Journal of Cancer Research and Clinical Oncology

, Volume 132, Issue 10, pp 617–626

The prevalence of BRCA1 and BRCA2 mutations in eastern Chinese women with breast cancer

Authors

  • Chuan-Gui Song
    • Department of Breast Surgery, Breast Cancer Institute, Cancer Hospital/Cancer InstituteFudan University
  • Zhen Hu
    • Department of Breast Surgery, Breast Cancer Institute, Cancer Hospital/Cancer InstituteFudan University
  • Jiong Wu
    • Department of Breast Surgery, Breast Cancer Institute, Cancer Hospital/Cancer InstituteFudan University
  • Jian-Min Luo
    • Department of Breast Surgery, Breast Cancer Institute, Cancer Hospital/Cancer InstituteFudan University
  • Zhen-Zhou Shen
    • Department of Breast Surgery, Breast Cancer Institute, Cancer Hospital/Cancer InstituteFudan University
  • Wei Huang
    • Chinese National Human Genome Center at Shanghai
    • Department of Breast Surgery, Breast Cancer Institute, Cancer Hospital/Cancer InstituteFudan University
Original Paper

DOI: 10.1007/s00432-006-0105-9

Cite this article as:
Song, C., Hu, Z., Wu, J. et al. J Cancer Res Clin Oncol (2006) 132: 617. doi:10.1007/s00432-006-0105-9

Abstract

To investigate the prevalence of BRCA1 and BRCA2 mutations among Chinese patients, we studied 70 Shanghai cases with early onset breast cancer and affected relatives, and mutation screening was performed in the whole-gene sequence of BRCA1 and BRCA2 by polymerase chain reaction-based denaturing high-performance liquid chromatography. Six disease-causing mutations in BRCA1 (8.6%) and two in BRCA2 (2.9%) were detected, including four novel mutations that were all in the BRCA1 gene (3449insA, IVS17-1G>T, IVS21+1G>C and 5587-1del8). Additional sequence variants identified included 30 polymorphisms (18 in BRCA1 and 12 in BRCA2) and a novel mis-sense mutation of unknown significance in BRCA2 (5911G>C). The 9.5 and 2.4% patients with breast cancer diagnosed before the age of 35 were BRCA1- and BRCA2-mutation carriers, and the prevalence of BRCA1 and BRCA2 mutations in the families with two or more affected individuals were 12.1 and 3.0%, respectively. In these families, all the BRCA1 and BRCA2 mutations were detected in the families containing at least one case diagnosed under the age of 40, and in the families whose youngest patients were diagnosed before the age of 35, the prevalence of BRCA1 and BRCA2 were as high as 40 and 20%, respectively. Based on this information, we conclude that genetic testing should be performed among patients with early onset breast cancer (<40 years), especially combined with family history.

Keywords

Breast cancerBRCA1 and BRCA2Sequence variationDHPLCChinese

Introduction

BRCA1 and BRCA2 gene (MIM# 113705 and MIM# 600185) germ-line mutation distribution and frequency in breast cancer patients have been widely studied in western populations, and the genetic screening guideline for high-risk women also have been established in some developed countries. But little is known regarding the detailed prevalence of BRCA1 and BRCA2 mutation in Chinese women, especially in mainland.

In our former study (Hu et al. 2003), we analyzed 22 breast cancer diagnosed before the age of 35 and 20 patients with a family history in Shanghai for BRCA1 germ-line mutation. Three novel disease-associated mutations (582C>T, 735C>T and 2790delT) were detected, and the prevalence of BRCA1 mutation is 9.1% in patients with early onset breast cancer and 5% in patients with family history, respectively.

This result gives us some preliminary information about prevalence of BRCA1 mutation in Chinese women, and suggests the necessity of genetic testing in Chinese breast cancer patients. However, some problems in the genetic testing of BRCA1 and BRCA2 mutation should be taken into consideration as follows: (1) who should receive genetic testing? The genetic screening of BRCA1 and BRCA2 mutations is too expensive and complicated to be widely used in developing countries, such as China. (2) Is there some ‘hot spot’ of BRCA1 and BRCA2 mutation in Chinese population? Screening the entire length for both BRCA1 and BRCA2 genes is also costly. To get more helpful information for genetic screening in Chinese high-risk women and also to investigate the prevalence of BRCA2 mutation in Chinese population, we conduct our further study of BRCA1 and BRCA2 analysis in Chinese breast cancer patients in Shanghai with early onset disease or a positive family history.

Materials and methods

Samples

All the primary breast cancer patients were diagnosed between the year 2002 and 2004 in the Department of Breast Surgery at Cancer Hospital/Institute, Fudan University, Shanghai, China. The inclusion criteria were: (1) at least one first-degree relative with breast cancer, regardless of age or (2) breast cancer diagnosed below age 35.

After having obtained a written informed consent, a 10-ml blood sample was obtained from each patient and mixed with 1.6 ml of ACD solution immediately after collection. Genomic DNA was isolated from the blood leukocytes using standard procedures. The blood samples of affected relatives were also collected, and the youngest patient in each family was analyzed for mutations.

Polymerase chain reaction

Polymerase chain reaction (PCR) primer pairs and conditions as former publications were used to amplify exon and intron–exon boundaries from genomic DNA (Arnold et al. 1999; Gross et al. 2000; Meyer et al. 2003). PCR was carried out in 50-μl volumes containing 100 ng of genomic DNA, 1.5 mM MgCl2, 50 mM KCl, 10 mM Tris–HCl (pH 8.3), 200 μM dNTPs, 0.5 μM of each primer and 0.2 μl of Taq polymerase (5 U/μl) (Qiagen, Valencia, CA, USA). The PCR conditions were: denaturing at 95°C for 5 min followed by 35 cycles of denaturing at 94°C for 30 s, annealing at 55°C for 45 s and extension at 72°C for 1 min; one final step at 72°C for 5 min. GeneAmp® 9700 PCR System (ABI Applied Biosystem, Foster City, CA, USA) was used for all PCR amplification.

Denaturing high-performance liquid chromatography

Prescreening for BRCA1 and BRCA2 mutations was carried out by denaturing high-performance liquid chromatography (DHPLC) analysis (WAVE DNA Fragment Analysis System, Transgenomic, Omaha, NB, USA) as described by former publication (Arnold et al. 1999). The DHPLC gradients and temperatures were determined by use of WAVE®maker software and properly adjusted. PCR products were subjected to an additional 3 min 95°C denaturing step followed by gradual re-annealing from 95 to 65°C over a period of 30 min prior to analysis, and then eluted with a linear acetonitrile gradient at a flow rate of 0.9 ml/min, finally detected by online ultraviolet absorbance monitoring at 254 nm.

DNA sequencing

After DHPLC analysis, DNA sequencing was performed in re-amplified PCR products from samples containing variants. Both strands were sequenced by an automated ABI 3730 DNA sequencer (Applied Biosystems, Foster City, CA, USA) following the manufacturer’s instructions. All nucleotide numbers referred to the wild type cDNA sequence of BRCA1 (U14860.1) and BRCA2 (U43746.1) as reported in GeneBank.

Result

Patients’ characteristics

In the total 70 patients, who fulfilled the inclusion criteria, 33 of them had at least one first-degree affected relative, 42 patients had breast cancer diagnosed before age 35, while five patients had both early onset breast cancer and affected relatives. In the 33 families with two or more cases, 18 contained two cases, 12 had three and 3 families had four cases. Eleven of the families contained at least one case diagnosed under age 40, 5 of them had at least one case diagnosed under age 35, while in 22 families all cases were diagnosed at age 40 or older.

Characterization of alterations in BRCA1 and BRCA2 sequence

Prevalence

A total of eight disease-associated mutations were detected in our series, six in BRCA1 and two in BRCA2. In the 42 patients with breast cancer under the age of 35, four BRCA1 mutations (9.5%) and one BRCA2 mutations (2.4%) were detected. In those families with two or more cases, four BRCA1 mutations (12.1%) and one BRCA2 mutation (3.0%) were found. In the families containing 2, 3 and 4 affected relatives, there were 3, 1 and 0 (16.7%, 8.3% and 0) family carrying BRCA1 mutations, and the number of BRCA2 mutations were 0, 1 and 0 (0, 8.3% and 0), respectively. In 22 families with all cases were diagnosed at age 40 or older, no BRCA1 or BRCA2 mutation was found, while four BRCA1 mutations (36.4%) and one BRCA2 mutation (9.1%) were detected in 11 families containing at least one patient diagnosed under age 40. In five families containing at least one case diagnosed under age 35, two families carried BRCA1 mutation (40%), and one family carried BRCA2 mutation (20%) (Table 1).
Table 1

Clinical features of cases and distribution of BRCA deleterious mutation

Characteristic

No.

No. of BRCA1 mutation (%)

No. of BRCA2 mutation (%)

Total cases

70

6 (8.6%)

2 (2.9%)

Age of onset <35 yrs

42

4 (9.5%)

1 (2.4%)

Families with two or more affected individuals

33

4 (12.1%)

1 (3.0%)

No. of affected individuals in the family

 2

18

3 (16.7%)

0 (0)

 3

12

1 (8.3%)

1 (8.3%)

 4

3

0 (0)

0 (0)

Age of youngest patients in the family

 ≥40 yrs

22

0 (0)

0 (0)

 <40 yrs

11

4 (36.4%)

1 (9.1%)

 <35 yrs

5

2 (40%)

1 (20%)

Disease-associated mutations

In the six BRCA1 disease-associated mutations, four of them were found in the families with two or more affected cases, they were 1100delAT (Fig. 1), IVS17-1G>T (Fig. 2), IVS21+1G>C (Fig. 3), and 5640delA (Fig. 4). 1100delAT was a frame-shift mutation, resulting in protein termination at codon 328, and have been reported in Breast Cancer Information Core (BIC) database. IVS17-1G>T and IVS21+1G>C were two novel splice-site alterations and had not been reported in BIC or published literature so far. Since in both cases the highly conserved consensus splice sites were affected, these variants were classified as disease-associated mutations. IVS17-1G>T was found in a breast cancer patient diagnosed at the age of 26 with her mother affected. IVS21+1G>C was harbored in an early onset patient diagnosed at the age of 33, whose mother developed breast cancer at 45. 5640delA was a frame-shift mutation, which resulted in a chain termination at codon 1842. The patient who carried this mutation was diagnosed at the age of 37, and the patient’s mother and maternal aunt also suffered from breast cancer. The other two BRCA1 disease-associated mutations were 3449insA and 5587-1del8 which were detected in two patients of early onset breast cancer with no reported family history. 3449insA was a novel frame-shift mutation and harbored in a patient with breast cancer diagnosed at the age of 31, causing a protein termination at codon 1114. 5587-1del8 was also a frame-shift mutation and had not been reported previously, which was found in a patient with breast cancer at the age of 30 and resulted in protein truncation at codon 1827 (Table 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-006-0105-9/MediaObjects/432_2006_105_Fig1_HTML.gif
Fig. 1

Mutation detection in exon 11B of the BRCA1 gene. A DHPLC chromatogram homoduplex formation (wild-type) and heteroduplex formation (mutant, arrowed). B DNA sequencing reveals a AT deletion (underlined) at nt1100, which resulted in chain termination at codon 328

https://static-content.springer.com/image/art%3A10.1007%2Fs00432-006-0105-9/MediaObjects/432_2006_105_Fig2_HTML.gif
Fig. 2

Mutation detection in exon 18 of the BRCA1 gene. A DHPLC chromatogram homoduplex formation (wild-type) and heteroduplex formation (mutant, arrowed). B DNA sequencing reveals a G→T substitution (underlined) at nt5194-1, which was a splice-site mutation

https://static-content.springer.com/image/art%3A10.1007%2Fs00432-006-0105-9/MediaObjects/432_2006_105_Fig3_HTML.gif
Fig. 3

Mutation detection in exon 21 of the BRCA1 gene. A DHPLC chromatogram homoduplex formation (wild-type) and heteroduplex formation (mutant, arrowed). B DNA sequencing reveals a G→C substitution (underlined) at nt5451+1, which was a splice-site mutation

https://static-content.springer.com/image/art%3A10.1007%2Fs00432-006-0105-9/MediaObjects/432_2006_105_Fig4_HTML.gif
Fig. 4

Mutation detection in exon 24 of the BRCA1 gene. A DHPLC chromatogram homoduplex formation (wild-type) and heteroduplex formation (mutant, arrowed). B DNA sequencing reveals a A deletion (underlined) at nt5640, which resulted in chain termination at codon 1842

Table 2

BRCA1 and BRCA2 disease-associated mutation in Chinese patients

Mutationa

Exon

AA change

Typeb

Previous reports

Age at diagnosis

Family history

BRCA1

1100delAT

Exon 11

Stop 328

FS

Yes

37

Mother

3449insA

Exon 11

Stop 1114

FS

No

31

Nil

IVS17-1G>T

Intron 17

SP

No

26

Mother

IVS21+1G>C

Intron 21

SP

No

33

Mother

5640delA

Exon 24

Stop 1842

FS

Yes

37

Mother, maternal aunt

5587-1del8

Exon 24

Stop 1827

FS

No

30

Nil

BRCA2

5802delAATT

Exon 11

Stop 1861

FS

Yes

47

Mother, sister

5950delCT

Exon 11

Stop 1910

FS

Yes

25

Nil

aGeneBank reference sequences: BRCA1 version # U14680.1 and BRCA2 version # U43746.1

bMutation types: FS frame-shift mutation, SP splice-site mutation

Two previously reported frame-shift mutations, 5802delAATT (Fig. 5) and 5950delCT,were detected in BRCA2 gene.5802delAATT was harbored in a 47-year-old breast cancer patient with her mother and sister affected by the same malignancy. A patient with breast cancer at the age of 25 carried 5950delCT, which resulted in the protein termination at codon 1910 (Table 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-006-0105-9/MediaObjects/432_2006_105_Fig5_HTML.gif
Fig. 5

Mutation detection in exon 11K of the BRCA2 gene. A DHPLC chromatogram homoduplex formation (wild-type) and heteroduplex formation (mutant, arrowed). B DNA sequencing reveals a AATT deletion (underlined) at nt5802, which resulted in chain termination at codon 1861

Mis-sense mutations of unknown significance

We detected one novel mis-sense mutation in BRCA2, which was 5911G>C (E1895Q) (Fig. 6) and located in exon 11. A patient with breast cancer diagnosed at age 35 was found to harbor the mutation. In our population, the allele frequency of this mutation was 0.007 (Table 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-006-0105-9/MediaObjects/432_2006_105_Fig6_HTML.gif
Fig. 6

Mutation detection in exon 11k of the BRCA2 gene. A DHPLC chromatogram homoduplex formation (wild type) and heteroduplex formation (mutant, arrowed). B DNA sequencing reveals a G→C transition (arrowed) at nt5911, which resulted in Glu→Gln at codon 1895

Table 3

BRCA1- and BRCA2-sequence variants of unknown significance and polymorphisms in Chinese patients

Sequence variantsa

Position

AA change

Allele frequencyb

Our study

Previous reports

BRCA1

IVS1-115T>C

Intron 1

Non-coding

0.33

Not mentioned (Karpukhin et al. 2002)

233G>A

Exon 3

Lys38Lys

0.007

0.04 (Tang et al. 1999)

IVS8-64delT

Intron 8

Non-coding

0.30

0.28 (Seo et al. 2004)

812G>A

Exon 11

Tyr231Tyr

0.02

Nil

2201C>T

Exon 11

Ser694Ser

0.43

0.40 (Hu et al. 2003)

2430T>C

Exon 11

Leu771Leu

0.11

0.38 (Hu et al. 2003)

2685T>C

Exon 11

Tyr856His

0.007

0.03 (Tang et al. 1999)

2731T>C

Exon 11

Pro871Leu

0.21

0.11 (Hu et al. 2003)

3232A>G

Exon 11

Glu1038Gly

0.17

0.24 (Tang et al. 1999)

3667A>G

Exon 11

Lys1183Arg

0.31

0.31 (Durocher et al. 1996)

IVS14+14A>G

Intron 14

Non-coding

0.007

0.005 (Seo et al. 2004)

IVS14-63C>G

Intron 14

Non-coding

0.11

Not mentioned (Reeves et al. 2004)

IVS16-92A>G

Intron 16

Non-coding

0.01

Not mentioned (Sng et al. 2003)

IVS16-68A>G

Intron 16

Non-coding

0.02

0.43 (Tang et al. 1999)

IVS17+65G>A

Intron 17

Non-coding

0.33

Nil

IVS18+66G>A

Intron 18

Non-coding

0.34

0.31 (Tang et al. 1999)

IVS19-28C>T

Intron 19

Non-coding

0.02

Nil

IVS22+32A>T

Intron 22

Non-coding

0.007

0.008 (Tang et al. 1999)

BRCA2

IVS1-26G>A

Intron 1

Non-coding

0.18

Nil

IVS4+67A>C

Intron 4

Non-coding

0.07

0.15 (Seo et al. 2004)

IVS4-79T>C

Intron 4

Non-coding

0.07

Nil

1342A>C

Exon 10

H372N

0.40

0.42 (Wagner et al. 1999)

1593A>G

Exon 10

S455S

0.05

0.11 (Wagner et al. 1999)

IVS10-51G>T

Intron 10

Non-coding

0.007

0.005 (Seo et al. 2004)

2457T>C

Exon 11

H743H

0.09

0.13 (Wagner et al. 1999)

3199A>G

Exon 11

N991D

0.01

0.15 (Wagner et al. 1999)

3624A>G

Exon 11

K1132K

0.04

0.34 (Wagner et al. 1999)

4035T>C

Exon 11

V1269V

0.13

0.23 (Wagner et al. 1999)

5911G>C

Exon 11

E1895Q

0.007

Nil

IVS14+53C>T

Intron 14

Non-coding

0.06

0.13 (Wagner et al. 1999)

IVS16-14T>C

Intron 16

Non-coding

0.39

0.048 (Wagner et al. 1999)

Sequence variants of unknown significance are highlighted in bold text

aGeneBank reference sequences: BRCA1 version # U14680.1 and BRCA2 version # U43746.1

bAllele frequency was expressed as the prevalence among 140 chromosomes

Polymorphism

We also detected 30 polymorphisms in our population, 18 in BRCA1 and 12 in BRCA2. In BRCA1, 15 polymorphisms were previously reported and 3 were novel one. Two novel and ten previously reported polymorphisms were found in BRCA2. In the five novel sequence variants, one was in exon 11 of BRCA1 and two were detected in intron, and all of them resulted in non-coding change (Table 3).

Discussion

We observed that the prevalence of BRCA1 and BRCA2 mutations in the patients with breast cancer diagnosed before the age of 35 was 9.5 and 2.4%, respectively. The prevalence of BRCA1 in these patients was similar to our former report (9.1%) (Hu et al. 2003) and other Chinese series (Sng et al. 2000; Tang et al. 1999). If we ruled out the cases with family history, the prevalence would be adjusted to 5.4% (2/37). Since the patients analyzed in our present study was nearly twofold as that in our previous study, the prevalence estimated here might be more accurate to reflect the distribution of BRCA1 in Chinese patients with early onset breast cancer. The prevalence of BRCA1 germ-line mutations in Singaporean Chinese with early onset breast cancer under the age of 35 and 40 were approximately 14 and 9%, respectively (Sng et al. 2000). In a series of Hong Kong, the prevalence was 8.0% in patients diagnosed under age 45 (Tang et al. 1999). The similar phenomenon was found in the western population, and the study in US and UK showed the prevalence in the patients with breast cancer before the age of 35 was about 5–7%. Little was known regarding the prevalence of BRCA2 mutation in the Chinese early onset breast cancer patients, and in our study, the prevalence of BRCA2 mutation in such kind of patients was 2.4%. These results gave us some preliminary information in Chinese population.

In our series, the 12.1 and 3.0% families with two or more affected individuals were BRCA1 and BRCA2 mutations carriers, respectively. As we knew, the prevalence of mutations in breast cancer patients with affected relatives was highly dependent on the number of affected relatives in the family and the age of onset of cancer. It had been partly confirmed in our study. In the families containing two or more affected relatives, all the BRCA1 and BRCA2 mutations were detected in the families containing at least one case diagnosed under the age of 40, and in the families whose youngest patients were diagnosed before the age of 35, the prevalence of BRCA1 and BRCA2 were high as 40 and 20%, respectively. So in the Chinese families containing more than one case of breast cancer, the age of onset in the youngest patients was an important predictive factor of the probability of mutations. On the other hand, in our series, the prevalence of BRCA1 in the families only containing two affected relatives (16.7%) was higher than that in the families with three or four affected relatives (8.3% and 0, respectively). The predictive value of affected relatives’ number in the probability of mutations, which had been reported widely in other population, could not be found in our breast cancer families. It was very likely to be due to the small sample size in our study, and further work should be performed to confirm it.

We found six BRCA1 and two BRCA2 disease-associated mutations, and all of them were in different locations. Combined with our former study (Hu et al. 2003), still no repeated mutations were found in our series. In the six BRCA1 mutations, two sequence variants were found in BIC or published literature. 1100delAT was previously described in a Jewish-Libyan family and two non-Jewish Caucasian individuals [Breast Cancer Information Core (BIC); FitzGerald et al. 1996; Gal et al. 2004], and it had not been found in Chinese population before. Another previously reported BRCA1 mutation was 5640delA, which was found in an American breast cancer family (Malone et al. 1998), and it was also reported in Chinese population for the first time. Two BRCA2 mutations in our study, 5802delAATT and 5950delCT, were all reported previously (Ikeda et al. 2001; Suter et al. 2004). 5802delAATT was observed in seven independent Japanese families, and haplotype analysis suggested it was a founder mutation in Japanese population (Ikeda et al. 2001). So our finding was very helpful for the study of the origin of Japanese nation. 5950delCT was reported in a case-control study performed in Shanghai breast cancer patients (Suter et al. 2004) and it was the first time to find a recurrent BRCA mutation in Shanghai population.

In our study, two BRCA1 deleterious mutations, 5640delA and 5587-1del8, both occurred in exon 24. Interestingly, in the Shanghai’s case-control study (Suter et al. 2004), 5589del8 was observed in two separate cases. Similarly, two mutations, 5802delAATT and 5950delCT, were found in BRCA2 gene and these mutation both were located in exon 11 adjacently. 5950delCT was also the recurrent BRCA2 mutation in Shanghai population. This information might be suggestive of the potentially ‘hot’ mutation regions in Chinese population, but we still needed to make more researches before we got the conclusion, because the sample sizes in these studies were very small.

In our population, we detected one novel mis-sense mutation 5911G>C (E1895Q) in BRCA2. Since the novel mis-sense mutation located in BRC repeat-region that is reported to interact with RAD1 protein, the variant was likely to have a deleterious effect. However, the analysis of biological function should be done before we classified it as a disease-associated mutation.

The other distinction between our present study and former report was that we had used different methods in mutation analysis. In the former report, we used PCR-single-strand conformation polymorphism (SSCP) assay for mutation detection. According to a blinded study (Eng et al. 2001), SSCP might miss 20–30% of sequence alterations, whereas DHPLC used in the present study allowed a highly sensitive and accurate screen for mutation detection in BRCA1/2 with sensitivity and specificity reported to range from 96 to 100% (Eng et al. 2001; Gross et al. 2000; Wagner et al. 1999).

Our two studies gave us some preliminary information of the prevalence of BRCA mutations in Shanghai population. Based on this information, genetic testing should be performed among patients with early onset breast cancer (<40 years), especially combined with family history. We also found some potentially ‘hot’ BRCA-mutation regions in Shanghai women, but we could not get any conclusion before the study of larger sample size. Still no founder mutation was revealed in Mainland China as yet, and more work needed to be done for comprehensive data on BRCA distribution in Chinese breast cancer patients. Furthermore, since China was a developing country, to choose a highly sensitive, cost-effective and fast screening method for use should be considered. In this setting, DHPLC could be an ideal tool applied to mutational analysis.

Acknowledgments

This research was supported in part by the following grants: Grant sponsor: National Key Project of China, Grant number: 2002AA711A08-34, 2001BA703BO5; Grant sponsor: Outstanding Young Investigator Award of National Natural Science Foundation of China, Grant number: 30025015; Grant sponsor: National Natural Science Foundation of China, Grant number: 30371580; Grant sponsor: The Grant from Shanghai Science and Technology Committee; Grant number: 03J14019, 04ZR14027.

Copyright information

© Springer-Verlag 2006