Special Feature

BRCA1 Mutation in Epithelial Ovarian Cancer

By David Gershenson, MD

The brca1 gene was isolated and mapped to chromosome 17q12-21 in 1994. The BRCA1 gene consists of 5592 nucleotides spread over 100 kb of genomic DNA. Evidence to date suggests that at least part of its function is as a tumor suppressor gene. At least 38 distinct mutations of BRCA1 exist. The original estimates of ovarian cancer risk in BRCA1 carriers were as high as 60%. These estimates were based on studies of large families with a relatively high number of cancer cases. More recent studies based on studies of Ashkenazi Jewish women indicate that the risk of ovarian cancer in BRCA1 carriers may be as low as 16% (Struewing, et al. N Engl J Med 1997;336:1401-1408).

Current estimates are that approximately 10% of all ovarian cancers have a genetic basis. BRCA1 mutations account for almost half of these cases. Other genetic causes of ovarian cancer include the BRCA2 gene mutation and the hereditary nonpolyposis colorectal cancer syndrome (HNPCC).

Since its discovery, the prognostic significance of a BRCA1 germ-line mutation in women with ovarian cancer has been of great interest. Early reports of women with BRCA1 germ-line mutation and breast cancer suggested that they had a better prognosis than those women with sporadic breast cancer. In 1996, Rubin and colleagues (N Engl J Med 1996;335:1413-1416) reported on 53 women with ovarian cancer who had germ-line mutations of BRCA1. Control patients were matched for age, stage, histologic grade, and histologic type. Actuarial median survival time for the 43 patients with advanced-stage disease was 77 months compared with 29 months for the matched controls. Rubin et al concluded that ovarian cancers associated with BRCA1 mutations appear to have a significantly more favorable clinical course compared with sporadic ovarian cancers.

Subsequently, a flood of letters was sent to Rubin et al; most were critical of the methods and the findings of this report. Criticisms of the methodology included the fact that no pathology review, with subtype and grade, was performed for this study. In addition, the amount of residual disease, type of chemotherapy, and substage were not specified. Others charged that there was considerable selection bias in this study since the BRCA1 patients and the controls were treated during different timeframes.

We now have two other bits of evidence suggesting that the conclusions drawn by Rubin et al may be off the mark. At the annual meeting of the Society of Gynecologic Oncologists (SGO) in February, an Israeli group presented the preliminary results of a study of 242 blood or tissue specimens analyzed for the presence of the 185delAG mutation of BRCA1 (Ben David Y, et al. Presentation at the 29th Annual Meeting of SGO, Orlando, Florida). Forty-five women (18.6%) were found to be positive. No significant difference was found between the carriers and the non-carriers with respect to median age, ethnic group, percentage of epithelial tumors, stage, and histologic grade. Importantly, no significant difference in survival was found between the carriers and non-carriers: the two-year survival rate was 72% for the carriers and 69.8% for the non-carriers. Although the investigators did not find a difference in survival between the carriers and non-carriers, we need to be cautious about drawing conclusions since the follow-up time is relatively short (2 years).

Also, in February, a Swedish research team reported the survival of 71 BRCA1-associated cancer patients (33 breast cancers, 7 breast and ovarian cancers, and 31 ovarian cancers from 21 families with BRCA1 germ-line mutations) diagnosed after 1958 and compared it with that of a population-based comparison group that consisted of all other invasive breast (n = 28,281) and ovarian (n = 7011) cancers diagnosed from 1958 to 1995, as well as to that of an age- and stage-matched control group (Johannsson OT, et al. J Clin Oncol 1998;16:397-404). No apparent survival advantage was found for BRCA1-associated breast cancers. For BRCA1-associated ovarian cancers, an initial survival advantage was noted that disappeared with time. Johannsson and associates conclude that survival of carriers of BRCA1 mutation may be similar to, or worse, than that for breast and ovarian cancer.

Although early studies indicated that BRCA1 gene mutation conferred a worse prognosis on a woman with ovarian cancer, more recent reports, with apparent superior methodology, suggest that ovarian cancer patients with a BRCA1 mutation have the same, or even a worse, prognosis. However, this story is just unfolding. All of the studies to date include a relatively small number of women with BRCA1 mutations. It is also possible that different mutations of BRCA1 are associated with different outcomes. And, especially in the Israeli study, the follow-up time is short. Therefore, it is important that we keep an open mind on this topic.