Breast Cancer Research and Treatment

, Volume 144, Issue 2, pp 379–389

Determinants of genetic counseling uptake and its impact on breast cancer outcome: a population-based study

Authors

  • Aurélie Ayme
    • Department of Genetic and Laboratory MedicineUniversity Hospitals of Geneva
    • Oncogenetics and Cancer Prevention Unit, Division of OncologyUniversity Hospitals of Geneva
  • Valeria Viassolo
    • Oncogenetics and Cancer Prevention Unit, Division of OncologyUniversity Hospitals of Geneva
  • Elisabetta Rapiti
    • Geneva Cancer Registry, Institute for Social and Preventive MedicineGeneva University
  • Gérald Fioretta
    • Geneva Cancer Registry, Institute for Social and Preventive MedicineGeneva University
  • Hyma Schubert
    • Geneva Cancer Registry, Institute for Social and Preventive MedicineGeneva University
  • Christine Bouchardy
    • Geneva Cancer Registry, Institute for Social and Preventive MedicineGeneva University
  • Pierre O. Chappuis
    • Oncogenetics and Cancer Prevention Unit, Division of OncologyUniversity Hospitals of Geneva
    • Division of Genetic MedicineUniversity Hospitals of Geneva
    • Geneva Cancer Registry, Institute for Social and Preventive MedicineGeneva University
    • INSERM, U946
Epidemiology

DOI: 10.1007/s10549-014-2864-3

Cite this article as:
Ayme, A., Viassolo, V., Rapiti, E. et al. Breast Cancer Res Treat (2014) 144: 379. doi:10.1007/s10549-014-2864-3

Abstract

Genetic counseling and BRCA1/BRCA2 genes testing are routinely offered in a clinical setting. However, no data are available on the proportion of breast cancer patients with a positive family history undergoing genetic counseling. By linking databases of the Oncogenetics and Cancer Prevention Unit at the Geneva University Hospitals and the population-based Geneva Cancer Registry, we evaluated the uptake of genetic counseling among 1709 breast cancer patients with familial risk of breast cancer and the determinants of such a consultation process. We also studied the impact of genetic counseling on contralateral breast cancer occurrence and survival. Overall, 191 (11.2 %) breast cancer patients had genetic counseling; this proportion was 25.1 % within the high familial risk group. Recent period of diagnosis, early-onset breast cancer, female offspring, high familial risk, tumor size, and chemotherapy treatment were statistically significantly associated with genetic counseling uptake in multivariate analysis. More than 2 % of patients had developed contralateral metachronous breast cancer. An increased risk of contralateral breast cancer of borderline significance was found for patients who had genetic counseling versus those who had not (Cox model adjusted hazard ratio 2.2, 95 % confidence intervals 1.0–5.2, P = 0.063). Stratification by BRCA1/BRCA2 mutation status showed that the occurrence of contralateral breast cancer was 8-fold higher among mutation carriers compared with non-carriers. Age-adjusted overall survival and breast cancer-specific survival were not significantly different between patients who underwent genetic counseling and those who did not. In conclusion, we observed a significant increase in the use of genetic counseling over time and found that breast cancer patients with high familial risk had more often genetic counseling than those with moderate familial risk. A more thorough evaluation of sociodemographic and clinical predictors to attend the cancer genetic unit may help improving the use of genetic counseling services for at-risk individuals at a population level.

Keywords

Breast cancerFamilialGenetic counselingSecond primaryEpidemiologyCancer

Introduction

Positive family history is one of the strongest predictors of a woman’s lifetime risk of developing breast cancer [1, 2]. BRCA1 and BRCA2 are the two major susceptibility genes involved in hereditary predisposition to breast cancer [3, 4]. Genetic counseling and testing are now routinely offered to individuals with increased probability of carrying BRCA1/BRCA2 mutations [5]. Therefore, breast cancer patients and their relatives dealing with the possibility of hereditary breast cancer risk have to face numerous decisions, including the choice of undergoing genetic counseling, testing for BRCA1 and BRCA2 mutations, screening, and prevention strategies such as chemoprevention and prophylactic surgery. Family history of breast cancer is also a well-established risk factor for contralateral breast cancer [610]. Carrying BRCA1/BRCA2 germ-line mutations has been associated with a high risk of contralateral breast cancer with a 10-year risk ranging from 18 to 33 % for BRCA1 mutation carriers and from 13 to 19 % for BRCA2 mutation carriers [1114]. To date, no study investigated the impact of genetic counseling among breast cancer patients with familial risk on breast cancer outcome.

Robust data exist on the uptake rate and determinants of BRCA1/BRCA2 testing in families diagnosed with deleterious BRCA1/BRCA2 mutations [1520]. Rates of testing among first- and second-degree relatives of index cases ranged from 27 to 44 %, and testing decision was more frequent among females, first-degree relatives, individuals with personal history of cancer, and individuals with offspring [1517, 19, 20]. On the contrary, little is known about the attendance rates and determinants of genetic counseling and testing in different populations, and the context according to personal and family history of breast cancer. In a systematic review evaluating the real and the hypothetical (defined as “being interested in testing”) uptake of breast cancer genetic testing in individuals with personal or family breast cancer history, a few characteristics such as Ashkenazi Jewish heritage, older age, and married status were associated with genetic testing uptake [21]. However, studies differed in recruitment setting, assessment of family history of breast cancer and uptake definitions. In particular, none of the previous studies has evaluated on a population-based level, the effects of other potentially important characteristics such as social class, private or public sector of care, and tumor characteristics in genetic counseling attendance and the effect of such consultation on breast cancer outcome [2124].

The main goals of the present study were to assess the proportion of breast cancer patients with a positive family history who decided to undergo genetic counseling in a population-based setting and to investigate the determinants of such a consultation process. We also investigated the impact of counseling uptake on breast cancer outcomes including contralateral tumor occurrence and survival. These issues were addressed by linking databases of the Oncogenetics and Cancer Prevention Unit at the Geneva University Hospitals and the population-based Geneva Cancer Registry in Switzerland.

Patients and methods

Oncogenetics and Cancer Prevention Unit and Geneva Cancer Registry databases

In 1994, a consultation unit providing cancer risk assessment, surveillance, and prevention recommendations was initiated by the Division of Oncology at the Geneva University Hospitals, Switzerland. It was formally set up as the Oncogenetics and Cancer Prevention Unit in 1996, and remained the only center providing genetic counseling for familial aggregation or hereditary breast cancer predisposition syndromes in the Geneva area. The counseling activity encompasses all levels of care (average, as well as moderate, and high cancer risk situations) and all types of familial cancer aggregation or hereditary cancer susceptibility syndromes. The cancer risk evaluation process involves the collection of personal and family history (at least for first-, second-, and third-degree relatives) and, whenever possible, confirmation of all cancer diagnoses through medical records. In case of personal or familial medical history suggestive of a hereditary cancer susceptibility syndrome, the possibility to undergo genetic testing is extensively discussed as part of the genetic counseling. Since 1994, more than 1,800 families have been seen at the Oncogenetics Unit and about 60 % of all probands consulted for breast cancer risk evaluation.

For the purpose of this study, 1,550 pedigrees of individuals (probands) who consecutively consulted the Oncogenetics and Cancer Prevention Unit between 1994 and 15 June 2012 were reviewed and all breast cancer diagnoses were registered. After having confirmed with the Geneva Cancer Registry that the residence of the patient and the site of breast cancer-related treatments were located in the Geneva Canton, a total of 469 individuals were retained for the study. Variables of interest included in this database are gender, date of birth, vital status, parity, date of breast cancer diagnosis, relationship with the family member attending genetic consultation, date of first consultation, carrying out BRCA1/BRCA2 testing, date of blood sampling, and genetic testing result. For the carriers of pathogenic BRCA1 or BRCA2 mutations, information concerning preventive mastectomy and/or oophorectomy (realized or not, and when relevant, age at surgery) was also considered.

The population-based Geneva Cancer Registry has been described in detail elsewhere [25]. Briefly, the database contains information on all patients with cancer diagnosed in the resident population of the canton of Geneva since 1970. The Geneva Cancer Registry extracts information from various sources and is considered accurate [26]. Recorded data include sociodemographic information, history of breast and ovarian cancers in first- or second-degree relatives, tumor characteristics [27], hormone receptor status, stage of disease at diagnosis, treatment during the first 6 months after diagnosis, occurrence of other primary cancers, and survival status. The Cancer Registry regularly assesses survival. The index date refers to the date of diagnosis confirmation or the date of hospitalization when it preceded the diagnosis and was related to the disease. In addition to passive follow-up (routine examination of death certificates and hospital records), active follow-up is performed yearly using the files of the Cantonal Population Office in charge of the registration of the resident population. Cause of death is systematically recorded and validated by consulting medical files or, when necessary, by sending a specific questionnaire to the patient’s physician. Cause of death is coded according to the International Statistical Classification of Diseases and Related Health problems [28].

Variables of interest included in this database are patient characteristics (date of diagnosis, age at diagnosis, marital status, offspring, level of familial risk, social class, country of birth, sector of care), tumor characteristics (method of detection, stage, size, axillary node involvement, histological subtype, differentiation, hormone receptor status, and HER2 status), types of treatment (surgery, radiotherapy, chemotherapy, and hormonotherapy), and outcome.

Social class was regrouped in three levels: low (manual employees, skilled and unskilled workers), middle (non-manual employees and administrative staff), and high (professionals, executives, administrators, entrepreneurs) based on the patient’s last occupation or, if unemployed, that of the spouse.

Breast cancer histology was classified as ductal carcinoma, lobular carcinoma, other and unknown (no microscopic confirmation). Staging was based on the pathologic tumor-node-metastasis (TNM) classification or, when absent, the clinical TNM classification [28]. Hormone receptor status was classified as positive (≥1 % cells expressing receptors) or negative (<1 % cells expressing receptors). Treatment was classified as surgery (breast-conservative surgery, mastectomy), radiotherapy (yes, no), chemotherapy (yes, no), and hormonotherapy (yes, no).

History of breast and ovarian cancers in first- or second-degree relatives is routinely recorded since 1990 [29]. Familial risk was categorized as high (at least 1 first-degree relative with breast/ovarian cancer diagnosed before the age of 50 years, or at least 2 first-degree relatives with breast/ovarian cancer at any age, or at least three cases of breast/ovarian cancer among first- or second-degree relatives), low (no affected first- or second-degree relatives with breast/ovarian cancer), or moderate (all other known family histories) according to a previous study of our group [29].

For the purpose of the current study, we identified all women with a first invasive breast cancer and a moderate or high familial risk recorded between 1990 and 2010 at the Geneva Cancer Registry (n = 1784). Linkage of this database with the one of Oncogenetics and Cancer Prevention Unit (n = 469) led us to exclude 203 individuals from the latter (9 male patients, 11 breast cancer patients diagnosed before 1990, 33 patients with in situ carcinoma and 150 patients classified as low familial risk). Finally, 75 women who had not themselves undergone genetic counseling were excluded from both databases.

Statistical analysis

We compared patient, tumor, and treatment characteristics among women who had genetic counseling versus women who had not using the Chi square test of heterogeneity. To assess which variables were independently associated from genetic consultation, we performed an unconditional logistic regression including in the model variables significantly associated in univariate analyses. Breast cancer patients who had genetic counseling were considered as cases and those who had not as controls.

Patients were followed for occurrence of metachronous contralateral breast cancer or death until December 31, 2010. Metachronous contralateral breast cancers were defined as all invasive breast cancers occurring after 6 months following the first breast cancer. Cumulative risks for developing metachronous contralateral breast cancer after the first breast cancer were calculated by using the Kaplan–Meier product-limit method and were compared between patients with and without genetic counseling by using the log-rank test. Cox proportional hazards regression was used to calculate hazard ratios and their 95 % confidence intervals (95 % CI). Observations were censored at the time of contralateral breast cancer, or death, whichever occurred first. Patients with synchronous contralateral cancer at the time of first breast cancer (n = 34) or having had a prophylactic bilateral mastectomy (n = 43) were excluded from this analysis.

We also evaluated the breast cancer-specific survival defined as the interval between the date of diagnosis and the date of death from breast cancer, and the overall survival defined as the interval between the date of diagnosis and the date of death from any cause. We used Cox regression models to evaluate the impact on survival of having genetic counseling.

All tests were two-sided. Statistical significance was established at P < 0.05. Analyses were performed using SPSS software (version 15.0.1, SPSS Inc. Chicago, IL, USA).

Results

The studied cohort included 1,709 breast cancer patients with moderate or high familial risk, 191 (11.2 %) had genetic counseling, and 1,518 (88.8 %) never had one. The mean duration of follow-up was not significantly different between patients who underwent genetic counseling and those who did not (79 months and 86 months, respectively; P = 0.095). Among the 191 patients who had genetic counseling, 119 (62.3 %) underwent BRCA1/BRCA2 genetic testing. Out of this group, pathogenic mutations were identified in 23 (19.3 %) patients, whereas the remaining 96 patients had non-informative testing results. Fifteen patients carried a BRCA1 mutation and eight patients a BRCA2 mutation.

Determinants of genetic counseling uptake

Characteristics of the entire patients’ cohort are presented in Table 1. Genetic counseling uptake increased from 8.4 % (20/238) for patients diagnosed in 1990–1994 to 16.3 % (91/560) for those diagnosed in 2005–2010, whichever was the date of genetic consultation. Among high familial risk patients, 25.1 % (94/374) had genetic counseling versus 7.3 % (97/1335) of patients belonging to moderate-risk group (P < 0.001). Compared to patients without genetic counseling, those who underwent genetic counseling were significantly younger (mean, 49.5 years vs 60.0 years, P < 0.001), of higher social class (P = 0.036), had more often female offspring (P < 0.001) and a higher familial risk (P < 0.001). Statistically, significant differences in family history between patients who underwent genetic counseling and those who did not were noted: 53.9 % of counseled patients versus 24.8 % of uncounseled patients had two or more first- or second-degree relatives affected by breast/ovarian cancer (P < 0.001); these figures were 14.1 and 4.6 %, respectively, when considering affected first-degree relatives only (P < 0.001). Patients who had genetic counseling had significantly more often tumors of smaller size (P = 0.001) and received more often chemotherapy (P < 0.001) than uncounseled patients. To assess which variables were independently associated with the uptake of genetic counseling, we included in a logistic regression model all variables significantly linked to genetic counseling attendance in univariate analyses (i.e., period of diagnosis, age at diagnosis, social class, female offspring, familial risk, tumor size, and chemotherapy). Patients with a significantly higher probability of having genetic counseling were those diagnosed in 2005–2010 compared to those diagnosed in 1990–1994 (odds ratio [OR] 2.2, 95 % CI 1.2–4.0), those with female offspring versus those without (OR 2.0, 95 % CI 1.4–2.9), those with high familial risk versus those with moderate familial risk (OR 5.1, 95 % CI 3.6–7.1), and those with versus without chemotherapy (OR 1.8, 95 % CI 1.2–2.6) (Table 2). By contrast, the ORs for genetic counseling use were significantly decreased for patients aged 50–69 years (OR 0.4, 95 % CI 0.3–0.5) or aged 70 years or more (OR 0.1, 95 % CI 0.1–0.3) compared with patients under 50 years of age. We performed sensitive analyses by adding in the model the four variables found with a P value between 0.05 and 0.10 in univariate analysis (i.e., method of breast cancer detection, tumor differentiation, surgery, and radiotherapy) and also using a backward stepwise logistic regression including all the 21 parameters studied in Table 1. The results were unchanged or were not modified. None of the interactions tested was significant: the P values were 0.93 for interaction between social class and age, 0.97 for interaction between tumor size and chemotherapy, 0.96 for interaction between tumor size and radiotherapy, and 0.46 for interaction between tumor size and surgery.
Table 1

Characteristics of breast cancer patients with increased familial risk according to uptake of genetic counseling

 

Genetic counseling

Pa

Yes, n = 191

No, n = 1,518

No. (%)

No. (%)

Patient characteristics

Period of cancer diagnosis

  

<0.001

 1990–1994

20 (11)

218 (14)

 

 1995–1999

27 (14)

380 (25)

 

 2000–2004

53 (28)

451 (30)

 

 2005–2010

91 (48)

469 (31)

 

Age at diagnosis, years

  

<0.001

 <50

98 (51)

348 (23)

 

 50–69

82 (43)

811 (53)

 

 ≥70

11 (6)

359 (24)

 

Marital status

  

0.971

 Ever married

162 (85)

1,286 (85)

 

 Never married

29 (15)

232 (15)

 

Female offspringb

  

<0.001

 No

68 (36)

858 (57)

 

 Yes

123 (64)

660 (44)

 

Familial risk

  

<0.001

 Moderate

97 (51)

1,238 (82)

 

 High

94 (49)

280 (18)

 

Social class

  

0.036

 High

51 (29)

276 (20)

 

 Medium

97 (55)

834 (61)

 

 Low

30 (17)

253 (19)

 

 Unknown

13 (−)

155 (–)

 

Country of birth

  

0.591

 Switzerland

98 (51)

824 (54)

 

 Southern Europe

52 (27)

363 (24)

 

 Other

41 (22)

331 (22)

 

Sector of care

  

0.297

 Private

90 (47)

776 (51)

 

 Public

101 (53)

742 (49)

 

Tumor characteristics

Method of detection

  

0.071

 Symptoms

34 (18)

358 (24)

 

 Breast self-examination

73 (38)

477 (31)

 

 Clinical screening

10 (5)

125 (8)

 

 Mammography

74 (39)

558 (37)

 

Stage

  

0.980

 I

80 (43)

630 (43)

 

 II

81 (43)

634 (43)

 

 III

21 (11)

168 (11)

 

 IV

5 (3)

48 (3)

 

 Unknown

4 (–)

38 (–)

 

Axillary node

  

0.130

 Negative

101 (54)

883 (60)

 

 Positive

86 (46)

594 (40)

 

 Unknown

4 (–)

41 (–)

 

Histological subtype

  

0.436

 Ductal

160 (84)

1,218 (80)

 

 Lobular

23 (12)

207 (14)

 

 Other

8 (4)

93 (6)

 

Differentiation

  

0.051

 Good

47 (26)

397 (28)

 

 Moderate

81 (44)

707 (50)

 

 Poor

56 (30)

318 (22)

 

 Unknown

7 (–)

96 (–)

 

Size, mm

  

0.001

 1–20

121 (78)

858 (65)

 

 21–40

23 (15)

379 (29)

 

 >40

11 (7)

90 (7)

 

 Unknown

36 (–)

191 (–)

 

Estrogen receptor statusc

  

0.204

 Positive

141 (84)

1,106 (87)

 

 Negative

27 (16)

159 (13)

 

 Unknown

23 (–)

253 (–)

 

Progesterone receptor statusc

  

0.429

 Positive

125 (74)

979 (77)

 

 Negative

43 (26)

290 (23)

 

 Unknown

23 (–)

249 (–)

 

HER2 receptor statusd

  

0.446

 Positive

20 (16)

137 (19)

 

 Negative

103 (84)

578 (81)

 

 Unknown

68 (–)

803 (–)

 

Treatment characteristics

Surgery

  

0.096

 No

4 (2)

89 (6)

 

 Mastectomy

51 (27)

393 (26)

 

 Breast conserving

136 (71)

1,036 (68)

 

Radiotherapy

  

0.096

 No

32 (17)

334 (22)

 

 Yes

159 (83)

1,184 (78)

 

Chemotherapy

  

<0.001

 No

74 (39)

920 (61)

 

 Yes

117 (61)

598 (39)

 

Hormonotherapy

  

0.966

 No

53 (28)

419 (28)

 

 Yes

138 (72)

1,099 (72)

 

a χ2 test of heterogeneity

bExtracted from the Cantonal Population Office

cRecorded since 1995

dRecorded since 2001

Table 2

Adjusted odds ratio (OR) for the uptake of genetic counseling

Variables

Adjusted ORa (95 % CI)

P(heterogeneity)

Period of diagnosis

 1990–1994

1 (Reference)

0.001

 1995–1999

0.8 (0.4–1.6)

 

 2000–2004

1.3 (0.7–2.5)

 

 2005–2010

2.2 (1.2–4.0)**

 

Age at diagnosis, years

 <50

1 (Reference)

<0.001

 50–69

0.4 (0.3–0.5)***

 

 ≥70

0.1 (0.1–0.3)***

 

Social class

 High

1 (Reference)

0.406

 Medium

0.8 (0.5–1.2)

 

 Low

0.7 (0.4–1.2)

 

Female offspring

 No

1 (Reference)

<0.001

 Yes

2.0 (1.4–2.9)***

 

Familial risk

 Moderate

1 (Reference)

<0.001

 High

5.1 (3.6–7.1)***

 

Tumor size, mm

 1–20

1 (Reference)

0.004

 21–40

0.4 (0.3–0.7)**

 

 >40

0.9 (0.4–1.9)

 

Chemotherapy

 No

1 (Reference)

0.004

 Yes

1.8 (1.2–2.6)**

 

aAdjusted for all other variables

* P < 0.05; ** P < 0.01; *** P < 0.001

Impact of genetic counseling uptake on breast cancer outcome

In total, 1,632 patients with a previous unilateral breast cancer diagnosis were included in the analysis of contralateral metachronous breast cancer. Among the 191 patients who underwent genetic counseling, ten women were excluded (four had synchronous bilateral breast cancer and six had contralateral preventive mastectomy associated to primary breast surgery). A total of 36 patients (2.2 %) had developed a contralateral invasive breast cancer in a median period of 2,363 days (range 430–5,984). The hormonal status of both first and second breast tumors was known for 24 of these 36 patients. A moderate agreement was found for the estrogen receptor status (kappa = 0.44, P = 0.028) whereas no agreement between the first and second tumors was observed for the progesterone receptor status (kappa = 0.08, P = 0.63). Table 3 shows patient, tumor, and treatment characteristics significantly associated with metachronous contralateral breast cancer occurrence in a univariate analysis. Genetic counseling use, diagnosis of cancer in the earlier periods, younger age at diagnosis, high familial risk, larger tumor size, mastectomy, lack of estrogen receptor of the first tumor, the absence of hormonotherapy (which is linked to estrogen receptor status) appeared as risk factors for the development of metachronous contralateral breast cancer in our study. Cumulative risks for developing metachronous contralateral breast cancer after the first diagnosis were significantly different between patients who underwent a genetic consultation versus those who did not (logrank test P < 0.001, Fig. 1). The delay for contralateral breast cancer occurrence was similar for patients with genetic counseling and for those without genetic counseling (2,550 and 3,194 days, respectively; P = 0.298). In a Cox model accounting for other variables significantly associated with contralateral breast cancer occurrence (i.e., period of diagnosis, age at diagnosis, familial risk, tumor size, breast surgery and hormonotherapy), the risk (hazard ratio [HR]) for a contralateral breast cancer for patients with versus without genetic counseling was 2.2 (95 % CI 1.0–5.2, P = 0.063). Further adjustment on other potential determinants of metachonous breast cancer, including social class, sector of care, method of breast cancer detection, tumor differentiation, surgery, and radiotherapy, yielded a HR 2.0 (95 % CI 0.9–4.9, P = 0.108).
Table 3

Factors associated with occurrence of metachronous contralateral breast cancer

Variables

n

Metachronous contralateral breast cancer n (%)

Pa

Genetic counseling

 No

1,451

26 (1.8)

0.001

 Yes

181

10 (5.5)

 

Period of diagnosis

 1990–1994

229

9 (3.9)

0.001

 1995–1999

386

15 (3.9)

 

 2000–2004

482

10 (2.1)

 

 2005–2010

535

2 (0.4)

 

Age at diagnosis, years

 <50

426

15 (3.5)

0.021

 50–69

857

19 (2.2)

 

 ≥70

349

2 (0.6)

 

Familial risk

 Moderate

1,278

22 (1.7)

0.011

 High

354

14 (4.0)

 

Tumor size, mm

 1–20

945

15 (1.6)

0.022

 21–40

377

9 (2.4)

 

 >40

94

6 (6.4)

 

 Unknown

216

6 (2.8)

 

Surgery

 No

91

0 (0.0)

0.044

 Mastectomy

384

14 (3.6)

 

 Breast conservation

1,157

22 (1.9)

 

Hormonotherapy

 No

452

18 (4.0)

0.002

 Yes

1,180

18 (1.5)

 

aχ2 test of heterogeneity; univariate analysis

https://static-content.springer.com/image/art%3A10.1007%2Fs10549-014-2864-3/MediaObjects/10549_2014_2864_Fig1_HTML.gif
Fig. 1

Cumulative risk for metachronous contralateral breast cancer according to uptake of genetic counseling

Among the 181 patients who benefited from genetic counseling included in the analysis of contralateral metachronous breast cancer, 111 (61.3 %) performed genetic testing and 18 of them (16.2 %) carried BRCA1/BRCA2 germ-line mutations. Stratification by BRCA1/BRCA2 mutation status showed that the occurrence of contralateral breast cancer was 8-fold higher among mutation carriers than among non-carriers (33 vs 4 %, P < 0.001). The risk of developing a contralateral breast cancer for patients with versus without BRCA1/BRCA2 mutations was significantly higher (crude HR 10.5, 95 % CI 2.9–38.5, P < 0.001).

The overall survival did not significantly differ between patients who underwent genetic counseling and those who did not (age-adjusted HR 0.8, 95 % CI 0.5–1.3, P = 0.906). Similarly, breast cancer-specific survival was not significantly different between the two groups (age-adjusted HR 0.8, 95 %CI 0.5–1.4, P = 0.484).

Discussion

To our knowledge, this is the first population-based study assessing the rate of genetic counseling uptake among breast cancer patients with a positive family history and to demonstrate the impact of such counseling on breast cancer outcome. This study has been carried out with the support of the Oncogenetics and Cancer Prevention Unit, which is the only center providing genetic consultation since 1994 for breast cancer familial aggregation or hereditary predisposition syndromes, and the Geneva Cancer Registry, which is the oldest registry in Switzerland to collect and analyze cancer data for the entire population of Geneva. In Switzerland, modalities of breast cancer treatment are chosen by a multidisciplinary staff including surgeons, chemo- and radiotherapists according to the international clinical guidelines. Procedures are the same in the public and the private sectors of care. Accordingly, all patients are offered the same options of treatment at first issue. Globally, 11.2 % of breast cancer patients that could potentially benefit from genetic counseling were effectively seen at the Oncogenetics and Cancer Prevention Unit. The geographic dispersal of some of the families and the cosmopolitan characteristics of the Geneva population with an important turn-over may partially explain this low proportion of uptake compared to less mobile populations. It is possible that some of breast cancer patients sought counseling somewhere out of the Geneva canton, so that 11.2 % would be a conservative estimate. Approximately, one out of four high familial risk breast cancer patients underwent genetic counseling. In our opinion, this is an important result since consultation process, genetic analysis, and specific surveillance/preventive measures are mostly addressed to high- rather than moderate-breast cancer risk patients. Our data are consistent with previous studies that found a statistically significant association between having a positive family history and hypothetical or real uptake of BRCA1/BRCA2 genetic testing [21, 30, 31].

In this study, we showed that recent period of breast cancer diagnosis, young age at diagnosis, female offspring, and high familial risk were statistically significant determinants of genetic counseling uptake. The increasing use of genetic counseling over time has previously been reported [32]. This is not surprising considering the better knowledge on prevalence and penetrance of BRCA1/BRCA2 mutations [14, 3338], the impact of risk management recommendations on breast cancer prevention and mortality [39, 40], and the growing interest in genetics showed by physicians and general population. Therefore, it is anticipated that a higher proportion of breast cancer patients and their relatives will benefit from appropriate clinical recommendations in the future. Early-onset breast cancer is a criterion to refer patients to genetic counseling [4143]. Thus, it was expected that breast cancer patients diagnosed before 50 years of age were more likely to undergo genetic counseling than patients diagnosed after 50 years of age. The association between genetic counseling uptake and female offspring found in our study is in agreement with results from previous studies on predictive BRCA1/BRCA2 genetic testing [1520]. These studies showed that one of the currently reported motivations to undergo genetic counseling is to determine the cancer risk of the offspring, which may in the future have an impact on their clinical surveillance.

With regard to a possible association between social class levels and use of genetic counseling, we cannot exclude that patients belonging to lower socio-economic classes could be less concerned by genetic counseling and medical screening than individuals of higher social level. Of note, this type of medical consultation, as well as the cost of testing for most of the cancer genetic predisposition syndromes, is covered by the mandatory health insurance system for all Swiss citizens. Consequently, in our opinion, genetic counseling-related costs per se would not or little influence the decision to uptake genetic counseling. The fact that genetic counseling uptake was higher among patients with high social level could be related to a more effective communication process between the physicians and their patients. Noteworthy, receiving chemotherapy treatment was another factor associated with the uptake of genetic counseling. This result could reflect a particular awareness about genetic and familial risk of medical oncologists compared to other physicians [44]. It could also reflect a heightened awareness regarding more adequate or more complete cancer treatment among breast cancer patients with affected family members who underwent genetic counseling.

We found a twofold higher risk of metachronous contralateral breast cancer among patients with genetic counseling uptake than for those without. This result is consistent with the fact that early-onset disease and strong positive family history are both criteria to refer people to genetic evaluation and that they are well-established risk factors for contralateral breast cancer [7, 8, 10, 45]. However, it could also be a consequence of a more intensive surveillance program/self-observation recommended by genetic consultants to moderate- and high-risk patients. Notably, in five out of 10 patients who underwent genetic counseling and developed metachronous breast cancer, contralateral cancer diagnosis occurred after the genetic consultation process and all of these patients followed an enhanced surveillance.

No significant differences in overall survival and in breast cancer-specific survival were observed between patients with and without genetic counseling uptake. Despite a study period of 20 years, the number of deaths from breast cancer was limited among patients who had genetic counseling (n = 15) and, accordingly, the statistical power was low to detect differences between the two groups.

The main limitation of the study is the fact that information concerning cancer family history was extracted from medical files or derived by questioning physicians and not directly from the patients. The familial risk of breast cancer was established integrating information available at the time of cancer diagnosis; therefore, patients who developed a positive family history after breast cancer diagnosis could have been missed. Moreover, family structure/size and details of the concerned branch of the family were not taken into account to categorize the familial risk. However, information on family history is accurate as attested by a previous study of our group [46]. Education is not routinely recorded by the Geneva Cancer Registry. We used patient’s last occupation, or if unemployed, that of the spouse in order to create a three-level indicator of socioeconomic status. Occupational social class and education measures have been shown to be highly correlated. Despite the fact that the 3 levels of socioeconomic indicators used can only be considered as an approximate indicator of education level, confounding from education is likely to be limited.

In conclusion, we evaluated the actual proportion of breast cancer patients with positive family history of breast cancer that benefit from genetic counseling and assessed the clinical outcome in terms of contralateral breast cancer occurrence and survival in our population. The uptake of genetic counseling among patients with personal and family history of breast cancer is notably increasing overtime. We found a significant proportion of high familial risk patients effectively undergoing genetic consultation process. A better knowledge of demographic and clinical determinants of attending a cancer genetic unit should be helpful to improve the genetic counseling activity and thus, to reach more at-risk breast cancer individuals concerned by effective surveillance and prevention measures.

Acknowledgments

This study was financially supported by grants KLS 02544-02-2010 and KLS 2946-02-2012 from the Swiss Cancer League. This investigation was also supported by a grant of the Geneva Cancer League and Fondation Dr. Henri Dubois-Ferrière Dinu Lipatti.

Conflict of interest

The authors have no conflicts of interest and do not have any disclosures to declare.

Copyright information

© Springer Science+Business Media New York 2014