ARID1A Mutations in Endometriosis-Associated Ovarian Carcinomas
Wiegand KC, Shah SP, Al-Agha OM, et al
N Engl J Med. 2010;363:1532-1543
Summary
An international team of investigators examining the genetic profile of clear-cell carcinoma of the ovary discovered a specific somatic mutation in 6 tumors among a group of 19 ovarian cancers (18 primary malignancies and 1 cell line). This abnormality occurred in ARID1A, which encodes for the protein BAF250a. The gene is an important part of a chromatin remodeling complex known as SWI-SNF, which is involved in several fundamental normal cellular processes and is believed to be important in tumor suppression. Of note, before this work, mutations of ARID1A in ovarian cancers had not been reported.
The investigators subsequently examined an additional 210 ovarian cancers for abnormalities in ARID1A and 455 for abnormalities in the expression of BAF25a. Overall, 46% of clear-cell carcinomas (55 of 119) and 30% of endometrioid carcinomas (10 of 33) exhibited mutations in ARID1A, vs no abnormalities in high-grade serous ovarian tumors (0 of 76). In addition, a strong correlation was noted between loss of BAF250a protein expression and the endometrioid or clear-cell subtypes. Finally, in several cases, loss of this protein was observed in adjacent atypical endometriosis but not in endometriosis found distant from the cancer.
Viewpoint
Recent data have demonstrated the prognostic and predictive value associated with the presence of BRCA1 and BRCA2 abnormalities, which are found in 5%-10% of women with ovarian cancer. In the relatively near future, patients with BRCA abnormalities are likely to be treated with therapies specifically targeted to this patient population.
The current study found that as many as one half of patients with clear-cell carcinoma and one third of patients with endometrioid cancer have a mutation in ARID1A, resulting in a loss of BAF250a expression. This raises the provocative and hopeful possibility that antineoplastic drugs targeted to this mutation or the loss of the corresponding protein may favorably affect the natural history of the malignancy while sparing normal tissues.
Furthermore, observation of an identical mutation in the apparently nonmalignant adjacent endometriosis suggests that this defect is a relatively early event in the transformation of normal tissue into malignancy. Finally, this finding strongly supports the conclusion that these malignancies have a unique pathway for development, as the genetic abnormality is not seen in the more common serous ovarian cancers.
Abstract