectal cancer, nor did the two-sample Mendelian randomization study support an association between leptin and the risk of colorectal cancer (P>0.05). Conclusion Although the case-control study suggests probable correlations of leptin with the risk of colorectal adenoma, colorectal cancer, and colorectal adenoma progression to colorectal cancer, Mendelian randomization studies did not support a causal association of leptin with the risks of colorectal adenoma, colorectal cancer, or colorectal adenoma progression to colorectal cancer.

Key words: leptin; colorectal adenoma; colorectal cancer; Mendelian randomization