Incidence of ovarian cancer after bilateral salpingo-oophorectomy in women with histologically proven endometriosis

Objective: To assess the incidence of ovarian cancer in women with histologically proven endometriosis after bilateral salpingo-oophorectomy (BSO).

Design: Retrospective nationwide cohort study.

Setting: Dutch pathology database.

Patient(s): Women with histologically proven endometriosis who had undergone BSO between 1990 and 2015 (n = 7,984). This study consists of 2 control cohorts: women with histologically proven endometriosis without BSO (n = 42,633) and women with a benign dermal nevus (n = 132,535).

Intervention(s): Observational study.

Main outcome measure(s): Number of histologic diagnoses of (extra-)ovarian cancers. Incidence rate ratios (IRR) were estimated for (extra-)ovarian cancer. The number needed to treat was calculated.

Result(s): We identified 9 (0.1%) (extra-)ovarian cancers in the BSO cohort and 170 (0.4%) and 444 (0.3%) ovarian cancers in the endometriosis and nevus control cohorts, respectively. We found an age-adjusted IRR of 0.34 (95% confidence interval [CI], 0.15-0.76) when the BSO cohort was compared with the endometriosis cohort. Comparing the BSO cohort with the nevus control cohort resulted in an age-adjusted IRR of 0.38 (95% CI, 0.17-0.85). The number needed to treat when the BSO cohort was compared with the endometriosis control cohort was 351 (95% CI, 272-591).

Conclusion(s): In this nationwide study, we found that the (extra-)ovarian cancer incidence in women with histologically proven endometriosis decreased to less than the background population risk after BSO. Additionally, we found a significant reduction of the incidence of ovarian cancer when compared with women with histologically proven endometriosis without BSO. Endometriosis surgery could in the future be a preventive strategy in women with endometriosis and a high-risk profile for ovarian cancer.