The cadherin 1 (CDH1) gene plays critical roles in the epithelial-mesenchymal transition process, potentially offering us a glimpse into the development of endometrial carcinoma (EC). The present study aimed to identify whether genetic variants in CDH1 affect EC susceptibility in Chinese Han women, using a strategy combining haplotype-tagging single nucleotide polymorphisms (htSNPs) association analysis with fine-scale mapping. A total of 9 htSNPs in CDH1 were genotyped among 516 cases and 706 age-matched cancer-free controls. Logistic regression analyses revealed 3 htSNPs (rs17715799, rs6499199 and rs13689) to be associated with increased EC risk and 3 htSNPs (rs12185157, rs10431923 and rs4783689) with decreased EC risk. Furthermore, 14 newly imputed SNPs of CDH1 were identified to be associated with EC risk (P<0.05) using genotype imputation analysis. Notably, multivariate logistic analysis demonstrated that rs13689, rs10431923 and rs10431924 could affect EC susceptibility independently (P≤0.001). Subsequent Generalized Multifactor Dimensionality Reduction analysis revealed several best fitting models for predicting EC risk, including SNP-SNP interactions among rs7100190, rs12185157, rs10431923, rs7186053, rs6499199, rs4783689, rs13689, rs6499197 and rs10431924, and SNP-environment interactions between related SNPs and number of childbirth. Moreover, functional annotations suggest that the majority of these susceptible variants may carry potential biological functions that affect certain gene regulatory elements. In summary, this study suggested that the genetic polymorphisms of CDH1 were indeed associated with EC susceptibility on several levels. If further additional functional studies could verify these findings, these genetic variants may serve as future personalized markers for the early prediction of endometrial cancer in Chinese Han women.