BRCA Mutations in Young Breast Cancer Patients Without a Family History Influence Histologic Features of the Tumors

abstract & commentary

Source: Armes JE, et al. Cancer 1998;83:2335-2345.

It is now understood that breast tumors that develop in women with germline mutations in cancer susceptibility genes (particularly BRCA1 and BRCA2) have distinct histologic phenotypes.1,2 However, the majority of the current data is derived from affected individuals from cancer families, but mutations are also observed in women without such a family history. In fact, the majority of women with breast carcinoma who carry a BRCA1 or BRCA2 mutation do not have any family history of breast or ovarian cancer. The current study was designed to determine whether the histologic phenotypes recognized for tumors in BRCA1/2 mutation carriers from cancer prone families apply to breast carcinoma occurring in young women with germline BRCA1/2 mutations but no family history of breast cancer.

The study undertook a histologic assessment of breast carcinomas diagnosed before age 40 years identified from a population-based study. Samples were derived from the Australian Breast Cancer Family Study (ABCFS). Of 467 breast cancer cases diagnosed between 1992 and 1995 in this age group, blood samples were available from 388 cases. BRCA mutations were detected by the protein truncation test and were confirmed by manual cycle sequencing.

For the morphologic study, tumors from 10 cases with germline BRCA1 mutations and nine cases with BRCA2 mutations were compared with tumors from 21 control patients (younger than 40 years old, without BRCA mutations).

Breast cancer in BRCA1 mutation carriers was associated with distinct histologic appearance. The tumors were higher grade with greater mitotic counts, a syncytial growth pattern, pushing margins and confluent necrosis. Furthermore, atypical medullary carcinoma was observed to a greater extent in this group. In contrast, all low-grade tumors and tumors with low mitotic rates were found in the control groups (those without germline BRCA1/2 mutations). Those with germline BRCA2 mutations were more likely to have pleomorphic histology exhibiting extensive intraductal growth.

Thus, there appears to be distinct histologic correlates of mutations in the BRCA genes. The authors suggest that such histologic patterns observed in young patients may be useful in predicting those for whom more extensive genetic testing may be fruitful.

COMMENTARY

Screening for BRCA mutations is a cumbersome and expensive task. This is primarily because these are large genes and numerous evenly distributed mutation sites have been observed. Accordingly, widespread screening is currently not feasible. The criteria used to determine which patients or families should be screened have yet to be established. The current report may be helpful in this regard. Young women with atypical medullary or pleomorphic medullary carcinomas that have histologic features of syncytial growth or prominent pushing margins are more likely to have mutations in BRCA genes.

This may well be the first demonstration, on a population basis, of a phenotypic-genotypic correlation with regard to BRCA mutation. Prior kindred studies3,4 also demonstrated similar histologic features, but in this report individuals were not selected for family history. The fact that such findings were clearly distinct in a small study such as this indicates just how powerful the association is.

High-grade tumors with similar histologic patterns have been identified with breast cancer in younger women in general.5 From the current analysis of 40 patients, it could be predicted that those histologic correlates of more aggressive tumors (seen more commonly in young patients) might well relate, in part, to BRCA mutations. Genetic screening for the families of these patients would seem appropriate.

References

1. Rubin SC, et al. N Engl J Med 1996;335:1413-1416.

2. Jass JR. J Clin Pathol 1997;50:892-895.

3. Marcus JN, et al. Cancer 1996;77:697-709.

4. Lakhani SR, et al. J Natl Cancer Inst 1998;90:1138-1145.

5. Marcus JN, et al. Monogr Natl Cancer Inst 1994; 16:23-34.