the Dana-Farber Cancer Institute
    Departments of Psychiatry and Medicine, Harvard Medical School, Boston MA
    Faculty of Pharmacy, Universite Laval, Québec, Canada.

    ABSTRACT

    PURPOSE: Patient communication with relatives about cancer genetic test results is the primary means for alerting those who may benefit from identification of hereditary risk. This study identifies factors predicting patterns of disclosure of BRCA1/2 test results to first-degree relatives (FDRs) among women tested in a clinical protocol.

    PATIENTS AND METHODS: A total of 273 women completed a family communication measure 4 months after BRCA1/2 result disclosure. 2 analyses and logistic regression models identified factors predicting sharing of the test result.

    RESULTS: Most FDRs were informed of the participant's test result by 4 months; female relatives were more likely to be informed than males. Tested women conveyed inconclusive results (variant or negative without known familial mutation) less frequently to their sisters than conclusive (positive/true negative) results (P = .03). Twenty-three percent of participants did not inform their father. Informing brothers was more likely when BRCA1/2 was inherited through paternal lineage (P = .04), but 29% of brothers were not informed. Women older than age 40 were less likely to share their result with their parents (P = .03) than were women  40. Children's ages influenced communication to offspring; most children were told.

    CONCLUSION: Demographic, health-, and test-related factors predicted genetic test result communication to FDRs. Additional research investigating the full spectrum of discussion within families and motives for incomplete sharing of genetic test results with relatives may suggest strategies for providers and targeted educational interventions for patients to enhance family communication.

    INTRODUCTION

    The distinguishing feature of genetic medicine is that the discovery of one person's disease or predisposition status has risk implications not only for that individual but also for his or her relatives. This is especially true for the patient's first-degree relatives (FDRs; ie, his or her siblings, parents, and children), who have the highest probability of sharing the altered allele. The transmission of genetic information to a patient's family is usually accomplished by one-to-one communication between the patient and each relative. Given the nature of modern families and the complexity of the information itself, this communication can be difficult or can even be deemed impossible in some circumstances by some patients. Patients bear primary responsibility for sharing their information with relatives, but when this fails, ethical issues may arise about the responsibilities of providers.1 Understanding how families communicate about hereditary risk information is important to considerations about the role of providers in enhancing family communication and in planning for genetic services.

    This descriptive study examines the sharing of BRCA1/2 results with all categories of FDRs among a sample of women at the Dana-Farber Cancer Institute (Boston, MA) and collaborating hospitals. We sought answers to the following questions. What is the extent of sharing of the result of one individual's BRCA1/2 testing with FDRs by 4 months postdisclosure. What factors predicted informing FDRs of BRCA1/2 results By examining the entire cohort of FDRs, we hoped to learn not only about patterns of family communication about genetic test results, but also to determine if there were systematic gaps in communication of BRCA1/2 results that might become the target for subsequent interventions to enhance communication. Potential predictors of whether participants inform FDRs were selected based on the nature of the medical impact of BRCA1/2, the existing literature, and our clinical experience.

    PATIENTS AND METHODS

    Study Participants

    The participants in this study were drawn from the 310 women participating in Project GRACE (Genetic Risk Awareness and Cancer Education; a National Human Genome Research Institute–funded randomized study of two genetic counseling interventions) who enrolled onto the program between December 1, 1998, and July 1, 2001. Initially, 442 women had been invited from Cancer Genetic Risk Assessment clinics or clinical oncology practices to participate; 132 declined participation. Of the 310 participants in Project GRACE, 298 women received their BRCA1/2 test result, of whom 274 completed both the enrollment packet and 4-month postdisclosure family communication questionnaire. Of that group, one woman did not have any living first-degree relatives, so the final analysis is based on 273 women.

    For the overall study, women were required to meet the following eligibility criteria: age 18 or older; personal or family history of breast, ovarian, or other cancer consistent with BRCA1/2 heredity with posterior probability of carrying an altered gene of  10% based on published probabilities or Bayesian calculations; and documentation of participant or family member cancer diagnosis. Project staff randomly assigned each participant to either a standard genetic counseling or an enhanced consent intervention administered by a trained medical oncology nurse. Both models included review of the patient's family history and personal risk estimation, description of the test procedure and, at minimum, some description of potential psychological and family issues related to the dissemination of genetic information. The genetic counseling intervention included an exercise in which the counselee was asked to imagine the experience of disclosing (or not disclosing) their result to certain relatives. Therefore, individuals in this intervention arm might have been expected to have thought through the sharing of results with relatives more carefully and to have been more conscious of adverse as well as positive outcomes. Participants were required to attend two visits at their study site, and received their BRCA1/2 test result during their second appointment. Enumeration of relatives at 25% to 50% risk of sharing a mutation with a participant testing positive, and the implications of the participant test result for close relatives were reviewed as part of the results disclosure session in both interventions. Genetic test charges were subsidized or covered by the project. The institutional review boards at the Dana-Farber Cancer Institute and participating institutions approved this study.

    Measure

    Sharing of test results was assessed using the Family Communication Measure (FCM), a self-report instrument designed by the authors to capture whether the participant informed or did not inform relatives after genetic testing. Pilot testing was carried out with 10 at-risk individuals to determine acceptability and clarity. The FCM was mailed to participants 4 months after results disclosure.

    The FCM asked participants to identify their living FDRs, listing relatives in age order within relationship categories with multiple entries. Participants were asked to indicate whether they had shared their test result with each listed relative. Instructions were given in a manner that conveyed neutrality about the informing family members.

    Statistical Analyses

    Analysis of the family communication data included calculation of the proportion of the participants' parents, siblings, and children with whom they had shared their test result. In each analysis group, the number of participants involved varied depending on the number of living relatives of the type being considered. Separate logistic regression models were used to identify individual and familial factors predicting the sharing of a participant's test result with each relative group.

    Factors considered a priori in the regression included BRCA1/2 test result (positive, negative, or inconclusive), participant age and cancer status, lineage carrying the BRCA1/2 mutation as determined from family history (ie, inherited via the maternal v paternal lineage), and randomized counseling intervention (genetic counseling v nursing consent session). Indeterminate negative results (ie, testing negative without a known familial mutation) and variants of unknown significance results were grouped together as inconclusive results in this study. Cancer status of the living parent was added for the analysis of informing the mother or father. For the analysis of disclosure to offspring, sex and ages of the participant's children (grouped as ages 6 to 13, 14 to 17, 18 to 30, and older than 30 years) were included. We anticipated that the participants might be more likely to share results with female relatives given that women are typically the kin-keepers2 (ie, the guardians of family history, including medical family history), and specifically, because BRCA1/2 confers less cancer risk to men. We also hypothesized that participants who received inconclusive results would be less likely to inform relatives because of the increased difficulty and complexity of interpreting the result. Furthermore, we also anticipated differences in family communication between participants who experienced different counseling models, although the direction of that difference was not predicted.

    Before carrying out the regression analysis, we examined whether there was collinearity among all the potential predictors using the procedure described in the SAS REG procedure (SAS Institute, Cary, NC).3 After considering the condition indices, the proportion of variance of the estimate accounted for by each principal component, and the variance inflation factors, we concluded that collinearity was not a problem in this data set. Interest in exploring further the telling of sisters led to a univariate analysis comparing the telling of conclusive and inconclusive test results to sisters. To control for family clusters, statistical modeling was carried out using the generalized estimated equation approach.4 Statistical analyses were conducted using SAS software.5

    RESULTS

    Participant Characteristics

    Participants were all white women, of whom 92% had at least some college education, and 60% had had either breast or ovarian cancer (Table 1). Participants across the age range had come for testing, with nearly equal representation of women in the age groups  40, 41 to 50, and older than 50 years. The 273 participants were from 244 families (mean, 1.12 participants per family) and 204 participants (75%) were the first in their family to undergo BRCA1/2 testing. The majority of participants (81%) were married, 11% were single, and 8% were divorced/widowed. Approximately half of our sample (52%) reported annual household incomes of $95,000 or more. Twenty-five percent of the group (n = 68) received a positive BRCA1 or BRCA2 result, 12% (n = 34) had true-negative results, and 63% (n = 171) had inconclusive results, showing either an unclassified variant or a negative result in a family without an identified mutation.

    There were no statistical differences between women who responded to the FCM at 4 months and those who did not in age, cancer status, education, income, marital status, race/ethnicity, or counseling randomized assignment (data not shown). The characteristics of the women in this sample generally are similar to those of women who have sought BRCA1/2 genetic testing.6,7

    With Whom Are Results Shared by 4 Months Postdisclosure

    Test results were told to female relatives more often than to males in all groups. Thus, mothers were told in higher proportion than were fathers (P = .05; Fig 1), sisters were told in higher proportion than were brothers (P < .0001; Fig 2), and daughters were told in higher proportion than were sons (P = .02; Fig 3).

    Factors in the Telling of One's BRCA1/2 Result to FDRs

    Table 2 summarizes the significant findings from our logistic regression models of the informing of parents, siblings, and children of test results within 4 months of disclosure. The a priori factors that did not predict the sharing of BRCA1/2 genetic test results with any group of FDRs included the mother's cancer status, the participant's cancer status, and the counseling intervention assignment. Factors predictive of test result are discussed in the following three sections.

    Informing parents. Older women (age older than 40 years) were less likely to tell their result to their mother (P = .03) and their father (P = .03) than were younger women age  40 years. When the participant's father had had cancer (type unspecified), participants were more likely to share their test result with their father (P = .04).

    Informing siblings. Participants were more likely to inform their brother(s) of their test result in families in which the BRCA1/2 mutation was inherited through the paternal line (P = .04). Univariate analysis examining the telling of sisters revealed a significant difference between the 18% of sisters (28 of 159) who were not informed of the participant's inconclusive result and the 9% (12 of 134) who were not informed of a positive or true-negative result (P = .03; data not shown).

    Informing children. In the logistic regression analysis, the telling of female children more than male children was statistically significant (P < .01). As anticipated, children ages 6 to 13 years were told their mother's BRCA1/2 test result less often (37%) than children who were  30 years old (93%; P < .01). However, there was no significant difference in the telling of adolescents age 14 to 17 (85%) or young adults age 18 to 30 (92%) compared with the telling of children age 30 or older. There was no significant interaction between age and sex of children told.

    DISCUSSION

    Data on which FDRs of women tested for BRCA1/2 are not informed of the test results have implications for the medical care of those individuals and also for physicians' duty to warn uninformed relatives about their hereditary cancer risk.8-10 Our study suggests the extent of nondisclosure may be more significant than previously considered,10 and that providers of genetic services are likely to encounter patients who will not notify some relatives for whom genetic information has significant implications.

    A range of factors figure into a complex equation differentiating those informed from those not informed. Like other studies of hypothetical and actual result communication, we observed that participants shared their result more often with female than male FDRs.10-12 Although the breast and ovarian cancer risk conveyed by BRCA1/2 mutations to female carriers is considerably greater, there are a number of cancer risks associated with BRCA1/2 mutation status in males.13-15 There are also, of course, direct implications for the female offspring of male carriers. The interest of men in BRCA1/2 testing and their deeply felt concern for the risks faced by their female relatives has been documented in a number of studies.16-19 It is possible that there is an underestimation of the desire of male members of BRCA1/2 families to be made aware of details of their family's hereditary breast/ovarian cancer risk. Communication about genetic test results in hereditary breast cancer families may improve through the development of materials targeted to informing men about a topic that is often thought of as an issue of concern only to females.

    Sex is not a complete predictor of family communication, however, as shown by the fact that 14% of participants' sisters were not told. The degree to which the sharing of test results is dependent on perceived direct medical benefit to the recipient may become clear in studies of family communication of genetic test results with roughly equal medical implications for men and women, such as hereditary nonpolyposis colorectal cancer testing, or results of concern predominantly to males, such as testing for mutations in prostate cancer genes.

    The conclusiveness of the result also seems to influence the sharing of a BRCA1/2 result. In a study of 43 sister pairs, Hughes20 reported that women tested for BRCA1/2 conveyed positive results to sisters more often than they conveyed inconclusive results. In our study we found similar trends for most categories of FDRs, although the difference reached significance only for communication to sisters. Participants may have been less likely to communicate their inconclusive result to relatives because the complexity of the message increases as the conclusiveness declines, or because of their perception that telling would be of little or no use to relatives.

    We also found that younger women were more likely to tell their test result to their parents than were women older than age 40, a pattern previously reported in the telling of third-degree relatives.11 Older women may wish to protect their typically older parents from unpleasant news16,21 and may consider older parents less able to understand the test result. Younger women may be emotionally closer to their parents and extended family. In addition, women younger than age 40 may have mothers who could still be at significant hereditary cancer risk themselves if they are carriers of BRCA1/2 mutations. In future research it may be useful to distinguish between telling as explanation to relatives who are, by virtue of their age or sex, at relatively low risk for hereditary cancer (such as the 70-year-old father of a BRCA1/2 female mutation carrier with no other daughter) and telling as warning to those who themselves or through their female offspring have much to gain from this information, given that they currently face significantly increased cancer risks (such as the 42-year-old mother of a BRCA1/2 mutation carrier).

    To our knowledge, ours was the first study of family communication to investigate the influence of medical factors on the informing of relatives. Of particular interest is our finding that brothers were told more often about a sister's BRCA1/2 test result in families in which the mutation is inherited through the paternal side. It seems likely that this reflects a misperception that BRCA1/2 hereditary risk differentially affects men in paternal lineage families. However, social reasoning rather than scientific reasoning is not uncommon in the lay understanding of genetics. Richards and Ponder,2 for example, found that lay people thought they shared more of their genes with their children than with their siblings, because their emotional bonds were often stronger with their children.

    Distortions about the meaning of the line of inheritance of BRCA1/2 mutations could also reflect incorrect information received from health care providers. A survey of physicians (internists, oncologists, and gynecologist/obstetricians) found only 36% knew that transmission of hereditary susceptibility for breast/ovarian cancer can come through the male lineage.22 Nonetheless, the increased telling of results to brothers in patrilinear families is particularly important because communication in hereditary cancer families has been described as least extensive between brothers and sisters.23 Primary care physicians can play important roles in correcting misperceptions of patients about the role of patrilinear inheritance of breast/ovarian cancer risk and in encouraging the wide dissemination of test results to male relatives.

    Participants in our study were more likely to communicate their result to fathers who themselves had had some type of cancer, possibly because lines of communication about cancer may have been opened previously during the father's illness or because the father is perceived as more interested in cancer-related topics. The results may be less marked with regard to telling mothers who have had cancer because discussion between mothers and daughters in breast/ovarian cancer families may already be extensive, regardless of the mother's cancer status.

    Sensitive ethical and psychological questions are raised by consideration of the ages at which children should be told about their mother's genetic test result.24 Our research supports that age per se is not the major consideration in the telling of children older than age 13, given that adolescents age 14 to 17 were told at rates that were not different from rates adult children were told. This is in line with research that established that children older than age 13 were capable of evaluating medical information as well as adults.25

    Finally, it was somewhat surprising that the positive versus true-negative nature of the test result was not a significant factor in the telling of FDRs, except in the telling of results to children. Although this could be a function of our sample size, others have also reported that mutation status per se does not explain communication between family members about test results.26 It seems increasingly clear that the earlier assumption that the absolute nature of an individual's test would be the major factor predicting the telling of relatives was simplistic and did not reflect the psychological complexity of family relationships.19

    Our methods have several limitations that we have taken into account in assessing our findings. We relied on participant self-report and did not explicitly verify that the information actually was shared, nor did we assess what was said. Relatives could also have been made aware of the result by other family members. Thus, the actual degree of noncommunication could be either higher or lower than that reported here. Future studies assessing indirect communication among family members including non–blood relatives, although valuable, will be methodologically and ethically challenging, especially given current Health Insurance Portability and Accountability Act regulations. In addition, we did not seek information about the informing of relatives after 4 months postdisclosure, although time since disclosure was not a factor in the telling of relatives in a Belgian study in which participants were from 6 to 29 months postdisclosure.11

    Our study enrolled only women. It would be of great interest to compare data on whom men tell about their BRCA1/2 test results and what facilitates or impedes telling. Our sample was not culturally diverse, which limits generalizability of the findings to heterogeneous populations. This has been a problem in much of the research on cancer genetic testing; in general, the populations that have been tested to date have been white. Finally, although our sample size was relatively large compared to that of many other studies in this area, our power to detect differences may have been limited.

    Our study describing gaps in the sharing of BRCA1/2 results with the full gamut of FDRs identified medical factors that affect family communication, and refined our consideration of age, sex, and test result as other predictive factors. Future research could help to further differentiate factors in noncommunication of test results, including patterns of direct and indirect communication within families of diverse cultural backgrounds, relationship of communication patterns to family cohesion and emotional connection between members, intrinsic aspects of the genetic information itself, medical and demographic characteristics of patient and recipient, and/or the patient's perceptions of the immediate utility of the information to the recipient. Such knowledge will be useful in assisting individuals and families undergoing genetic testing to optimize family communication, and may also reduce potential ethical burdens on genetics professionals.

    Authors' Disclosures of Potential Conflicts of Interest

    Althoguh all authors completed the disclosure declaration, the following author or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

    AuthorsEmploymentLeadershipConsultantStockHonorariaResearch FundsTestimonyOther

    Judy E. GarberNovartis Pharmaceuticals (A)

    Dollar Amount Codes (A) <$10,000 (B) $10,000-99,999 (C) $100,000 (N/R) Not Required

    Author Contributions

    Conception and design: Andrea Farkas Patenaude, Michel Dorval, Lisa S. DiGianni, Judy E. Garber

    Financial support: Judy E. Garber

    Administrative support: Judy E. Garber

    Provision of study materials or patients: Lisa S. DiGianni, Judy E. Garber

    Collection and assembly of data: Lisa S. DiGianni, Katherine A. Schneider, Anu Chittenden, Judy E. Garber

    Data analysis and interpretation: Andrea Farkas Patenaude, Michel Dorval, Lisa S. DiGianni, Katherine A. Schneider

    Manuscript writing: Andrea Farkas Patenaude, Michel Dorval, Lisa S. DiGianni, Katherine A. Schneider, Judy E. Garber

    Final approval of manuscript: Andrea Farkas Patenaude, Michel Dorval, Lisa S. DiGianni, Katherine A. Schneider, Anu Chittenden, Judy E. Garber

    Acknowledgment

    We thank Michel Gaudet, MSc, for statistical analyses and Kylie Smith, MS, for her assistance with database management. Other Project GRACE (Genetic Risk Awareness and Cancer Education) investigators who contributed to this research include Caroline Block, MD, Kevin Hughes, MD, Joel Schwartz, MD, Michael Seiden, MD, and Tracey Weisberg, MD.

    NOTES

    Supported by Grant No. 5RO1HG01244 from the Ethical, Legal, and Social Implications (ELSI) branch of the National Human Genome Research Institute, National Institutes of Health. M.D. currently holds a "Chercheur-boursier" award from the Fonds de la Recherche en Santé du Québec and is supported by the INHERIT BRCAs research program funded by the Canadian Institutes of Health Research.

    A.F.P. and M.D. contributed equally to this work.

    Presented in part in a poster at the Annual Meeting of the American Society of Clinical Oncology, New Orleans, LA, June 5-8, 2004.

    Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

    REFERENCES

    Leung W-C: Results of genetic testing: When confidentiality conflicts with a duty to warn relatives. BMJ 321:1464-1466, 2000

    Richards M, Ponder M: Lay understanding of genetics: A test of a hypothesis. J Med Genet 33:1032-1036, 1996

    Besley D, Kuh E, Welsch RE: Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. New York, NY, John Wiley & Sons, 1980, pp 310

    Ziegler A, Blettner M, Kastner C, et al: Identifying influential families using regression diagnostics for generalized estimating equations. Genet Epidemiol 15:341-353, 1998

    SAS Institute: SAS/STAT User's Guide, Version 6. Cary, NC, 1990

    Lerman C, Narod S, Schulman K, et al: BRCA1 testing in families with hereditary breast-ovarian cancer: A prospective study of patient decision-making and outcomes. JAMA 275:1885-1892, 1996

    Bottorff JL, Ratner PA, Balneaves LG, et al: Women's interest in genetic testing for breast cancer risk: The influence of sociodemographics and knowledge. Cancer Epidemiol Biomarkers Prev 11:89-95, 2002

    Keeling SL: Duty to warn of genetic harm in breach of patient confidentiality. J Law Med 12:235-253, 2004

    Offit K, Groeger E, Turner S, et al: The "duty to warn" a patient's family about hereditary disease risks. JAMA 292:1469-1473, 2004

    Julian-Reynier C, Eisinger F, Chabal F, et al: Disclosure to the family of breast/ovarian cancer genetic test results: Patients' willingness and associated factors. Am J Med Genet 94:13-18, 2000

    Claes E, Evers-Kiebooms G, Boogaerts A, et al: Communication with close and distant relatives in the context of genetic testing for hereditary breast and ovarian cancer in cancer patients. Am J Med Genet 116A:11-19, 2003

    Costalas JW, Itzen M, Malick J, et al: Communication of BRCA1 and BRCA2 results to at-risk relatives: A cancer risk assessment program's experience. Am J Med Genet 119C:11-18, 2003

    Liede A, Karlan BY, Narod SA: Cancer risks for male carriers of germline mutations in BRCA1 or BRCA2: A review of the literature. J Clin Oncol 22:735-742, 2004

    Lynch HT, Deters CA, Snyder CL, et al: BRCA1 and pancreatic cancer: Pedigree findings and their causal relationships. Cancer Genet Cytogenet 158:119-125, 2005

    Breast Cancer Linkage Consortium: Cancer risks in BRCA2 mutation carriers. J Natl Cancer Inst 4:91:1310-1316, 1999

    d'Agincourt-Canning L: Experiences of genetic risk: Disclosure and the gendering of responsibility. Bioethics 15:231-247, 2001

    McAllister MF, Evans DG, Ormiston W, et al: Men in breast cancer families: A preliminary qualitative study of awareness and experience. J Med Genet 35:739-744, 1998

    Lodder L, Frets PG, Trijsburg RW, et al: Men at risk of being a mutation carrier for hereditary breast/ovarian cancer: An exploration of attitudes and psychological functioning during genetic testing. Eur J Hum Genet 9:492-500, 2001

    Smith KR, West JA, Croyle RT, et al: Familial context of genetic testing for cancer susceptibility: Moderating effect of siblings' test results on psychological distress one to two weeks after BRCA1 mutation testing. Cancer Epidemiol Biomarkers Prev 8:385-392, 1999

    Hughes C, Lerman C, Schwartz M, et al: All in the family: Evaluation of the process and content of sisters' communication about BRCA1 and BRCA2 genetic test results. Am J Med Genet 107:143-150, 2002

    Bonadona V, Saltel P, Desseigne F, et al: Cancer patients who experienced diagnostic genetic testing for cancer susceptibility: Reactions and behavior after the disclosure of a positive test result. Cancer Epidemiol Biomarkers Prev 11:97-104, 2002

    Escher M, Sappino A-P: Primary care physicians' knowledge and attitudes towards genetic testing for breast-ovarian cancer predisposition. Ann Oncol 11:1131-1135, 2000

    Kenan R, Arden-Jones A, Eeles R: Healthy women from suspected hereditary breast and ovarian cancer families: The significant others in their lives. Eur J Cancer Care 13:169-179, 2004

    Tercyak KP, Peshkin BN, DeMarco TA, et al: Parent-child factors and their effect on communicating BRCA1/2 test results to children. Patient Educ Counsel 47:145-153, 2002

    Weithorn L, Campbell S: The competency of children and adolescents to make informed consent decisions. Child Dev 53:1589-1598, 1982

    Koehly LM, Peterson SK, Watts BG, et al: A social network analysis of communication about hereditary nonpolyposis colorectal cancer genetic testing and family functioning. Cancer Epidemiol Biomarkers Prev 12:304-313, 2003 2

查询更多Women相关信息在本站>>

《临床肿瘤学医学期刊》2006年2月第24卷第2期