SEARCH

SEARCH BY CITATION

Keywords:

  • BRCA1;
  • BRCA2;
  • genetic testing;
  • pedigree analysis

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Thousands of individuals have undergone mutational analysis of BRCA1 and BRCA2. The Ohio State University Clinical Cancer Genetics program has identified 466 individuals from 289 families with a mutation in BRCA1 or BRCA2. Excluding Ashkenazi Jewish founder mutations, we observed 9 deleterious BRCA mutations five or more times in ostensibly unrelated families and another 13 mutations in 3–4 families. We hypothesized that some of the rarer recurrent mutations observed in our population were due to different branches of the same family being tested independently without knowledge of previous testing of relatives. We examined 90 pedigrees for individuals with the same mutations that were seen three or more times for shared reported family medical history or surnames. Familial links were made in four instances out of a total of 22 shared mutations despite the fact that individuals were not aware that another family member had been tested. As more individuals undergo BRCA testing, we propose that this phenomenon will become more common. Being unaware of previous testing in a family not only affects the risk assessment but also likely increases the costs associated with the genetic testing and subsequent cancer screening in many cases.

Clinical mutational analyses of BRCA1 and BRCA2 to identify individuals with hereditary breast and ovarian cancer syndrome (HBOC) have been available since the mid 1990s. The identification of a BRCA gene mutation in a family allows probands and subsequently tested family members to make informed decisions regarding their risks of breast, ovarian and other cancers. Identifying a BRCA gene mutation in a proband also allows for targeted, site-specific mutation analysis in relatives. A positive result in the relatives of a BRCA mutation carrier indicates they also have the BRCA-associated cancer risks and can consider risk reduction options. A negative result in this context is equally powerful and is considered a ‘true negative’. In most cases, in the absence of unexplained family history or other risk factors, the risk for these individuals is considered to be much lower than that of their BRCA-positive relatives. In some cases, however, a BRCA mutation is not identified in a seemingly high-risk family. In this situation, unaffected family members are still considered to have higher cancers risks than individuals without a family history of cancer and are often recommended to adhere to more aggressive breast cancer screening protocols [1, 2].

Sharing of genetic test results is of great importance given the potential impact on relatives, and it is part of standard genetic counseling practice to discuss how to share results with family members.[3] Some potential barriers to notifying family members of cancer genetic testing results have been documented in the literature. Most of these studies have focused on dissemination of information to first degree relatives (FDRs) [4-6]. Fewer studies have assessed the sharing of information with more distant relatives. In a 2010 study, Cheung et al. surveyed 1103 BRCA testers regardless of test result. Slightly less than half of respondents reported informing a first cousin of their test results (47%) [7]. A Dutch study that used semi-structured interviews, in conjunction with psychological assessments and questionnaires, found that although most individuals who had undergone BRCA gene testing informed their FDRs of their results, none of the BRCA-positive individuals told all of their cousins and only 58% told some of their cousins. The most commonly cited reasons for not telling distant relatives of their genetic testing results were (i) they had little contact with these relatives, (ii) they assumed that the relatives were informed by other family members, and (iii) they felt their results were confidential [8].

Given these barriers to familial communication, and as BRCA gene testing becomes more common, it is likely that distant family members are being tested without the knowledge that other family members are also being evaluated.

The Ohio State University Clinical Cancer Genetics program (OSU CCG) has offered clinical genetic testing of BRCA1 and BRCA2 since 1996. Owing to the large number of individuals who have been tested, we hypothesized that some mutation-positive individuals from the same family have been tested without the knowledge of the other. Here, we describe our study population and the number of families that were ‘linked’ (found to share an identifiable common ancestor or likely common ancestor) after genetic testing.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We performed retrospective database queries of the OSU Wexner Medical Center (OSUWMC), Division of Human Genetics database to determine the frequency of all identified BRCA gene mutations. We then compared recurrent mutations to those collected from a worldwide collection by the Consortium of Modifiers of BRCA1/2 (CIMBA) and identified nine mutations that were not reported as the most frequent (top 10 of all mutations). We tested whether the most common non-Ashkenazi Jewish (AJ) founder mutations (found in 59 families) or these rarer recurrent mutations (found in 31 families) were shared by different familial branches that were undergoing testing independently by examining 90 pedigrees with these shared mutations. We evaluated pedigrees for relatives with identical names and concordant reports of cancer family history. This study was approved by the Institutional Review Board at the Ohio State University.

Costs of breast cancer screening were obtained from the 2012B Physician Fee Schedule from the Centers for Medicare and Medicaid Services [9].

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Demographics of BRCA1 and BRCA2 population

The OSUCCG began offering clinical testing of mutations in BRCA1 and BRCA2 in 1996. From retrospective database queries over a 16 year period, we determined that we tested 2509 individuals including 466 individuals from 289 families with a pathogenic BRCA1 or BRCA2 mutation (Table S1, Supporting Information). Through the collection of family histories, we confirmed genetic test results on an additional 318 relatives who were tested elsewhere.

Cancers in mutation carriers

BRCA1 and BRCA2 mutations confer a greatly increased risk of developing cancer. As expected, the most common cancers observed in females in our cohort are breast (n = 199) and ovarian/fallopian tube/primary peritoneal cancer (n = 54). Three males had breast cancer. The age range of breast cancer diagnoses in our female BRCA mutation carriers was 25–80 years and for ovarian/fallopian tube/peritoneal cancer was 22–74 years. A number of cancers that are not typically associated with BRCA1 and BRCA2 were observed in mutation-positive individuals (Table S2), but these occurred in less than 5% of our population and therefore do not appear to be in excess of population risk.

Characteristics of mutations

In our patient cohort of 289 families, we identified 169 different deleterious BRCA mutations; many of which were only observed in one proband or family. The most common mutations were the Ashkenazi founder mutations BRCA1 187delAG (14 individuals), BRCA1 5385insC (14 individuals) and BRCA2 6174delAG (9 individuals). Excluding the AJ founder mutations, we observed nine deleterious mutations five or more times in ostensibly unrelated families (Table 1). These include BRCA1 mutations C61G (5 families), 3875del4 (5 families), 2800delAA (6 families) and Exon13ins6kb (5 families), and BRCA2 mutations 5910C>G (6 families), 8395G>C (6 families), 7297delCT (5 families), 6503delTT (6 families), and IVS15-1G>A (5 families). As very common mutations can be due to mutational hotspots or more distantly related shared ancestors, we looked at recurrent, but less frequent mutations. We identified 13 additional mutations that were observed in 3–4 families tested (Table 1). Overall, in 4 of the 22 groups of pedigrees with mutations in common, links were made despite the fact that individuals were not aware that another branch of their family had been tested. Two families (BRCA1 A1623G and BRCA1 2722C>G) were linked through a more distant relative with a shared surname, but two families (BRCA2 7990del3ins2 and BRCA1 4987C>G) had first cousins who were tested without knowledge of the other (Fig. 1 and data not shown).

Table 1. Most commonly observed non-Ashkenazi Jewish mutations
BRCA1 mutationaTimes observedbBRCA2 mutationaTimes observed
  1. a

    Breast Cancer Information Core (BIC) nomenclature.

  2. b

    Number of separate families.

2800delAA65910C>G6
3875delGTGC58395G>C6
Exon13ins6kb56503delTT6
300T>G (C61G)57297delCT5
2576delC4IVS15-1G>A5
2722C>G41983del53
1294del4032041insA3
4987C>G32157delG3
3600del1135578delAA3
3889delAG32041delA3
  9610C>T3
  7990del3ins23
image

Figure 1. Linking individuals through pedigree and mutation status. Four probands from two families, represented in panels (a) and (b), were identified as being related due to shared mutation and common ancestors. The mutation in the family shown in panel (a) is BRCA2 7990del3ins2 and the mutation in the family shown in panel (b) is BRCA1 A1623G. In both of these families, first cousins were seen for cancer genetics risk assessment and genetic testing without knowledge of previous genetic testing results in the family.

Download figure to PowerPoint

Cost implications

Because knowledge of mutational status of distant family members can impact cost of testing and screening, we estimated HBOC-related costs for relatives. We compared estimated cost differences for a 32-year-old female with a strong family history of breast cancer who was not found to carry a mutation in BRCA1 or BRCA2 (Table 2) to 32-year-old female who tested negative for a known family mutation. Current list prices for testing through Myriad Genetics, Inc. are $475 for single site analysis and $3350 for Comprehensive BRACAnalysis. Breast cancer screening recommendations were based on the National Comprehensive Cancer Network guidelines for breast cancer screening and diagnosis for a woman with a >20% lifetime risk of breast cancer. [10, 11] From these estimates, a 32-year-old woman who undergoes Comprehensive BRACAnalysis because of a strong family history could have genetic testing and screening costs of $4346, whereas the same woman who tests negative for the family's known BRCA mutation would have testing and screening costs of roughly $545, almost 8-fold less (Table 2).

Table 2. Incurred genetic testing and breast cancer screening costs for a theoretical case involving uninformative negative BRCA test resulta
Procedure/testSourceEstimated costRecommended if family BRCA mutation knownRecommended if unaware of BRCA mutation in family
  1. HCPS, Healthcare Common Procedure Coding System; MRI, magnetic resonance imaging.

  2. a

    Cancer screening recommendations and estimated cost differences for a 32-year-old female who is tested for BRCA gene mutations because of strong family history (lifetime risk greater than 20% based on family history) and has a negative test result. HCPCS costs are estimated based on 2012B estimates from the Centers for Medicare and Medicaid Services Fee Schedule.

Comprehensive BRCA analysisList price$3350NoYes
Single site BRCA analysisList price$475YesN/A
Annual mammogramHCPCS: G0202$140NoYes
Annual breast MRIHCPCS: 77059$716NoYes
Annual clinical breast examHCPCS: 99213$70YesYes (×2)
Total (1 year cost)  $545$4346

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Our findings have implications for genetic counseling of individuals undergoing BRCA1 and BRCA2 analysis and could be applicable to other hereditary cancer syndromes. This analysis has shown that multiple individuals who are of relatively close relationship were unaware of genetic testing outcomes in their family. The linking of related families is crucial for appropriate screening and management but is also important in relationship to cost benefit as site-specific testing is much less expensive than comprehensive BRCA analysis.

In the theoretical scenario presented, not only are increased costs associated with the comprehensive gene analysis (compared to site-specific analysis), but also the costs of high-risk cancer screening and prevention could be unnecessarily incurred. The cost of screening for a woman with a true negative result could be as little as an 8th the cost of a woman who is following screening recommendations in the high-risk setting unnecessarily. The Medicare Physician Fee Schedule was used as a reference given the variability of fees between facilities and providers. Many young women requiring high-risk screening do not have Medicare/Medicaid and the amount billed to their insurance or the patient could be significantly higher. Additionally, these codes are rarely billed in isolation and are often associated with other technical and professional fees that are not included.

A recent study of the knowledge and attitudes of individuals involved with an advocacy group for HBOC about genetic discrimination found that the most commonly cited reason for not pursing genetic testing was cost [11]. On the basis of this information, it is possible that some individuals who have been offered genetic testing and declined would choose differently if a mutation had already been identified in their family. Knowledge of a familial mutation BRCA could also add to the perceived importance of the testing and greatly reduce the likelihood of an uninformative result.

In addition to the cost benefits associated with up-front knowledge of a familial BRCA mutation, this information could have a positive clinical impact for family members who receive ‘true negative’ results by allowing for appropriate management recommendations and informed decision making. Given promising reports from early PARP inhibitor trials, knowledge of BRCA mutation status may also guide chemotherapeutic treatment options in the future [12].

There are some limitations to this study. Because information collected clinically about distant relatives differs and information provided differs among patients, it is possible that some ‘linkable’ pedigrees were not identified. Likewise, we classified two families as ‘potentially linked’ because it was not clear if the likely common ancestor actually had the BRCA gene mutation because they were never tested and were deceased. We may also have missed a similar number of ‘linkable’ families simply because the proband of a given branch tested negative for the ancestral BRCA mutation and thus was not included in our analysis. As three-generation pedigrees are standard for clinical risk assessment in cancer genetics [3], more distant relationships may have gone undetected. This implies that some individuals who should be considered to be ‘true negatives’ in mutation-positive families are unaware and are likely participating in high-risk cancer screening programs based on strong ‘BRCA-like’ family history and their residual risks of cancer.

Clearly, sharing of genetic test results with relatives is critical. While there have been many discussions surrounding a health care professional's duty to warn family members about cancer genetic testing performed in the family, it is generally considered necessary when there is a high likelihood of harm if the relative(s) are not warned, the relative(s) are identifiable, or resulting from failure to disclose results would outweigh the harm of the disclosure. Complicating matters further is that breaching confidentiality could violate protections provided by the Health Information Portability and Accountability Act (HIPAA). In most cases, it is considered the duty of the healthcare provider to discuss the importance of sharing results with family members and encouraging the patient to follow through. Genetic counselors typically offer to aid families in sharing information with relatives such as through sample letters, but it may be beneficial to take a more active approach. This could include providing a letter to share with relatives, encouraging patients to bring family members with them to genetics appointments, and the development of electronic education tools that patients and family members both could utilize. Clinicians should also address common barriers to sharing information directly during result disclosure.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

This study was supported in part by the OSU Comprehensive Cancer Center. Carolyn Craven, Erin O'Toole, and Marguerite Goulet helped to accrue study participants and record patient data. We thank Katherine Nathanson and the CIMBA consortium members for providing BRCA1 and BRCA2 mutation frequencies. We thank the families for participating in our research studies.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
cge12211-sup-0001-TableS1.docWord document33KTable S1:Demographics of OSU CCG BRCA+ individuals.
cge12211-sup-0002-TableS2.docWord document33KTable S2:Cancers observed in carriers other than breast and ovarian.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.