Objective: To explore the association between endometriosis and cancer risk, and to evaluate the potential causal relationship through Mendelian Randomization (MR) analysis. Methods: Female participants were divided into two groups according to the presence or absence of endometriosis. Multivariate Logistic regression model was used to analyze the relationship between endometriosis and the risk of cancer. The possible confounding factors such as age, body mass index (BMI), race, education level, smoking status, marital status, menarche age, parity, family poverty to income ratio (PIR), diabetes and hypertension were gradually controlled. Subgroup analysis and interaction testing were performed. To test potential causal relationships, single nucleotide polymorphisms (SNPs) associated with endometriosis were selected as instrumental variables on the basis of genome-wide association studies (GWAS), and the causal effects were mainly estimated by inverse variance weighting(IVW). Sensitivity analysis was performed using MR-Egger and Cochrane’s Q test. Results: A total of 3,987 participants were included in NHANES data. Multivariate Logistic regression showed that endometriosis was associated with an increased incidence of malignant tumors \[odds ratio (OR):2.160,95% confidence interval(95%CI):1.429~3.264,P < 0.001\]. The results of subgroup analysis were consistent(all P for interaction>0.05). Two-sample Mendelian randomization analysis showed that endometriosis was associated with an increased risk of ovarian cancer(OR=1.493,95%CI:1.220~1.827,P< 0.001), endometrial cancer(OR=1.125,95%CI:1.037~1.219,P=0.004). There was a causal relationship between the increased risk of cervical cancer(OR=1.186,95%CI:1.063~1.323,P=0.002), and endometriosis was the most closely related to ovarian cancer. Further MR Analysis of the association between endometriosis and different histological subtypes of ovarian cancer revealed that: There was a significant association between endometriosis and high-grade serous ovarian cancer(OR= 1.122,95%CI:1.054~1.195,P<0.001), low-grade serous ovarian cancer(OR= 1.269,95%CI:1.085~1.485,P=0.002), ovarian clear cell carcinoma(OR=1.491,95%CI:1.277~1.741,P<0.001), and endometrioid carcinoma(OR= 1.392,95%CI:1.266~1.530,P<0.001). Although there was an association between endometriosis and low-grade mucinous ovarian cancer (OR = 1.148,95%CI:1.027~1.283,P =0.015), this association was no longer statistically significant after FDR correction. There was no significant association between endometriosis and invasive mucinous ovarian cancer(OR=1.123,95%CI:0.987~1.279,P=0.079). Conclusion: The combined evidence from the NHANES-NDI cohort and MR Analysis suggests an increased risk of ovarian cancer in women with endometriosis, as seen with other gynecologic malignancies, and provides a new epidemiological basis for exploring the potential causal relationship between endometriosis and gynecologic cancers.