Do Breast Cancer Patients Tested in the Oncology Care Setting Share BRCA Mutation Results with Family Members and Health Care Providers?

BRCA genetic test results provide important information to manage cancer risk for patients and their families. Little is known on the communication of genetic test results by mutation status with family members and physicians in the oncology care setting. As part of a longitudinal study evaluating the impact of genetic counseling and testing among recently diagnosed breast cancer patients, we collected patients' self-reported patterns of disclosure. Descriptive statistics characterized the sample and determined the prevalence of disclosure of BRCA test results to family members and physicians. Of 100 patients who completed the baseline and the 6-month followup survey, 77 reported pursuing testing. The majority shared test results with female first-degree relatives; fewer did with males. Participants were more likely to share results with oncologists compared to surgeons, primary care physicians, or other specialty physicians. These findings suggest that while breast cancer patients may communicate results to at-risk female family members and their medical oncologist, they may need education and support to facilitate communication to other first-degree relatives and providers.


Introduction
Mutations in the BRCA1 and BRCA2 (BRCA) genes place individuals at higher risk of developing breast and ovarian cancer, compared to those without a mutation [1,2]. A recent study of cancer risks in BRCA mutation carriers in a large US-based sample estimated the cumulative breast cancer risk at age 70 years to be 46% in BRCA1 carriers and 43% in BRCA2 carriers. Cumulative ovarian cancer risk was 39% in BRCA1 carriers and 22% in BRCA2 carriers [3]. Although less well-established, elevated cancer risks in men from BRCA families have been reported in male breast cancer (6-8%) [4][5][6][7][8][9][10] and prostate cancer (20-30%) [5,8,[10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27]. Individuals who undergo genetic testing and discover that they carry a BRCA mutation can manage their cancer risk through intensive screening, prophylactic surgery, and/or chemoprevention [28][29][30][31]. Criteria set forth by the National Comprehensive Cancer Network to identify and refer breast cancer patients to a genetics professional [32] include a personal history of early onset breast cancer (i.e., diagnosed ≤ age 50), triple negative breast cancer, and/or ≥2 primary cancers. Women diagnosed with breast cancer at any age may also be referred when there is a family history of early onset breast cancer, ovarian cancer at any age, and/or male breast cancer. Additionally, women with a family history of breast cancer and cancers considered to be part of the Hereditary Breast Ovarian Cancer spectrum (e.g., pancreatic, thyroid) may also warrant genetic evaluation.
Clinical guidelines indicate the most informative clinical testing strategy is to first test a family member with cancer [33]. Then, if a familial mutation is identified, testing can be offered to unaffected family members to determine whether they have the already identified BRCA mutation [33]. If there was a previously identified mutation in the family and the proband's test way negative (i.e., true negative), the proband generally is considered to be at general population risk for cancer. In this situation, there is a high level of reassurance that BRCA mutations are not the underlying cause of cancer, which may reduce anxiety as well as unnecessary screening and surveillance in the proband and at-risk family members. Conversely, test results are considered uninformative if an individual is the first person in the family to have testing and receives a negative result or if an alteration in the BRCA genes is detected but the clinical significance for the alteration is unknown (i.e., variant of uncertain significance). In this situation, cancer risk is determined based on personal and family cancer history.
In order for genetic test results to maximally benefit and impact clinical management, test results must be communicated to others. To impact the health of family members, probands must share results with at-risk relatives [34]. For management of the proband's personal cancer risk, healthcare providers must also be aware of test results. The healthcare system's emphasis on patient confidentiality, clinical practices in which genetic testing results may not be included as part of the "main" medical record, and position statements from professional organizations such as the American Society of Clinical Oncology [35] and the American Society of Human Genetics [36] indicating that providers generally do not share genetic testing information and results with family members who are not their patients may limit the ability of providers to participate in dissemination of test results among family members. As such, the responsibility of communicating BRCA test results likely falls on the probands themselves. Prior studies of communication of test results to family members have found that the majority of participants share test results with family members, typically within the first few months of receiving test results. However, frequency of disclosure may vary based on test result as well as the family members' age, gender, and degree of relatedness (e.g., first-degree versus second-degree relative) [37][38][39][40][41][42][43][44][45][46][47][48]. Qualitative and quantitative studies also demonstrate that probands sometimes express difficulty in sharing results with at-risk relatives, particularly when relationships are emotionally distant or test results are uninformative [43,44,49].
An understudied aspect of communication regarding BRCA test results is whether results are shared with individuals outside of the family structure, namely, health care providers. Given that probands are often identified due to a personal cancer diagnosis and subsequently receive counseling and testing in the oncology care setting, it is important that as they transition into survivorship and shift health care back into the community setting, primary care providers are also made aware of BRCA test results [50,51]. Despite the important role that health care providers play in providing care to patients based on BRCA test results, only two studies have examined communication of BRCA test results to health care providers. In a study of 69 patients referred to a cancer risk assessment program, the majority of patients shared results with both oncology (64-88%) and primary care providers (74-81%) [50]. In a larger study of 312 patients who underwent BRCA testing at a large comprehensive cancer center, most (72%) shared genetic test results with health care providers outside of the oncology care setting [51].
Some data suggest patients affected with cancer may react differently than unaffected probands to the cancer genetic counseling and testing process [52][53][54] including communication of test results [40,51,55]. It is possible that having genetic counseling and testing following a cancer diagnosis may delay or reduce the likelihood that patients disseminate test results to at-risk family members due to both the physical and emotional stressors of diagnosis and treatment. Despite the arguable equal importance of affected probands sharing test results with both at-risk family members and health care providers, to our knowledge, no prior study has examined disclosure patterns to family members and physicians in a group of affected patients in a single study. To that end, the primary focus of our paper is to describe the frequency of communication of test results among breast cancer patients who have undergone genetic counseling and testing in the breast oncology care setting. The findings from this study will inform the development of comprehensive interventions that facilitate communication of test results by probands to both family members and health care providers.

Participant Recruitment.
This substudy regarding communication of BRCA test results is part of a larger longitudinal investigation of the impact of genetic counseling on psychosocial and behavioral outcomes among breast cancer patients referred for genetic counseling in the oncology care setting. Eligibility criteria for the larger study included: (a) meeting National Comprehensive Cancer Network cancer genetics referral criteria [32], (b) ≥18 years of age; (c) confirmed personal breast cancer diagnosis based on medical records review; (d) no previous participation in genetic counseling (GC) and/or testing for hereditary breast and ovarian cancer (HBOC); (e) capable of speaking and reading standard English; (f) having a mailing address and working telephone number; and (g) having a GC appointment scheduled at Moffitt Cancer Center.
Upon Institutional Review Board approval, recruitment took place between April 2009 and July 2010 with followup completed in February 2011. Data were collected: (a) after scheduling but prior to the pretest genetic counseling appointment (T1), (b) within two to three weeks after participants completed pretest genetic counseling (T2), and (c) six months after completing genetic testing (T3). The current report is focused on the communication of BRCA mutation test results; therefore, analyses are based on T1 (sociodemographic and clinical) and T3 (communication of test results) data.
All patients in the current study received in-person pretest GC by a Masters-level prepared genetic counselor through the clinical GC and testing service at the Cancer Center. For those who proceeded with testing, the majority received testing by the sole clinical laboratory for BRCA Journal of Cancer Epidemiology 3 testing in the US. Those who were uninsured or unable to obtain insurance coverage for testing were informed of a research study at another institution where they were able to receive BRCA testing free of charge. Those who proceeded with testing were subsequently scheduled for either an inperson or telephone-based posttest results disclosure session. Those patients then received followup letters summarizing their pre-and posttest GC sessions.
The study team reviewed the GC appointment schedule weekly for women meeting study eligibility criteria. Eligible patients were mailed an introductory letter with a toll-free number to opt out of further contact by the study team, the T1 survey, two consent forms, and a preaddressed envelope. Approximately one week from the mailing date, patients who did not opt out were contacted via telephone to confirm receipt of study materials and to answer any questions about the study. For those not reachable by telephone prior to their scheduled GC appointment, the study coordinator met briefly with patients after their GC session to determine whether the T1 survey was complete. Those who did not complete the T1 survey before attending their pretest GC session were considered decliners.
For the first six weeks of study recruitment, patients who failed to attend their pretest GC appointment were considered ineligible. However, this strategy precluded the opportunity to include patients who rescheduled and attended their GC appointment. Thus, recruitment procedures were revised so that patients who scheduled a new appointment were mailed an additional introductory study packet. Patients who failed to reschedule their appointment between the date of their canceled appointment and July 2010 (end of recruitment period) were considered ineligible. Participants received a $25.00, $20.00, and $30.00 gift card for completing T1, T2, and T3, respectively. Those patients who completed the T1 and T3 assessments and reported BRCA genetic test results were included in the current analyses.

Measures
Sociodemographic and clinical characteristics obtained from patient questionnaires or medical records review included: current age, age at diagnosis, time since diagnosis, current stage of breast cancer (1, 2/3, 4, unstaged, other [e.g., unknown]), previous surgery (yes, no), and primary payor at diagnosis (private insurance, public insurance, no insurance, other). In addition, we classified BRCA mutation status as: positive, true negative (i.e., tested negative for known familial BRCA mutation), and uninformative (tested negative in the absence of a known familial BRCA mutation or had an indeterminate or variant of uncertain significance result). We excluded 15 people who indicated they either did not know or had an "other" test result from the analyses examining disclosure of test results to family members and physicians. Additional data collected via selfreport questionnaires included: education (completed high school or less; vocational school and some college; college graduate and beyond), total household income at time of diagnosis (≤35,000; >35-≤50,000; >50,000), marital status (married/living with partner, other), and race (Black, White, other).
Family communication of BRCA test results was assessed using an abbreviated version of the Family Communication Measure, developed by Patenaude and colleagues [43]. Respondents were asked to complete the following information for each first-degree relative : relative still living (yes, no), age, and status of result disclosure (shared, did not share).
Communication of results to health care providers was assessed with a single item developed for the current study where participants were first asked whether they shared their results with a health care provider (yes, no, do not know). Those who responded "yes" were asked to select the specialties for each of the providers with whom they shared the information (surgeon, oncologist, obstetrician/gynecologist [OBGYN], primary care doctor, doctor of another specialty, or other health care provider).

Data Analysis.
Descriptive statistics were calculated to characterize the sample and to determine the prevalence of disclosure of BRCA test results to family members and physicians. To determine the prevalence of BRCA-test result disclosure, the proportions of living relatives with whom test results were shared stratified by family member (e.g., mother) and participant BRCA status (positive, true negative, uninformative) were calculated. Statistical analyses were conducted using SAS (Version 9.2, Cary, NC).

Results
As shown in Figure 1, a total of 223 patients were identified by the genetic counselor as potential participants who had scheduled an appointment for GC. Of these, 87 patients did not meet eligibility requirements (e.g., previous participation in GC and/or testing for HBOC, rescheduled GC appointment more than three times, nonEnglish speaking). Of the remaining 136 eligible participants, 114 consented and completed the T1 survey, resulting in an 83.8% response rate. Of those 114, 100 completed the T3 assessment that included questions about communication of test results. Data analyses were conducted on 77 women who reported pursuing BRCA testing.
As detailed in Table 1, study participants were on average 48 years old at the time of diagnosis. At the time of the T1 survey, participants were on average 52 years old and a median of 9.7 months had passed since they were diagnosed. The majority was married or living with a partner (66.2%); white (81.8%); at least a college graduate (46.8%); and covered by private insurance (57.1%). The greatest proportion had an annual household income of more than $50,000 at the time of diagnosis (39.0%). Regarding clinical variables, the greatest proportion of participants (  As seen in Table 2, most probands communicated results with family members, regardless of the test results. Of participants who tested positive (n = 3), all shared their test results with their mothers, fathers, and sisters, whereas participants shared with half of daughters aged 18 or older and brothers, on average. Of those who tested true negative (n = 31), sharing of results with first-degree relatives ranged from an average of only 67% of daughters younger than age 18 up to 100% of sons younger than 18. Participants with uninformative results (n = 11) shared results with all sons or daughters age 18 years or older. In general, a greater proportion of participants reported sharing their test result with all female relatives compared to male relatives. Participants who tested true negative reported sharing results with an average of 90-100% of female relatives aged 18 or older compared to 83-89% of male relatives. Those with an uninformative result shared with an average of 80-100% of female adult relatives compared to 56-100% of male adult relatives. Regarding disclosure to minor children (age 18 years or younger), those who tested true negative shared their result with an average of 67% of minor daughters and 100% of minor sons. Those with an uninformative test result shared with half of minor daughters and none of minor sons. Figure 2 summarizes patients' reports of communication of genetic test results with health care providers. Overall, 74% shared their test results with their oncologist and 51% with their surgeon. They were less likely to share results with their primary care physicians (40% of OBGYNs and 49% of primary care physicians) or other specialty physicians (9%).

Discussion
Overall, our study suggests that frequency of disclosure of genetic test results by probands varies by mutation status, gender of at-risk relatives, and age of at-risk relatives. For health care providers, disclosure rates varied by specialty.
Although based on a limited sample size, our findings suggest variability in disclosure patterns based on mutation status. Women who tested positive were most likely to communicate results to both parents and all sisters (100%), compared to those who tested true negative or uninformative. It is possible that the differences between positives and true negatives may be due to the fact that women who were in the true negative category were being tested for a previously identified mutation in the family. Thus, some parents and sisters may have already been tested, making it less likely that patients would see the need to disclose test results to other relatives who already have definitive information about their mutation status. This explanation is supported by the observation that those who were true negative were more likely to share results with brothers and children (i.e., those not likely to have had prior testing) when compared to those  who tested positive. It is also possible that those who test negative for a known familial mutation may feel a sense of guilt in sharing results with relatives who are positive (most likely parents or sisters who have been tested) [56,57].
Similar to prior studies [43], those who received uninformative results were also less likely to share results with family members compared to those who tested positive. Women with these test results may have difficulty understanding and communicating test results to family members due to both cognitive and emotional factors. A few studies of women who received uninformative test results suggest that they are likely to express uncertainty, misinterpret their test results [58,59], or report negative emotional reactions such as anger and frustration [59]. Communication of uninformative test results may occur less often because of probands'concern that at-risk relatives would take the results to mean they were not at increased risk for breast cancer [42]. This is evidenced by a recent study of 39 relatives of probands who received uninformative test results; relatives' medical decisions were based on their own perceptions of increased cancer risk and generally did not correlate with information actually communicated to them by the proband [34].
As observed in prior studies [43,45,46,48,55], probands were more likely to disclose test results to female relatives in each test result category. This finding is likely due to the higher cancer risks and greater availability of prevention/risk management options for female patients. However, given the equal risks of transmitting BRCA mutations to male and female offspring, increasing documentation of male cancer risks [60], and emerging surveillance guidelines for male BRCA mutation carriers [61], it is critical that female patients understand the importance of sharing test results with all atrisk relatives.
Another interesting observation from our study that is consistent with recent investigations on this topic is the 6 Journal of Cancer Epidemiology  disclosure of test results to minor children. While there are specific recommendations against BRCA testing for minors [35,62], there are less clear guidelines regarding disclosure of parents' test results to children. Because there are no recommended surveillance or risk reduction options prior to age 25 for known BRCA mutation carriers, there has been debate about the possible negative sequelae associated with disclosing cancer risk to minor children. Yet, several studies have documented that parents share test results with their minor offspring [37,38,[63][64][65]. In our study, of the three women with positive BRCA test results, none had a daughter or son below the age of 18. Those who tested true negative shared results with an average of 67% of minor daughters and shared uninformative test results with at least one daughter. The largest published study to date on this topic included 253 parents who had undergone BRCA testing and 505 offspring. Not surprisingly, for those who shared true negative results, children often expressed relief. However, parents perceived distress more frequently among offspring learning about their parent's BRCA positive or variant of uncertain significance result [38]. Thus, consideration of developmentally appropriate psychosocial interventions to support children who learn of BRCA mutation status, particularly positive and uninformative results, is warranted.
With regard to sharing results with health care providers, it appears that women were more likely to share results with oncology care providers, particularly medical oncologists. Prior studies document that fewer women attend GC (and therefore pursue testing) prior to surgical treatment for breast cancer [66,67]. Thus, the sharing of results with medical oncologists likely reflects the greater proportion of women referred by medical oncologists who generally assume their care after surgery. This finding is also supported by several studies of provider utilization of GC and testing that suggest when compared to other oncology care physicians (e.g., surgeons, radiation oncologists), medical oncologists have higher levels of knowledge and utilization of BRCA testing [68][69][70]. It is possible that the rates of sharing results with providers in the oncology care setting may also reflect probands' assumptions that all providers from a single institution would be made aware of or have access to genetic test results. However, at some institutions, BRCA results may be kept out of the "main" medical record and not accessible to all physicians involved in a patient's oncology care [71]. The health professional providing preand posttest GC should also discuss the patient's preferences for informing the relevant oncology providers based on specific institutional policies about documentation of BRCA test results.
Overall, participants reported sharing test results less often with primary care providers compared to oncology care providers. The frequency of sharing results (40-49%) is similar to another study that examined sharing of BRCA genetic test results with nononcology health care providers (30-44%) [51]. It is noteworthy that although patients in our study were only ∼5 months from receiving test results, compared to the other study where participants may have been several years from testing (as early as 1997), rates of test result disclosure were similar for both populations; this suggests low rates of disclosure with primary care providers may persist over time. As time elapses, most breast cancer patients tested in the oncology care setting will transition into survivorship and their care will move back into the primary care setting [72]. Primary care physicians will assume more responsibility for prevention or risk management of secondary cancers in these patients. Thus, it is important to encourage breast cancer patients to share results with their primary care providers.
While our study presents important information about the frequency with which genetic information is shared with at-risk relatives and health care providers, findings should be considered in light of certain limitations. First, our sample size is small and did not allow for statistical analyses of differences in disclosure based on variables such as mutation status, gender of relative, or provider type. However, ours is the first study to focus on this issue exclusively among affected breast cancer patients and provides preliminary data to support larger studies focused on understanding and supporting disclosure of BRCA test results among patients receiving counseling and testing at or near the time of a breast cancer diagnosis. Second, we evaluated sharing of test results during a relatively short time period (∼5 months post receipt of BRCA test results). However, prior studies suggest that the majority of test result disclosures with firstdegree relatives occurs within weeks to a few months of when the proband receives their test results [43,47]. Similarly, we observed consistent rates of communication with health care providers between our study and one that included breast cancer survivors who were several years from the time of testing [51]. This implies that if communication does not occur early, it will not likely occur later. Third, we did not assess detailed communication beyond firstdegree relatives. However, given that first-degree relatives are at highest risk for inheriting a BRCA mutation and most likely to be informed by a proband's test results, we selected to focus on this group to minimize participant burden of a lengthy questionnaire that would be required to assess additional at-risk relatives. Our sample for the present study represents a subset of women from a larger study who remained in the study for the six-month assessment and were also willing to share test results. In this sense, selection bias should be considered when interpreting out results; it is possible that our results do not capture the full variability of disclosure patterns among women in this population. Finally, we did not directly assess whether patients had a previously identified BRCA mutation in the family. Patients who were undergoing testing for a known family mutation may not need to communicate results with other at-risk relatives given this information may have already been provided by the initial relative who tested positive.

Conclusion
Communication of BRCA test results is a critical step toward realizing the benefit of genetic technology to identify cancer risk both for individuals and family members. In light of current ethical and legal concerns regarding privacy of patient genetic information [71,73], probands are often responsible for sharing test results both to at-risk relatives and health care providers. Prior research suggests that, to a large extent, they are willing to accept and carry out that responsibility [74]. However, our study demonstrates key areas where breast cancer patients who choose to undergo GC and genetic testing may need further education and support to facilitate accurate and effective communication to inform at-risk family members' medical decisions and care while minimizing psychosocial burden and familial distress.