Timothy R. Rebbeck

BACKGROUND: Risk-reducing salpingo-oophorectomy (RRSO) is widely used by carriers of BRCA1 or BRCA2 (BRCA1/2) mutations to reduce their risks of breast and ovarian cancer. To guide women and their clinicians in optimizing cancer prevention strategies, we summarized the magnitude of the risk reductions in women with BRCA1/2 mutations who have undergone RRSO compared with those who have not. METHODS: All reports of RRSO and breast and/or ovarian or fallopian tube cancer in BRCA1/2 mutation carriers published between 1999 and 2007 were obtained from a PubMed search. Hazard ratio (HR) estimates were identified directly from the original articles. Pooled results were computed from nonoverlapping studies by fixed-effects meta-analysis. RESULTS: Ten studies investigated breast or gynecologic cancer outcomes in BRCA1/2 mutation carriers who had undergone RRSO. Breast cancer outcomes were investigated in three nonoverlapping studies of BRCA1/2 mutation carriers, four of BRCA1 mutation carriers, and three of BRCA2 mutation carriers. Gynecologic cancer outcomes were investigated in three nonoverlapping studies of BRCA1/2 mutation carriers and one of BRCA1 mutation carriers. RRSO was associated with a statistically significant reduction in risk of breast cancer in BRCA1/2 mutation carriers (HR = 0.49; 95% confidence interval [CI] = 0.37 to 0.65). Similar risk reductions were observed in BRCA1 mutation carriers (HR = 0.47; 95% CI = 0.35 to 0.64) and in BRCA2 mutation carriers (HR = 0.47; 95% CI = 0.26 to 0.84). RRSO was also associated with a statistically significant reduction in the risk of BRCA1/2-associated ovarian or fallopian tube cancer (HR = 0.21; 95% CI = 0.12 to 0.39). Data were insufficient to obtain separate estimates for ovarian or fallopian tube cancer risk reduction with RRSO in BRCA1 or BRCA2 mutation carriers. CONCLUSION: The summary estimates presented here indicate that RRSO is strongly associated with reductions in the risk of breast, ovarian, and fallopian tube cancers and should provide guidance to women in planning cancer risk reduction strategies.

The mu (GSTM1 and theta (GSTT1) members of the glutathione S-transferase multigene family are candidate cancer susceptibility genes because of their ability to regulate the conjugation of carcinogenic compounds to excretable hydrophilic metabolites. Deletion variants that are associated with a lack of enzyme function exist at both these loci. Individuals who are carriers of homozygous deletions in the GSTM1 of GSTT1 genes may have an impaired ability to metabolically eliminate carcinogenic compounds and may therefore be at increased cancer risk. Molecular epidemiological studies have provided three pieces of information about the relationship of GSTM1 and GSTT1 with cancer susceptibility. First, the frequencies of homozygous GSTM1 and GSTT1 deletion carriers is very high (i.e., 20-50%) in most populations studied to date. Second, GSTM1 and, possibly, GSTT1 may be involved in the etiology of cancer at more than one site. Third, the risk conferred to individuals who carry homozygous deletions in GSTM1 and GSTT1 appears to be small in magnitude. (e.g., odds ration of < 2). However, the magnitude of risk is larger (e.g., odds ratio of 3-5) when interactions of GSTM1 of GSTT1 with other factors (e.g., cigarette smoking) are considered. These findings have implications for studies of GSTM1 and GSTT1 in cancer susceptibility and for future applications of these biomarkers in cancer prevention of control strategies. First, molecular epidemiological studies should consider both the common frequency of deletion genotypes and the relatively low cancer risk these deletion genotypes may impart. For example, the common frequency of deletion variants may improve statistical power in some molecular epidemiological studies, but large samples may still be required to detect relatively small effect sizes or important interaction effects. Second, the fact that deletion genotypes are common implies that the proportion of cancer attributable to these variants may be large in the general population. However, these genotypes may be less suited for individual cancer risk assessment because of their relatively small contribution to the absolute risk of cancer.

BACKGROUND: Previously, small studies have found that BRCA1 and BRCA2 breast tumors differ in their pathology. Analysis of larger datasets of mutation carriers should allow further tumor characterization. METHODS: We used data from 4,325 BRCA1 and 2,568 BRCA2 mutation carriers to analyze the pathology of invasive breast, ovarian, and contralateral breast cancers. RESULTS: There was strong evidence that the proportion of estrogen receptor (ER)-negative breast tumors decreased with age at diagnosis among BRCA1 (P-trend = 1.2 × 10(-5)), but increased with age at diagnosis among BRCA2, carriers (P-trend = 6.8 × 10(-6)). The proportion of triple-negative tumors decreased with age at diagnosis in BRCA1 carriers but increased with age at diagnosis of BRCA2 carriers. In both BRCA1 and BRCA2 carriers, ER-negative tumors were of higher histologic grade than ER-positive tumors (grade 3 vs. grade 1; P = 1.2 × 10(-13) for BRCA1 and P = 0.001 for BRCA2). ER and progesterone receptor (PR) expression were independently associated with mutation carrier status [ER-positive odds ratio (OR) for BRCA2 = 9.4, 95% CI: 7.0-12.6 and PR-positive OR = 1.7, 95% CI: 1.3-2.3, under joint analysis]. Lobular tumors were more likely to be BRCA2-related (OR for BRCA2 = 3.3, 95% CI: 2.4-4.4; P = 4.4 × 10(-14)), and medullary tumors BRCA1-related (OR for BRCA2 = 0.25, 95% CI: 0.18-0.35; P = 2.3 × 10(-15)). ER-status of the first breast cancer was predictive of ER-status of asynchronous contralateral breast cancer (P = 0.0004 for BRCA1; P = 0.002 for BRCA2). There were no significant differences in ovarian cancer morphology between BRCA1 and BRCA2 carriers (serous: 67%; mucinous: 1%; endometrioid: 12%; clear-cell: 2%). CONCLUSIONS/IMPACT: Pathologic characteristics of BRCA1 and BRCA2 tumors may be useful for improving risk-prediction algorithms and informing clinical strategies for screening and prophylaxis.

Established guidelines for causal inference in epidemiological studies may be inappropriate for genetic associations. A consensus process was used to develop guidance criteria for assessing cumulative epidemiologic evidence in genetic associations. A proposed semi-quantitative index assigns three levels for the amount of evidence, extent of replication, and protection from bias, and also generates a composite assessment of 'strong', 'moderate' or 'weak' epidemiological credibility. In addition, we discuss how additional input and guidance can be derived from biological data. Future empirical research and consensus development are needed to develop an integrated model for combining epidemiological and biological evidence in the rapidly evolving field of investigation of genetic factors.