This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by van Dijk, S. Articles by Otten, W. Search for Related Content PubMed PubMed Citation Articles by van Dijk, S. Articles by Otten, W. Social Bookmarking                            What's this?

Clinical Characteristics Affect the Impact of an Uninformative DNA Test Result: The Course of Worry and Distress Experienced by Women Who Apply for Genetic Testing for Breast Cancer

Sandra van Dijk, Daniëlle R.M. Timmermans, Hanne Meijers-Heijboer, Aad Tibben, Christi J. van Asperen, Wilma Otten

From the Department of Medical Decision-making, and Center for Human and Clinical Genetics, Department of Clinical Genetics, Leiden University Medical Center, Leiden; Department of Social Medicine, Free University, Amsterdam; and the Department of Clinical Genetics, Erasmus Medical Center, Rotterdam, the Netherlands

Address reprint requests to Sandra van Dijk, MSc, Department of Medical Decision-making, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, the Netherlands; e-mail: dijk{at}lumc.nl

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
PURPOSE: DNA mutation testing for breast cancer usually yields an uninformative result, which is a negative result in the absence of a known BRCA mutation within the family. However, few data are available on the psychological impact of this result. Moreover, the clinical heterogeneity within this group has not yet been considered. This study provides prospective data about the course of cancer-specific worry and distress for different groups of test applicants.

PATIENTS AND METHODS: All DNA test applicants (n = 238) completed three questionnaires: before and 1 and 7 months after disclosure of a DNA mutation test. With repeated-measures analysis of variance, differences were assessed between BRCA1/2-positive women (n = 42), BRCA1/2–true-negative women (n = 43), and women with an uninformative test result (n = 153).

RESULTS: On the group level, women with an uninformative result seemed to be reassured after disclosure (P < .001), but to a lesser extent than those women who received a true-negative result. However, not all women with an uninformative result reacted similarly: higher levels of worry and distress could be explained by relatively straightforward clinical variables, namely a personal history of cancer (P .001) and a higher pedigree-based risk (P .005). Furthermore, these clinical variables determined whether these women were either comparable to women who received a true-negative result or to BRCA mutation carriers.

CONCLUSION: Women with an uninformative result form a heterogeneous group of test applicants. The subpopulation of those with both a personal history of cancer and a relatively high pedigree-based risk expressed the highest levels of worry 7 months after DNA testing.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Since the identification of the BRCA1 and BRCA2 genes, many individuals have requested genetic testing for hereditary breast and ovarian cancer. From meta-analytic studies we now know that BRCA mutation testing generally does not lead to a decline in patient well-being.^1,2

Those reviews are based on women who opt for informative testing. However, given that the known BRCA mutations only account for 20% to 25% of familial aggregation,³ the majority of women applying for BRCA testing receive an uninformative result. Data on this group of test applicants who receive a negative result in the absence of a known BRCA1/2 mutation are relatively scarce. Despite concern about the possible harmful effects of continuing uncertainty associated with the result, no increased levels of distress have been observed to date.⁴ However, one of the few studies that compared women who received an uninformative result with women who learned that they carry the high-risk BRCA1/2 mutation did observe the same levels of distress 6 months after disclosure.⁵

A possible explanation for these seemingly contrasting findings may be the clinical heterogeneity of the group of women receiving an uninformative result. Although a positive family history is commonly a prerequisite for DNA mutation testing, some families remain suspect with regard to a hereditary cancer syndrome after an uninformative result, whereas within other families a hereditary pattern of cancer transmission is less obvious. Thus, women tend to differ with regard to the extent of their pedigree-based breast cancer risk, which might have an important impact on levels of worry and distress. Another important distinction is whether a woman has a personal history of cancer. The aim of this study was to assess whether the pedigree-based familial risk estimation and the personal cancer history can explain cancer worry and distress among women who receive an uninformative DNA test result. In addition, we compared groups of women with an uninformative DNA test result with women who received either a true-negative or a positive DNA test result.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Data Collection and Genetic Counseling
The study comprised all women who made an initial appointment for familial breast cancer counseling at the Department of Clinical Genetics in Leiden or Rotterdam (the Netherlands) in the period 1998 to 2002. Ethical approval was obtained from the hospitals' research ethics committees. Eligible women were at least 18 years old and had not received genetic counseling elsewhere. Referrals for genetic counseling were based on national guidelines.^6,7

In the first consultation, DNA testing was offered for individuals from families in which a pathogenic BRCA mutation was previously detected, and for individuals in whom the probability of mutation detection was approximately 10% or more. Usually this is an affected family member.⁸ All women who opted for DNA testing were invited to an in-person counseling session about the personal implications regarding either their own result or the result of their affected family member when the DNA test became available. In this session, the pedigree-based familial lifetime risk was derived from the Claus tables,⁹ and was conveyed to patients. Four risk categories were distinguished: general population risk (approximately 10%); slightly increased risk (10% to 15%); moderately increased risk (15% to 30%); and highly increased risk (30% or more). Counselees were provided with a letter that summarized all of the information discussed.

All participants completed questionnaires at several points in time: T₁, a pretest-disclosure questionnaire, sent up to 2 weeks after the first counseling session (thus, when this first questionnaire was completed, women had received risk information already, in line with the standard counseling protocol); T₂, a postdisclosure questionnaire that was sent 1 month after women had received the summary letter; and T₃, a follow-up questionnaire that was sent 6 months after completion of the postdisclosure questionnaire.

Measures
Patient characteristics. Information about age, educational level, marital status, and number of children was collected. In addition, all relevant medical information was obtained from counselees' medical records.

Breast cancer worry. In all questionnaires, we assessed breast cancer–related worries with one single item, "During the last two weeks, how often did you worry about developing breast cancer (again)?" on a 4-point scale ranging from 1 = almost never to 4 = almost all the time.¹⁰ Throughout the text, we refer to this measure as worry.

Breast cancer–specific distress. In each questionnaire, we included the Impact of Event Scale (IES),¹¹ which assesses the level of intrusion and avoidance, tailored to breast cancer, on a 4-point scale, in which 0 = not at all, 1 = seldom, 3 = sometimes, and 5 = often. The reliability of the scale was good (Cronbach's range, .89 to .92). We refer to this breast cancer–specific distress in the text as distress or IES.

Statistical Analyses
The SPSS statistical package version 11.5 (SPSS Inc, Chicago, IL) was used to analyze the data. Repeated-measures analysis of variance was used to assess differences between and within groups regarding courses of worry and distress.

For comparisons among women who receive an uninformative result, we dichotomized the personal risk estimation into relatively low risk (ie, < 30%) or relatively high risk ( 30%). This cutoff was chosen in line with the cutoff points of the Claus tables,⁹ and to obtain an optimal distribution. However, all women with a relatively low risk had a sufficiently high risk to receive DNA testing. In conjunction with having had a personal history of breast or ovarian cancer (yes or no), we first tested with repeated-measures analysis of variance whether significant effects of these two variables could be detected. Subsequently, on the basis of these two clinical variables we created four different groups of women with an uninformative result and compared them with women with either a positive or a true-negative result.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Patient Characteristics
Of the 997 eligible women, 762 consented to participate in the study (response rate, 76.4%: Leiden, n = 652; Rotterdam, n = 110). Not all women were eligible for BRCA1/2 mutation testing or chose to have a test. Furthermore, of the remaining women, not all (fully) completed three questionnaires over time (response rate 69.2%). Figure 1 represents a flow chart of those available for the analyses (n = 238).¹²

View larger version (15K): [in this window] [in a new window]   Fig 1. Flow of participants. (*) We have reported on women with a variant of uncertain clinical significance (VUCS) elsewhere.12 For two women with an uninformative result, a BRCA mutation was detected before T3. () These women developed cancer or died in between T1 and T3. () Because having no breast tissue is assumed to affect breast cancer distress, we excluded these women.

Groups of Test Applicants
Of the 238 women who received an in-person DNA mutation test disclosure session, 42 were carriers of a BRCA1/2 mutation, 43 received a true-negative result, and 153 received an uninformative test result. (In four instances a tested woman as well as her untested relative was included. We assessed whether the inclusion of these relatives could potentially affect our results by repeating all analyses without these women, or with either the tested or the untested women. Inclusion of these women did not weaken the results or conclusions. Therefore, we decided to include these women in our final analyses.) Groups did not differ regarding sociodemographic characteristics (Table 1). Furthermore, no differences with respect to breast cancer history were observed between women with an uninformative result and BRCA mutation carriers. Finally, within the group of women with an uninformative result, the familial pedigree-based risk was not associated with having a previous breast cancer diagnosis (² = 1.38; P = .24).

View larger version (10K): [in this window] [in a new window]   Fig 2. Course of breast cancer worry for groups of test applicants.

View larger version (11K): [in this window] [in a new window]   Fig 3. Course of breast cancer distress for groups of test applicants.

We assessed whether this impact of DNA mutation testing applied to all women who received an uninformative test result. Both a personal cancer history as well as a higher pedigree-based risk were independently associated with higher levels of breast cancer worry and distress (personal cancer history: worry F = 11.64, P = .001; IES F = 27.85; P < .001; risk: worry F = 10.10, P = .002; IES F = 8.10, P = .005). In addition, for personal cancer history a linear interaction with time was observed, indicating less relief after disclosure among women with a personal history of cancer (worry with time F = 5.51, P = .02; IES with time F = 7.17, P = .008).

Subgroup analyses revealed that this interaction effect of personal breast cancer history with time could only be observed among women at a relatively high risk (worry F = 4.64, P = .033; IES F = 5.75, P = .018). Among women with a relatively low risk, unaffected women did not react differently than women with a personal history of breast cancer (worry F = 1.13, P = .29; IES F = 1.78, P = .18). Three of the four subgroups of women with an uninformative result reported a significant linear decrease in both worry and distress over time (P scores for worry and distress ranged from < .001 to .033). In contrast, for women with a prior history of cancer and with a relatively high risk, no linear changes in worry or distress over time were observed (worry F = 0.55, P = .46; IES F = 0.50, P = .48).

True-Negative BRCA Test Result
Women who learned that they have not inherited the BRCA1 or BRCA2 mutation that was detected within their family previously report a linear decrease in worry and distress over time (Figs 2 and 3: worry F = 29.64, P < .001; IES F = 10.27, P = .003). They seemed to be relieved 1 month after the disclosure, and these decreased levels of worry and distress remained rather stable up to 7 months after disclosure.

BRCA1/2 Mutation Carriers
The amount of distress of women who learn that they carry a BRCA1 or BRCA2 mutation did not change over time (F = 0.05; P = .95). However, with regard to breast cancer worry, a quadratic interaction effect was observed (F = 5.08; P = .03). Immediately after disclosure a slight increase was reported. However, in the period after disclosure the overall level of worry decreased significantly (T₂ to T₃: F = 8.26; P = .006).

Some BRCA mutation carriers (19%) had undergone prophylactic mastectomy within the period under study, which could be a sufficient explanation for the decrease in worry between T₂ and T₃. This factor (ie, having had prophylactic mastectomy) neither affected the level of worry over time (F = 1.35; P = .27) nor the amount of distress over time (F = 0.75; P = .48). Nevertheless, the observed effect regarding the decreasing levels of worry between T₂ and T₃ was no longer significant if we excluded the eight women who had had prophylactic mastectomy (T₂ to T₃: F = 3.74; P = .062).

Furthermore, we could not detect different overall levels of worry or distress between affected and unaffected women (worry F = 2.36, P = .13; IES F = .81, P = .38). However, an interaction effect over time was observed for distress (F = 4.17; P = .048). Within the group of BRCA-mutation carriers with a personal cancer history, the amount of distress decreased slightly, whereas for unaffected BRCA mutation carriers, the level of distress increased slightly after disclosure. Still, unaffected BRCA mutation carriers did not differ from affected BRCA mutation carriers at any point in time regarding both worry and distress (Table 2).

Seven months after disclosure of a DNA test result, unaffected women with an uninformative result at a relatively low risk reported the same low levels of worry and distress as women with a true-negative result (T₃ worry F = 0.55, P = .46; IES F = .41, P = .53). In contrast, these unaffected women with an uninformative result at a relatively low risk were highly dissimilar from unaffected DNA mutation carriers 7 months after disclosure (T₃ worry F = 15.76, P < .001; IES F = 22.86, P < .001).

In addition, affected BRCA mutation carriers reported about the same levels of worry and distress 7 months after disclosure as affected women with an uninformative result at a relatively low risk (T₃ worry F = 0.31, P = .58; IES F = 0.26, P = .61). In addition, the reported level of distress 7 months after disclosure did not differ between affected BRCA mutation carriers and affected women with an uninformative result at a relatively high risk (T₃ F = 2.03; P = .16). Remarkably, however, the reported level of worry for the latter group of women with an uninformative result was higher than that of affected BRCA mutation carriers (T₃ F = 4.54; P = .034).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
An uninformative result was not associated with psychological harm. Quite the contrary, on a group level, these women seemed to be reassured on learning their result, but to a lesser extent than those women who received a true-negative result. However, the overall results probably provide an incomplete picture of the psychological impact of BRCA1/2 mutation testing for individuals who receive an uninformative result. This is because relatively straightforward clinical variables influenced the level of distress and worry. The women with an uninformative result with a previous cancer diagnosis reported higher levels of worry and distress than those who were unaffected. In addition to the influence of a personal cancer history, we also found an overall effect of familial breast cancer risk. Within the group of women who received an uninformative result, those with a relatively strong family history reported higher levels of worry and distress than women with a less elevated risk, before and after DNA test disclosure.

The latter is quite adequate from a clinical point of view. The likelihood that a high-risk mutation is actually present is lower after an uninformative test result, but this is especially true for women with a family history less suggestive of breast or ovarian cancer. Moreover, the results corroborate an important but subtle difference within the group of women with an uninformative result with regard to the primary aim of DNA mutation testing. Whereas women with a relatively low risk may opt for testing to rule out the relatively small possibility of having a BRCA mutation, those women with a relatively strong family history may undergo DNA testing with the motive of making a BRCA mutation manifest.

Hallowell et al¹³ described that affected women with an uninformative result reacted with anger and frustration at being unable to confirm the etiology of their own and their family cancer history, which would also provide the opportunity for informative DNA testing for their unaffected family members. Furthermore, in a study of Loader et al,¹⁴ about half of the affected women waiting for their DNA test result wished for a positive result, presumably for the same reasons as indicated by the women in the qualitative study of Hallowell et al.¹³ In concordance with these observations, affected women, and especially those at higher familial risk for cancer, not only expressed higher levels of distress and worry, they were also less relieved by their test result than unaffected women. Moreover, for these affected women with a relatively high risk, no changes in worry or distress were observed 7 months after disclosure, compared with their baseline values. At follow-up, one fourth even had distress levels above a cutoff score, indicating traumatic distress¹⁵ (score > 35; data not shown).

The categorizations of women who received an uninformative result also determined whether worry or distress levels of these women were either more or less comparable to those of women who received an informative result. Worry or distress levels in unaffected women at a relatively low risk who received an uninformative result were comparable to women who received a true-negative result. In addition, worry or distress levels in affected women with an uninformative result, independently of whether they had a relatively low or high pedigree-based risk, were rather comparable to affected BRCA mutation carriers 7 months after disclosure.

With regard to BRCA mutation carriers, 19% had had prophylactic mastectomy 7 months after disclosure. In line with the results in a recent study,¹⁶ we could not find clear-cut evidence that having a prophylactic mastectomy reduced the levels of worry or distress. This finding is alarming and requires additional research, given that the major psychological benefit from prophylactic surgery is assumed to be relief from anxiety.

Several limitations of the current study should be noted. First, subgroup analyses were conducted on relatively small sample sizes. In addition, a relatively high number of women did not fully complete the set of three questionnaires. However, women who completed all questionnaires reported the same levels of worry and distress as women who completed less than three questionnaires. Therefore, we believe that the results are still representative for the population currently seeking genetic testing.

Another limitation may be the selection of only two clinical characteristics to characterize women with an uninformative result. One reason that these clinical variables were selected was that they are relatively straightforward in the practice of genetic counseling. However, we do not want to claim that there are no other, and perhaps even more important, predictors of worry and distress among women with an uninformative result. For example, a previous study described the presence of negative life events as predictors of distress among counselees.¹⁷ Moreover, describing different levels of worry and distress, as we have done in our study, is not the same as explaining the wide variety of psychological responses to BRCA1/2 mutation testing. Future studies should have a longer follow-up period and larger numbers of respondents than in the current report. Furthermore, other measures in addition to distress, but also more well-validated measures of distress, may demonstrate the impact of genetic testing for women with an uninformative result in an even more meaningful way.

Traditionally, standard protocols for psychological counseling have especially focused on healthy women from hereditary breast and ovarian cancer families, analogous to the protocol for Huntington's disease.¹⁸ In contrast to this, our results with regard to an uninformative result suggest that women with a personal history of breast cancer, and especially among those with a high familial risk, may have levels of worry and distress after DNA mutation testing that are at least as high as those who have a positive test result. These women with an uninformative result may be a relatively neglected but vulnerable subpopulation of DNA mutation test applicants.

	Authors' Disclosures of Potential Conflicts of Interest

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
The authors indicated no potential conflicts of interest.

	Author Contributions

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 

Conception and design: Sandra van Dijk, Daniëlle R.M. Timmermans, Wilma Otten Provision of study materials or patients: Hanne Meijers-Heijboer, Christi J. van Asperen Collection and assembly of data: Sandra van Dijk, Hanne Meijers-Heijboer, Christi J. van Asperen, Wilma Otten Data analysis and interpretation: Sandra van Dijk, Daniëlle R.M. Timmermans, Wilma Otten Manuscript writing: Sandra van Dijk, Daniëlle R.M. Timmermans, Hanne Meijers-Heijboer, Aad Tibben, Christi J. van Asperen, Wilma Otten Final approval of manuscript: Daniëlle R.M. Timmermans, Hanne Meijers-Heijboer, Aad Tibben, Christi J. van Asperen, Wilma Otten

 

	ACKNOWLEDGMENTS

 
We thank Nanny van Duijn for her help on data collection, and all of the women who took part in the study.

	NOTES

 
Supported by the Dutch Cancer Society Grant No. UL 98-1740.

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

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
1. Meiser B, Butow P, Friedlander M, et al: Psychological impact of genetic testing in women from high-risk breast cancer families. Eur J Cancer 38:2025-2031, 2002[Medline]

2. Broadstock M, Michie S, Marteau T: Psychological consequences of predictive genetic testing: A systematic review. Eur J Hum Genet 8:731-738, 2000[CrossRef][Medline]

3. Nathanson KL, Wooster R, Weber BL: Breast cancer genetics: What we know and what we need. Nat Med 7:552-556, 2001[CrossRef][Medline]

4. Bish A, Sutton S, Jacobs C, et al: No news is (not necessarily) good news: Impact of preliminary results for BRCA1 mutation searches. Genet Med 4:353-358, 2002[Medline]

5. Schwartz MD, Peshkin BN, Hughes C, et al: Impact of BRCA1/BRCA2 mutation testing on psychological distress in a clinic-based sample. J Clin Oncol 20:514-520, 2002[Abstract/Free Full Text]

6. De Bock GH, Vliet Vlieland TP, Hageman GC, et al: The assessment of genetic risk of breast cancer: A set of GP guidelines. Fam Pract 16:71-77, 1999[Abstract/Free Full Text]

7. Van Asperen CJ, de Bock GH, Van der Horst F, et al: Screening for breast cancer on basis of individual risk assessment for women ineligible for the national population screening program. Ned Tijdschr Geneeskd 145:120-125, 2001[Medline]

8. Statement of the American Society of Clinical Oncology: Genetic testing for cancer susceptibility. J Clin Oncol 14:1730-1736, 1996[Abstract/Free Full Text]

9. Claus EB, Risch N, Thompson WD: Autosomal dominant inheritance of early onset breast cancer. Cancer 73:643-651, 1994[CrossRef][Medline]

10. Lerman C, Trock B, Rimer BK, et al: Psychological and behavioral implications of abnormal mammograms. Ann Intern Med 114:657-661, 1991[Abstract/Free Full Text]

11. Horowitz M, Wilner N, Alvarez W: Impact of Event Scale: A measure of subjective stress. Psychosom Med 41:209-218, 1979[Abstract/Free Full Text]

12. Van Dijk S, Van Asperen CJ, Jacobi CE, et al: Variants of uncertain clinical significance as a result of BRCA1/2 testing: The impact of an ambiguous breast cancer risk message. Genet Test 8:235-239, 2004[CrossRef][Medline]

13. Hallowell N, Foster C, Ardern-Jones A, et al: Genetic testing for women previously diagnosed with breast/ovarian cancer: Examining the impact of BRCA1 and BRCA2 mutation searching. Genet Test 6:79-87, 2002[CrossRef][Medline]

14. Loader S, Shields CG, Rowley PT: Impact of genetic testing for breast-ovarian cancer susceptibility. Genet Test 8:1-12, 2004[Medline]

15. Neal LA, Busuttil W, Rollings J, et al: Convergent validity of measures of post-traumatic stress disorder in a mixed military and civilian population. J Trauma Stress 7:447-455, 1994[CrossRef][Medline]

16. Watson M, Foster C, Eeles R, et al: Psychosocial impact of breast/ovarian (BRCA1/2) cancer-predictive genetic testing in a UK multi-centre clinical cohort. Br J Cancer 91:1787-1794, 2004[CrossRef][Medline]

17. Kim Y, Duhamel KN, Valdimarsdottir HB, et al: Psychological distress among healthy women with family histories of breast cancer: Effects of recent life events. Psychooncology 14:555-563, 2005[CrossRef][Medline]

18. Bleiker EM, Grosfeld FJ, Hahn DE, et al: Psychosocial care in family cancer clinics in The Netherlands: A brief report. Patient Educ Couns 43:205-209, 2001[CrossRef][Medline]

Submitted August 5, 2005; accepted May 19, 2006.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by van Dijk, S. Articles by Otten, W. Search for Related Content PubMed PubMed Citation Articles by van Dijk, S. Articles by Otten, W. Social Bookmarking                            What's this?