It has become increasingly clear that the four main histological subtypes of epithelial ovarian cancer (EOC), high-grade serous, endometrioid, clear cell and mucinous, are entities with different epidemiologies, clinical presentations, responses to treatment, and ultimate outcomes. In fact, for all intents and purposes, they can be considered different diseases, their only common denominator being that they frequently involve the ovary and pelvic organs. However, clinical practice has not caught up with these insights and the treatment of EOC is that of a single disease entity. In part, this is because we lack detailed knowledge of the molecular mechanisms driving the pathogenesis of each disease, which is vital in order to develop therapeutic approaches against common driver events. In the last few years, mutations in ARID1A and PIK3CA have been described in a substantial fraction of cases of ovarian clear cell carcinoma, yet the paper by Yamamoto et al in this issue of The Journal of Pathology reveals that PIK3CA mutations can be detected in precursor endometriosis tissues. These and other recent observations underscore the importance of investigating whether mutations in the eutopic endometrium actually predispose to endometriosis and eventually to malignancy.