Familial Cancer

, Volume 9, Issue 3, pp 297–304

Uptake of predictive testing among relatives of BRCA1 and BRCA2 families: a multicenter study in northeastern Spain

Authors

    • Cancer Genetic Counselling Unit, Medical Oncology DepartmentHospital de la Santa Creu i Sant Pau
    • Clinical Genetics Unit, Medical Oncology DepartmentHospital Universitari Sant Joan
  • Teresa Ramón y Cajal
    • Cancer Genetic Counselling Unit, Medical Oncology DepartmentHospital de la Santa Creu i Sant Pau
  • Asunción Torres
    • Clinical Genetics Unit, Medical Oncology DepartmentHospital Universitari Sant Joan
  • Esther Darder
    • Cancer Genetic Counselling Program, Catalan Institute of OncologyHospital Josep Trueta
    • Girona Biomedical Research Institute, IdIBGi
  • Neus Gadea
    • Medical Oncology DepartmentHospital Universitari Vall d’Hebron
  • Angela Velasco
    • Cancer Genetic Counselling Program, Catalan Institute of OncologyHospital Josep Trueta
    • Girona Biomedical Research Institute, IdIBGi
  • Daniel Fortuny
    • Medical Oncology DepartmentHospital Universitari Vall d’Hebron
  • Consol López
    • Cancer Genetic Counselling Unit, Medical Oncology DepartmentHospital de la Santa Creu i Sant Pau
  • David Fisas
    • Cancer Genetic Counselling Unit, Medical Oncology DepartmentHospital de la Santa Creu i Sant Pau
  • Joan Brunet
    • Cancer Genetic Counselling Program, Catalan Institute of OncologyHospital Josep Trueta
    • Girona Biomedical Research Institute, IdIBGi
  • M. Carmen Alonso
    • Cancer Genetic Counselling Unit, Medical Oncology DepartmentHospital de la Santa Creu i Sant Pau
  • Judith Balmaña
    • Medical Oncology DepartmentHospital Universitari Vall d’Hebron
Article

DOI: 10.1007/s10689-009-9313-1

Cite this article as:
Sanz, J., Ramón y Cajal, T., Torres, A. et al. Familial Cancer (2010) 9: 297. doi:10.1007/s10689-009-9313-1

Abstract

Identifying a BRCA mutation among families with hereditary breast and ovarian cancer enables distinguishing those who may benefit from a specific medical management. This study aimed to evaluate the uptake of predictive testing among close relatives of a proband in Spanish families with a BRCA1 or BRCA2 mutation, and to determine the associated demographic and clinical predictors. A retrospective cohort of families undergoing clinical genetic testing at four university hospitals in northeastern Spain was considered. From 108 unrelated BRCA1/2 families, 765 close relatives of probands were analyzed. Sixty percent of the first-degree and 28% of the second-degree relatives underwent predictive testing within a median time of 2 and 6 months, respectively, since the mutation disclosure to the proband. Relatives undergoing genetic testing were more likely to be female, first-degree, and belong to a family with a proband who had a high educational level. Relatives were also more likely to have offspring, a previous cancer diagnosis, and to be aged between 30 and 64 years. Among second-degree relatives, having a first-degree relative with cancer was highly correlated with uptake. In conclusion, uptake of BRCA1/2 predictive testing among close relatives was notably high and within a short period of time after disclosure of the mutation to the proband. Being female, a high educational level of the proband, and having a close relative with cancer were associated with uptake among relatives. Further studies are warranted to determine whether information is disseminated properly by probands and to learn about the reasons for those not undergoing testing.

Keywords

BRCA1BRCA2Predictive testingFactorsRelativesUptake

Introduction

For over a decade, medical management of families with hereditary breast and ovarian cancer (HBOC) syndrome have included genetic counselling and testing for BRCA1 and BRCA2 mutations. Recent data on the Spanish population estimate that the cumulative risk for breast cancer is 52% in BRCA1 and 47% in BRCA2 females, and for ovarian cancer it is 22 and 18% respectively [1]. Once a deleterious mutation is identified in the proband of an HBOC family, predictive testing in his/her relatives can differentiate carriers from non-carriers, thereby determining those who will benefit from more intensive surveillance or prophylactic surgery and those for whom such intervention is unnecessary.

Before BRCA1/2 testing became clinically available, several studies investigating the interest among HBOC families for mutation screening showed a positive attitude [2, 3]. Further research once testing was available found a high rate of communication of BRCA1/2 mutations to close relatives [49]. Therefore, uptake of predictive testing was also expected to be high among BRCA1/2 families. Nevertheless, studies in many countries have demonstrated lower uptake rates than anticipated [4, 6, 8, 1012].

The goals of our study were to assess the uptake of predictive testing among first- and second-degree relatives of the proband in Spanish families with a BRCA1 or a BRCA2 deleterious mutation, and to determine the demographic and clinical predictors associated with uptake.

Study population and methods

Participants

In this retrospective, multicentre cohort study we revised data concerning those families who underwent genetic testing for BRCA1 or BRCA2 genes between October 1995 and December 2007 and were found to carry a deleterious mutation. Inclusion criteria were that the center had updated follow-up information of the proband and his/her first- and second-degree relatives, and that the mutation had been disclosed to the proband at least 6 months before data collection. Individuals considered for the analysis had to be alive, over 18 years of age, and first- or second-degree relatives of the bloodline at risk. Offspring of non-carriers were excluded from the analysis. From a total of 108 unrelated families with a BRCA1 or a BRCA2 deleterious mutation that fulfilled the inclusion criteria, 765 relatives (399 first-degree and 366 second-degree) were recorded. Clinical and demographic data for each individual who underwent genetic testing were collected through questionnaires in personal interviews. Information about those who did not undergo genetic testing was collected from their relatives.

Twenty-five families with a deleterious BRCA1 or BRCA2 mutation were excluded due to loss of follow-up with the family members. Two relatives with a severe mental disorder were also excluded from the analysis.

Probands and their families were attended at the family cancer units from four university hospitals in the Catalan region, Spain: Hospital de Sant Pau and Hospital Vall d’Hebron in Barcelona, Hospital Sant Joan in Reus, and Hospital Josep Trueta in Girona. The contents and dynamics of genetic counselling and testing were similar in all participating centres and regional guidelines were followed [13]. Pre- and post-test assessment visits were given individually or collectively, by choice of the proband and according to family dynamics. Probands were strongly encouraged to inform their relatives about the familial cancer risk and to recommend them to seek specialized genetic assessment, if a mutation was identified. After disclosure of genetic results, all probands were provided with written information that included the genetic result and the personal and familial implications to facilitate the communication process within the family. Therefore, dissemination of information concerning the family’s cancer risk depended initially on probands. If the family made no further contact with a family cancer unit, no active measures were taken to promote family communication. Probands consented to participate in institutional review board (IRB) approved studies.

Methods

For probands, we collected data on gender, educational level, parenthood, previous diagnosis of cancer, type of cancer and age at diagnosis, BRCA1 or BRCA2 mutation, age at disclosure of mutation, family history of cancer (number of breast/ovarian cancer cases, presence of any relative’s death due to cancer before 50 years of age). For relatives, we obtained data concerning gender, parenthood, previous diagnosis of cancer and type of cancer. For those relatives who undertook genetic testing we recorded age at uptake, and time from disclosure of the mutation to the proband until testing. The median time was used rather than the mean as, at the centers where genetic testing has been available for more than 10 years, means are strongly influenced by the length of follow-up. For relatives who did not undergo the test, we recorded age at time of data collection. Age was analyzed according to 3 groups, ranged from 18–29, 30–64 and more than 64 years. For first-degree relatives, we recorded the relationship with the proband (offspring, sibling, parent), and for second-degree relatives, whether they had any first-degree relative with cancer.

To evaluate the differences between tested and non-tested groups, we used the T-test for continuous variables and Fisher’s Exact Test for categorical variables. A stepwise regression analysis was conducted to determine the predictors of uptake of genetic testing among relatives. All variables were included in the multivariate analysis. A P-value of 0.05 was considered statistically significant. All statistical analyses were performed with SPSS V14 (Chicago, IL).

Results

Characteristics of the study population

Tables 1 and 2 show the sociodemographic and clinical data of probands and first- and second-degree relatives, respectively. Almost all probands were women (97%), had offspring (82%), were affected with breast and/or ovarian cancer (93%), and had a median age of 40 years at first diagnosis. Seventeen percent had completed primary school, 59% secondary school and 24% had a university degree. Among first-degree relatives, 55% were female, 64% had offspring and 11% had breast and/or ovarian cancer. Among second-degree relatives, 53% were female, 51% had offspring, 6% had breast and/or ovarian cancer, and 59% had at least one-first-degree relative with cancer.
Table 1

Demographics of probands

 

N (%)

Individuals

108

Gender

 Female

105 (97)

 Male

3 (3)

Previous children

 Yes

89 (82)

 No

19 (18)

Educational level

 Primary school

17 (17)

 Secondary school

59 (59)

 University degree

24 (24)

Previous cancer

 Breast

79 (73)

 Ovary

12 (11)

 Breast + ovary

10 (9)

 Prostate

1 (1)

 Unaffected

6 (6)

Median age at 1st diagnosis

40 (21–82)

Number of HBOC-related cancer cases in the family

 1–2 cases of BC/OC

15 (14)

 >2 cases of BC/OC

92 (86)

Cancer antecedents in relatives

 Cancer death of a relative <50 years

67 (63)

 No cancer deaths among relatives <50 years

40 (37)

Mutated gene

 BRCA1

52 (48)

 BRCA2

56 (52)

Median age at disclosure of genetic result

50 (28–83)

HBOC hereditary breast and ovarian cancer, BC/OC breast cancer/ovarian cancer

Table 2

Demographics of first- and second-degree relatives

 

First-degree relatives N (%)

Second-degree relatives N (%)

Total

Individuals

399

366

765

Gender

 Female

219 (55)

194 (53)

413 (54)

 Male

180 (45)

172 (47)

352 (46)

Previous children

 Yes

246 (64)

155 (51)

401 (52)

 No

139 (36)

149 (49)

288 (38)

Previous cancer

 Breast

41 (10)

21 (6)

62

 Ovary

2 (1)

0

2

 Breast + ovary

0

0

0

 Prostate or others

11 (3)

12 (3)

23

 Unaffected

344 (86)

333 (91)

677

Specific relationship with proband (for FDR)

 Offspring

145 (36)

 Siblings

213 (54)

 Parent

41 (10)

Having a FDR with cancer (for SDR)

 Yes

216 (59)

 No

150 (41)

Mean age at time of genetic testing (for testers)

45 (±16.7)

45 (±18.3)

45 (±18.8)

FDR first-degree relatives, SDR second-degree relatives

Uptake of genetic testing

Uptake of genetic testing was calculated for all close relatives together and separately, for first-degree and second-degree relatives. Table 3 shows the univariate analyses for the three groups.
Table 3

Uptake of genetic testing among relatives (univariate analyses)

 

All relatives together N (%)

P value

First-degree relatives N (%)

P value

Second-degree relatives N (%)

P value

Relationship with proband

 First-degree relative

238 (60)

<0.001

    

 Second-degree relative

102 (28)

     

Gender

 Female

239 (58)

<0.001

167 (76)

<0.001

72 (37)

<0.001

 Male

101 (29)

 

71 (39)

 

30 (17)

 

Previous children

 Yes

216 (54)

<0.001

158 (64)

0.13

58 (37)

0.001

 No

113 (39)

 

78 (56)

 

35 (23)

 

Proband’s educational level

 Primary school

42 (30)

 

28 (41)

 

14 (20)

 

 Secondary school

186 (44)

<0.001

128 (60)

<0.001

58 (27)

0.021

 University degree

88 (68)

 

70 (79)

 

18 (44)

 

Relative’s previous cancer diagnosis

 Yes

58 (66)

<0.001

45 (82)

<0.001

13 (39)

0.15

 No

282 (42)

 

193 (56)

 

89 (27)

 

Age at genetic testing (only for testers)

 18–29

68 (41)

 

45 (55)

 

23 (27)

 

 30–64

215 (54)

0.004

157 (63)

0.313

58 (38)

0.154

 65+

55 (41)

 

35 (57)

 

20 (27)

 

Specific relationship with proband (for FDR)

 Offspring

 

86 (59)

 

 

 Sibling

 

120 (56)

0.35

 

Having a FDR with cancer (for SDR)

 Yes

 

 

75 (35)

 

 No

 

 

27 (18)

<0.001

Number of HBOC-related cancer cases

 1–2 cases of BC/OC

31 (44)

 

27 (54)

 

4 (20)

 

 >2 cases of BC/OC

308 (44)

0.5

210 (60)

0.241

98 (28)

0.3

Cancer antecedents in relatives

 Cancer death of a relative <50 years

205 (41)

 

132 (54)

 

73 (28)

 

 No cancer deaths among relatives <50 years

134 (52)

0.003

105 (69)

0.002

29 (27)

0.47

Mutated gene

 BRCA1

164 (43)

 

107 (57)

 

57 (30)

 

 BRCA2

176 (45)

0.4

131 (62)

0.26

45 (25)

0.35

FDR first-degree relative, SDR second-degree relative, HBOC hereditary breast and ovarian cancer, BC/OC breast cancer/ovarian cancer

Uptake among all close relatives

From a total of 765 eligible relatives, 44% (95% confidence interval (CI), 41–48) underwent genetic testing. The median time for testing was 3.3 months (range 0–113) after mutation disclosure to the proband. A greater number of first-degree relatives (60%) uptook genetic testing than second-degree relatives (28%, P < 0.001). Other variables associated with a significant higher uptake among relatives were female gender (58 vs. 29%, P < 0.001), to have offspring (54 vs. 39%, P < 0.001), a previous diagnosis of cancer (66 vs. 42%, P < 0.001) and a higher educational level of the proband (68 vs. 44 vs. 30%, P < 0.001). Finally, relatives in the 30–64 year age group underwent testing more often than the other two age groups (54 vs. 41 vs. 41%, P = 0.004).

The odds of higher uptake were being a first-degree relative (OR: 3.5), female gender (OR: 3.9), relative of a proband with a university degree (OR: 5.9), age 30–64 years (OR: 2.4), having offspring (OR: 1.7) and a previous diagnosis of cancer (OR: 2.8). Complete information is shown in Table 4.
Table 4

Predictors of uptake of genetic testing among all relatives

 

OR (95% CI)

P value

Degree of relationship with proband

 First-degree relative

3.5 (2.3–5.1)

<0.001

 Second-degree relative

1

 

Gender

 Female

3.9 (2.7–5.9)

<0.001

 Male

1

 

Proband’s educational level

 Primary school

1

 

 Secondary school

2.1 (1.2–3.4)

0.005

 University degree

5.9 (3.1–11.2)

<0.001

Previous children

 Yes

1.7 (1.0–2.7)

0.039

 No

1

 

Previous cancer

 Yes

2.8 (1.4–5.5)

0.003

 No

1

 

Age group (years)

 18–29

1.6 (0.8–3.3)

0.165

 30–64

2.4 (1.4–4.0)

0.001

 +65

1

 

Variables included in the model: relative’s gender, degree and type (offspring, sibling, parent) of relationship with the proband, previous children, previous diagnosis of cancer, proband’s educational level, family history of cancer (number of breast/ovarian cancer cases, death of any relative due to cancer before 50 years of age), BRCA1 or BRCA2 mutation, age at uptake (for testers), age at time of data collection (for non testers), having a first-degree relative with cancer (for second-degree relatives) and participating center

Uptake among first-degree relatives

Sixty percent of relatives uptook genetic testing within a median time of 2 months (range 0–113) after mutation disclosure to the proband. Relatives who underwent testing were significantly more likely to be female (76 vs. 39%, P < 0.001), to have a proband with a higher educational level (79 vs. 60 vs. 41%, P < 0.001) and to have a previous diagnosis of cancer (82 vs. 56%, P < 0.001).

Independent predictors of testing uptake for first-degree relatives (Table 5) were female gender (OR: 5.0), having offspring (OR: 1.7), a previous diagnosis of cancer (OR: 3.0) and being a relative of a proband with a university degree (OR: 7.5).
Table 5

Predictors of uptake of genetic testing among first-degree relatives

 

OR (95% CI)

P value

Gender

 Female

5.0 (3.0–8.4)

<0.001

 Male

1

 

Proband’s educational level

 Primary school

1

 

 Secondary school

2.0 (1.0–3.9)

0.053

 University degree

7.5 (3.1–18.2)

<0.001

Previous cancer

 Yes

3.0 (1.1–7.9)

0.025

 No

1

 

Previous children

 Yes

1.7 (1.0–3.0)

0.039

 No

1

 

Variables included in the model: relative’s gender, degree and type (offspring, sibling, parent) of relationship with the proband, previous children, previous diagnosis of cancer, proband’s educational level, family history of cancer (number of breast/ovarian cancer cases, death of any relative due to cancer before 50 years of age), BRCA1 or BRCA2 mutation, age at uptake (for testers), age at time of data collection (for non testers), having a first-degree relative with cancer (for second-degree relatives) and participating center

Uptake among second-degree relatives

Twenty-eight percent of relatives underwent genetic testing within a median time of 6 months (range 0–73) after mutation disclosure to the proband. Factors associated with a significant higher uptake were female gender (37 vs. 17%, P < 0.001), having offspring (37 vs. 23%, P = 0.001), a higher educational level of the proband (44 vs. 27 vs. 20%, P = 0.021), and having at least one-first-degree relative diagnosed of cancer (35 vs. 18%, P < 0.001).

Independent predictors of genetic testing for second-degree relatives (Table 6) were female gender (OR: 3.3), being a relative of a proband with a university degree (OR: 5.7), age 30–64 years old (OR: 3.3), and having a first-degree relative with cancer (OR: 3.3).
Table 6

Predictors of uptake of genetic testing among second-degree relatives

 

OR (95% CI)

P value

Gender

 Female

3.3 (1.8–6.1)

<0.001

 Male

1

 

Having a first-degree relative with cancer

 Yes

3.3 (1.6–6.7)

0.001

 No

1

 

Proband’s educational level

 Primary school

1

 

 Secondary school

2.7 (1.2–5.9)

0.016

 University degree

5.7 (2.1–15.6)

0.001

Age group

 18–29 years

1.9 (0.8–4.6)

0.16

 30–64 years

3.3 (1.5–6.8)

0.002

 +65 years

1

 

Variables included in the model: relative’s gender, degree and type (offspring, sibling, parent) of relationship with the proband, previous children, previous diagnosis of cancer, proband’s educational level, family history of cancer (number of breast/ovarian cancer cases, death of any relative due to cancer before 50 years of age), BRCA1 or BRCA2 mutation, age at uptake (for testers), age at time of data collection (for non testers), having a first-degree relative with cancer (for second-degree relatives) and participating center

Discussion

This is the first study of genetic testing uptake among BRCA1/2 families in Spain. Our results showed that almost half the eligible first- and second-degree relatives of the proband underwent predictive testing within 6 months of the mutation being disclosed to the family. The most powerful independent predictive factors for uptake were female gender and being a first-degree relative of the proband. Other independent factors were being a relative of a proband with a high educational level, having a diagnosis of cancer, having offspring, and age 30–64 years.

The observed 44% overall rate of uptake among close relatives is higher than that reported in Netherlands (36%) [8], France (27%) [11] and Israel (29%) [14]. Among first-degree relatives, 60% rate is also higher than that reported in Norway (43%) [15], Scotland (39%) [10], and France (26%) [4]. Two studies from the US [6, 12] report a similar uptake to ours, although direct comparisons cannot be made due to differences in study design. A 28% uptake of genetic testing among second-degree relatives is also higher than rates reported by others, as 6% in France [4]. This notably high uptake is meaningful in terms of reflecting the current clinical impact of identifying a BRCA1 or BRCA2 mutation in families in our setting. Is noteworthy that the analysis was performed without any previous selection of the study population or any specific intervention for the purpose of the current research, thereby precluding a selection bias and reflecting common clinical practice among families with a BRCA1/2 mutation in our region. The high uptake among our families may be related to the close relationship within family members in our culture. In addition, our study encompasses a longer follow-up period than other studies, facilitating the access to testing by relatives.

As anticipated, the strongest predictive factors of uptake were female gender and being a first-degree relative of the proband, which are consistent with the majority of studies in other countries [6, 11, 1518]. Unexpectedly, we found that 41% of the proband’s offspring did not uptake testing. Once the gender of this group was analyzed, we observed that 74% were male and consequently they might have been less motivated to undergo genetic testing, as results may not highly influence their own medical management. However, these individuals may transmit the mutated allele to their offspring, which may lately have an impact on their clinical management. Therefore, we must assure that adequate information regarding the autosomal dominant pattern of BRCA1/2 inheritance is being understood in the family.

First-degree relatives who had been diagnosed of cancer had a higher uptake than those who were non-affected, which is in agreement with results from other authors [14, 16, 19] and may merely reflect the fact that they wanted to confirm their mutation status. We believe that efforts should be made to enhance genetic testing among non-cancer relatives as they may greatly benefit from the predictive test in terms of surveillance and preventive practices. At a time that effective screening recommendations for breast cancer exist [20] and early prophylactic surgery of the ovaries is associated with a decreased risk in cancer and mortality, unaffected relatives are likely to benefit from knowing their mutation status.

The fact that only one every five-second-degree relatives without a first-degree relative with cancer underwent genetic testing suggests that strategies to involve second-degree relatives in genetic testing, independently of the cancer status of their first-degree relatives, are also warranted in light of the incomplete cancer penetrance of BRCA1 and BRCA2 mutations. It is likely that cancer risk perception among these relatives is low, and we must assure that knowledge of the existing genetic susceptibility in the family is being transmitted properly.

Many authors have reported that one of the main motivations for individuals to undergo genetic testing is gaining awareness of the cancer risk for their offspring [12, 16, 2123]. Accordingly, our data showed that first- but not second-degree relatives with offspring were more likely to take the test than those without children.

The finding that uptake was higher among relatives of a proband with a university degree might be supported by the fact that the proband’s degree of knowledge about the risk of transmission of BRCA1/2 mutations facilitates the communication process. In consequence, this would be significantly associated with the testing uptake among first-degree relatives [4, 12], while the proband’s difficulties in informing family members appears to be related to their poor understanding of the information [4, 8]. Individuals with a higher educational level likely have a better understanding of the genetic information and are able to communicate this more confidently to their relatives [2426]. Although we did not record specific data about the proband’s knowledge or understanding of the hereditary cancer risk, our results support this explanation. Consequently, efforts should be invested in our setting to develop easy-to-understand educational tools to better transmit information on BRCA cancer susceptibility among the family members. In addition, implementing more family-driven informative sessions could help to communicate genetic information within the relatives.

Our study must be considered in light of several limitations. The cross-sectional design of the analysis did not allow us to determine whether individuals who did not performed genetic testing had not been informed or had chosen not to come. To answer this question we would need a prospective analysis assessing the communication process within the family and the reasons to pursue or reject genetic testing among relatives. The exclusion of families whose proband had no contact at all with his/her first- or second-degree relatives, and those families that were lost to follow up by the centers may have biased positively the results. However, these families only accounted for a small percentage of the overall study population and we do not think that they would influence much the final results.

In conclusion, uptake of predictive testing among close relatives in our population was notably high and was undertaken within a short period of time after disclosure of the mutation to the proband. To learn about those not undergoing genetic testing, further studies are warranted to determine whether information is disseminated properly by probands and the reasons to reject genetic testing by informed relatives. Development of simple educational tools and implementing family informative sessions could help to communicate genetic information within the families.

Acknowledgments

The authors wish to thank Drs. O. Díez, M. Rodríguez, M. Baiget, A. Lasa, S. Gutiérrez, C. Lázaro, M. Domènech and M. Cornet for mutation analysis; Alberto Ameijide for help with the statistical analysis; Carolyn Newey for technical assistance; and Fundació Roses Contra el Càncer.

Copyright information

© Springer Science+Business Media B.V. 2009