Editors' ChoiceENDOMETRIOSIS

Driver mutations take the wheel in invasive yet nonmalignant disease

See allHide authors and affiliations

Science Translational Medicine  07 Jun 2017:
Vol. 9, Issue 393, eaan8194
DOI: 10.1126/scitranslmed.aan8194

Abstract

Despite their low risk for malignant transformation, infiltrating endometriotic lesions harbor cancer-associated mutations.

Endometriosis is a relatively common disease defined by the ectopic growth of endometrial tissue that normally lines the uterus, which usually manifests as endometriotic lesions in the peritoneal cavity and on the ovaries. It affects 5 to 10% of reproductive-aged women and confers a 30 to 50% risk of infertility. Ovarian endometriosis—in particular, atypical endometriosis—is associated with an elevated risk for certain subtypes of ovarian cancer, including endometrioid and clear-cell ovarian carcinomas, and can be considered a precursor lesion to these malignancies. Nonetheless, the vast majority of lesions—including deep infiltrating endometriotic lesions that locally invade pelvic tissues and cause debilitating pain—never progress to cancer.

Although lesions in patients with concurrent cancer have been found to carry cancer-associated driver mutations, including in KRAS and PTEN, detailed genomic analysis of noncancer-associated lesions has been lacking. It is unclear what role cancer-associated mutations play in disease that does not progress to cancer and how that role impacts clinical outcome and response to surgical or medical (typically hormone-based) intervention. Thus, in a recent report, Anglesio and colleagues sought to determine the extent to which noncancer-associated lesions harbor somatic driver mutations.

The authors used multiple mutational analysis platforms, including exome sequencing, targeted sequencing, and digital PCR, along with laser-capture microdissection and immunohistochemistry to carefully study lesion-specific somatic mutations in 27 patients with deep infiltrating endometriotic lesions. Although none of the patients showed any sign of cancer or dysplasia, somatic mutations were found in 79% of them, and 5 patients were found to have mutations known to drive cancer development, including in PIK3CA, KRAS, and ARID1A. In one instance, the same KRAS mutation was discovered in anatomically distinct lesions in the same patient but not in eutopic endometrial tissue and not in lesion-adjacent stromal tissue.

These results offer clues about the pathogenesis of endometriosis, for instance suggesting that multiple lineages of lesions arise in some patients and implicating the “metastatic” spread of lesions that bear identical mutations in other patients. This study also opens up questions regarding the role of known cancer driver mutations in benign disease and hints at the possibility of treating mutation-bearing lesions with anticancer drugs that target the mutated pathways. The prevalence and diversity of somatic mutations in endometriosis also suggests a need for a multivariate or multigene approach to analyzing patient fluids and lesions when screening, diagnosing, and molecularly classifying both benign and malignant diseases.

Highlighted Article

View Abstract

Navigate This Article