We conducted a meta-analysis of studies reporting on the risk of extra-ovarian malignancies among women with endometriosis. Summary relative risk (SRR) and 95% confidence intervals (CI) were calculated through random effect models. We explored causes of between-studies heterogeneity and assessed the presence of publication bias. We included 32 studies published between 1989 and 2018. We found an increased risk of endometrial (SRR 1.38, 95%CI 1.10-1.74) and thyroid cancer (SRR 1.38, 95%CI 1.17-1.63), and inverse association with cervical cancer (SRR 0.78, 95%CI 0.60-0.95). No association emerged for breast cancer (SRR 1.04, 95%CI 0.99-1.09) and melanoma (SRR 1.31, 95%CI 0.86-1.96). Between-study heterogeneity was large for breast and endometrial cancer and melanoma. Associations were generally stronger in case-control, cross-sectional, and cohort studies with internal control group, compared to cohort studies with external control group. No indication for publication bias was found. Our conclusions need to be confirmed in properly designed cohort studies with clinical confirmation of endometriosis.