Both the authors have contributed equally to this work and therefore are regarded as joined first authors.
High prevalence of BRCA1 founder mutations in Greek breast/ovarian families
Article first published online: 20 OCT 2013
© 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Special Issue: BRCA1 and BRCA2
Volume 85, Issue 1, pages 36–42, January 2014
How to Cite
High prevalence of BRCA1 founder mutations in Greek breast/ovarian families., , , , , , , , , , , , .
The authors declare that they have no conflict of interest.
- Issue published online: 12 DEC 2013
- Article first published online: 20 OCT 2013
- Accepted manuscript online: 6 SEP 2013 09:33AM EST
- Manuscript Revised: 4 SEP 2013
- Manuscript Accepted: 4 SEP 2013
- Manuscript Received: 25 JUL 2013
- Greek General Secretary for Research and Technology (GSRT) Program
- Research in Excellence II
- Ministry of Education, Lifelong Learning & Religious Affairs
- breast cancer;
- germline mutation;
- ovarian cancer
We have screened 473 breast/ovarian cancer patients with family history, aiming to define the prevalence and enrich the spectrum of BRCA1/2 pathogenic mutations occurring in the Greek population. An overall mutation prevalence of 32% was observed. Six BRCA1 recurrent/founder mutations dominate the observed spectrum (58.5% of all mutations found). These include three mutations in exon 20 and three large genomic deletions. Of the 44 different deleterious mutations found in both genes, 16 are novel and reported here for the first time. Correlation with available histopathology data showed that 80% of BRCA1 carriers presented a triple-negative breast cancer phenotype while 82% of BRCA2 carriers had oestrogen receptor positive tumours. This study provides a comprehensive view of the frequency, type and distribution of BRCA1/2 mutations in the Greek population as well as an insight of the screening strategy of choice for patients of Greek origin. We conclude that the Greek population has a diverse mutation spectrum influenced by strong founder effects.
Germline mutations in the BRCA1 (MIM# 11370) and BRCA2 (MIM# 600185) genes cause hereditary breast and/or ovarian cancer (HBOC) [1, 2]. Up to date, more than 5000 distinct mutations and sequence alterations have been reported within both genes since they were first identified in the mid-1990s. The frequency and distribution of pathogenic mutations in BRCA1 and BRCA2 vary across populations, mainly due to founder effects [3, 4]. Some populations appear to have a very high percentage of founder mutations, while other, genetically heterogeneous populations present a broader mutational spectrum.
In our previous work on the prevalence of BRCA1 and BRCA2 mutations in the Greek population, we revealed a broad and more variable mutational spectrum of mainly BRCA1 but also BRCA2 mutations, with four BRCA1 mutations accounting for 63% of all mutations detected in both genes, while the rest were unique or low-frequency mutations, reflecting the population's genetic heterogeneity . Here we present additional data, after screening a further 473 breast/ovarian cancer families, offering valuable insights on the screening strategy that should be employed on individuals of Greek origin in order to reduce costs and testing time.
Materials and methods
Patient study group
Index patients from 473 families with breast and/or ovarian cancer were screened for BRCA1 and BRCA2 mutations. Patients were enroled ad hoc from several hospitals located in various geographical areas of Greece in collaboration with the Hellenic Cooperative Oncology Group (HeCOG). Eligibility criteria were based on pedigree information (two relatives with either breast cancer below 60 years of age or ovarian cancer at any age), histopathology (triple-negative breast cancer below 50 years and a relative with breast or ovarian cancer at any age) and rare phenotypes [male breast cancer, synchronous or metachronous breast and/or ovarian cancer, early onset (<35 years) of disease]. The study was approved by the Bioethics committee of NCSR ‘Demokritos’ (240/EH▵/11.3) and was in agreement with the 1975 Helsinki statement, revised in 1983. Written informed consent was obtained from all individuals.
BRCA1/2 mutation analysis
DNA isolation from blood, polymerase chain reaction (PCR) amplification and DNA sequencing were performed as previously described . BRCA1 and BRCA2 genes were amplified using intronic primer pairs flanking each exon. Three diagnostic PCR reactions were also performed to detect the Greek founder genomic rearrangements involving exons 20, 23 and 24 ( [7, 8]; Pertesi et al. in preparation). In 164 cases of high-risk families negative for BRCA1/2 mutations by Sanger sequencing, genomic DNA was also examined by multiplex ligation-dependant probe analysis (MLPA) analysis (MRC-Holland, Amsterdam, the Netherlands) . Mutations were confirmed in an independent blood sample by sequencing in both forward and reverse directions. All nucleotide numbers refer to the wild-type cDNA reference sequence of BRCA1 NM_007294 and BRCA2 NM_000059. Mutation nomenclature is according to Human Genome Variation Society (HGVS) guidelines . Primer sequences and protocols are available upon request.
In silico prediction tools
The functional consequences of the variants of unknown clinical significance (VUS) identified were investigated using the in silico prediction tools sift (http://sift.jcvi.org/), polyphen (http://genetics.bwh.harvard.edu/pph2/) and aligngvgd (http://agvgd.iarc.fr/). Also, the ESE finder splice site prediction program (http://rulai.cshl.edu/cgi-bin/tools/ESE3/esefinder.cgi?process=home) was used to predict the variants' influence on splicing (default threshold).
The study was performed on 473 Greek patients with breast and/or ovarian cancer history, of which 467 female and 6 male. 153 families were found to carry a mutation. The overall mutation rate for both genes was 32.3% (153/473). The frequencies of BRCA1 and BRCA2 mutations were 28.5% (135/473) and 3.8% (18/473), respectively.
BRCA1 mutation spectrum
In BRCA1, 28 different deleterious mutations were found in 135 families. Of these, nine are novel (Table 1). Fourteen out of the total 28 BRCA1 mutations are frameshift, six are nonsense mutations, two are splice site mutations that cause exon skipping (c.5406+5G>C  and c.5467G>A ), and two are missense mutations previously classified as deleterious (p.Cys61Gly  and p.Gly1738Arg ). All four large genomic rearrangements detected are confined to the Greek population ( [5, 7, 8]; Pertesi et al. in preparation).
|Exon||Mutation HGVS (BIC)||Times observed||Patient numbera|
|5||c.181T>G - p.Cys61Gly (300T>G - C61G)||6||406 (O-39), 849 (B-45), 944 (O-59), 950 (B-48), 1126 (B-37), 1164 (B-46)|
|7||c.331delG (449delG-ter118)b||1||1085 (B-55)|
|11||c.1210insCT (1329insCT)b||1||1181 (B-39)|
|c.1504_1508delTTAAA (1623del5)||1||932 (B-47)|
|c.1505T>G - p.Leu502X (1624T>G - L502X)b||1||1152 (O-51)|
|c.1687C>T - p.Gln563X (1806C>T - Q563X)b||1||831 (B-37)|
|c.1953_1956delGAAA (2072del4)||1||940 (B-43)|
|c.2648insGGCA- (2767insGGCA)b||1||870 (O-44)|
|c.2963C>A - p.Ser988X (3082C>A - S988X)b||1||796 (B-29)|
|c.3178G>T - p.Glu1060X (3297G>T - E1060X)||1||1016 (B-34)|
|c.3375_3376delTC (3494delTC)||2||483 (B-30), 1144 (B-20)|
|c.3607C>T - p.Arg1203X (3726C>T - R1203X)||5||430 (B-40), 865 (B-53), 1004 (B-57), 1117 (B-40), 1025 (B-32)|
|c.3625delA (3741delA)b||1||840 (B-30)|
|c.3700_3704delGTAAA (3819del5)||7||558 (B-30), 657 (O-48), 699 (B-42), 749 (B-33), 918 (B-42), 964 (B-41), 1072 (B-40)|
|c.3755_3758delTGTC (3874del4)||2||914 (B-50), 1008 (B-38)|
|12||c.4168_4169delTG (4286delTG)b||1||824 (B-33)|
|14||c.4391_4393delCTAinsTT (4510delCTAinsTT)||1||1127 (B-36)|
|20||c.5212G>A - p.Gly1738Arg (5331G>A - G1738R)||17||418 (B-43), 419 (B-29), 466 (B-46), 634 (B-43), 707 (B-26), 710 (B-35), 739 (B-24), 750 (B-35), 769 (B-56), 798 (B-39), 877 (B-54), 915 (O-67), 1045 (B-33), 1048 (B-74), 1079 (O-53), 1129 (O-51), 1150 (B-35)|
|c.5251C>T - p.Arg1751X (5370C>T - R1751X)||8||439 (O-48), 601 (B-47), 637 (B-51), 737 (B-45), 863 (B-47), 952 (O-55), 1142 (B-42), 1198 (B-30)|
|c.5266dupC (5382insC)||26||412 (B-57), 454 (B-40), 520 (O-35), 521 (O-40), 522 (B-49), 578 (B-35), 635 (B-49), 655 (B-30), 658 (B-41), 709 (B-39), 777 (B-29), 825 (B-66), 887 (B-33), 890 (B-45), 899 (B-50), 933 (B-29), 946 (O-57), 951 (O-38), 954 (O-45), 963 (B-40), 1078 (O-63), 1080 (O-58), 1147 (B-59), 1159 (B-36), 1186 (O-48), 1205 (O-56), 1228 (B-34), 1236 (B-36)|
|c.5256_5277+3179del3200 (g.71660_74860del3200)||7||428 (O-56), 441 (B-50), 725 (B-40), 817 (B-34), 900 (O-65), 1024 (B-46), 1160 (B-33)|
|21||c.5328delC (5447delC)||1||553 (B-32)|
|22||c.5406+5G>C (IVS22+5G>C) c||1||968 (B-31)|
|23||c.5431C>T - p.Gln1811X (5550C>T - Q1811X)b||3||500 (B-29), 757 (B-30), 960 (B-41)|
|c.5467G>A (5586G>A)c||2||576 (B-43), 628 (B-55)|
|c.5406+664_*8273del (g.80280_91331del11052)||22||416 (B-37), 422 (B-33), 455 (O-30), 543 (B-31), 607 (B-35), 661 (B-47), 686 (B-32), 795 (B-33), 851 (B-26), 891 (B-40), 902 (B-33), 930 (B-56), 976 (B-29), 1036 (B-51), 1055 (O-68), 1056 (O-39), 1057 (O-49), 1058 (O-59), 1073 (B-40), 1103 (B-36), 1214 (O-34)|
|24||c.5492delC (5611delC)||1||782 (B-29), 1247 (O-35)|
|c.5468-285_5592 +4019del4429_insCACAG (g.82651_87079del4429_ins5)||13||440 (O-48), 442 (B-38), 468 (B-38), 469 (O-51), 514 (B-33), 567 (B-29), 736 (B-28), 893 (O-50), 896 (O-59), 1053 (B-38), 1059 (B-29), 1130 (O-49), 1131 (B-37)|
The mean age of diagnosis for first primary cancer among the 135 probands carrying a BRCA1 mutation is 40.2 years for breast cancer (range 20–56 years, 102 carriers) and 51.2 years for ovarian cancer (range 30–68 years, 33 carriers).
BRCA2 mutation spectrum
Analysis of the BRCA2 gene revealed 16 different deleterious mutations in 18 families, 7 of which are novel (Table 2). Apart from the apparently causative frameshift and nonsense mutations, two splice site mutations were found: c.7976G>A which causes in-frame skipping of exon 17  and c.7806-2A>T, a novel mutation predicted to cause the same splicing effect as the Slovenian founder c.7806-2A>G . It is worth highlighting that three of six male patients that were screened were found to be BRCA2 mutation carriers. The mean age of diagnosis for first primary cancer among the 15 female probands carrying a BRCA2 mutation is 39.9 years for breast cancer (range 28–50 years).
|Exon||Mutation HGVS (BIC)||Times observed||Patient numbera|
|8||c.658_659delGT (886delGT)||1||1111 (B-38)|
|11||c.2339C>G - p.Ser780X (2567C>G - S780X)||1||1040 (mB-71)|
|c.2808_2811delAAAC (3036del4)||1||683 (mB-44)|
|c.4415_4418delAGAA (4643del4)||2||967 (B-30), 1204 (B-42)|
|c.4767delA (4997delA)||1||427 (B-38)|
|c.5722_5723delCT (5950delCT)||1||653 (B-49)|
|c.6490C>T - p.Gln2164X (6718C>T - Q2164X)||1||969 (B-42)|
|c.6600-6601delTT (6828delTT)||1||1002 (B-41)|
|17||c.7806-2A>T (IVS16-2A>T) b, c||1||560 (B-31)|
|c.7976G>A (8204G>A )||1||1210 (B-45)|
|19||c.8364G>A - p.Trp2788X (8592G>A - W2788X)||1||1124 (B-43)|
|22||c.8930delA (9158delA)||1||1069 (B-41)|
|23||c.8990_9022del32 (9218del32)||1||465 (B-28)|
|c.9097dupA (9325insA)||2||630 (mB-61), 989 (B-33)|
|25||c.9376C>T - p.Gln3126X (9604C>T - Q3126X)||1||666 (B-47)|
|27||c.9748_9749insT (9976insT)||1||543 (B-50)|
Polymorphisms and variants of uncertain clinical significance in BRCA1 and BRCA2
Five BRCA1 and 11 BRCA2 VUS reported here for the first time are shown in Table 3. The pathogenic potential of putative novel splicing mutations was analysed by ESE finder and none of the novel VUSs identified were suggested to affect splicing. Analysis by in silico prediction tools (sift, polyphen and aligngvgd) classified 4 of the 11 changes as 'probably damaging', but only variant p.L82P in exon 6 seems to be most likely to be pathogenic. Existing functional data showing a severe effect of this change on the BRCA1-BARD1 heterodimer formation support this prediction .
|Gene/exon||Nucleotide||Codon||Base change||AA change||Mutation HGVS||polyphen-2a prediction||polyphen score||sift||Align gvgd|
|BRCA1/6||c.245||82||T>C||Leu to Pro||p.Leu82Prob||Probably damaging||0.998||Damaging||C65|
|BRCA1/8||c.457||153||A>G||Ser to Gly||p.Ser153Gly||Possibly damaging||0.799||Tolerated||C0|
|BRCA2/4||c.374||125||A>T||Asp to Val||p.Asp125Val||Benign||0.437||Damaging||C0|
|BRCA2/7||c.569||190||C>T||Pro to Leu||p.Pro190Leub||Probably damaging||0.997||Damaging||C15|
|BRCA2/10||c.1022||341||G>A||Cys to Tyr||p.Cys341Tyr||Benign||0.001||Tolerated||C0|
|BRCA2/11||c.3628||1210||G>A||Asp to Asn||p.Asp1210Asn||Benign||0.007||Tolerated||C0|
|BRCA2/11||c.3669||1223||T>G||His to Gln||p.His1223Gln||Benign||0.168||Tolerated||C0|
|BRCA2/11||c.4584||1528||C>G||Ser to Arg||p.Ser1528Argb||Probably damaging||0.999||Tolerated||C25|
|BRCA2/11||c.4648||1550||G>A||Glu to Lys||p.Glu1550Lys||Benign||0||Tolerated||C0|
|BRCA2/11||c.5200||1734||G>A||Glu to Lys||p.Glu1734Lys||Benign||0.410||Tolerated||C0|
|BRCA2/27||c.9863||3288||C>T||Thr to Ile||p.Thr3288Ileb||Probably damaging||0.999||Damaging||C15|
BRCA1-BRCA2 carriers' histopathology
The triple-negative tumour immunophenotype has been strongly associated with BRCA1-related hereditary breast cancers, while BRCA2-related cancers resemble sporadic breast cancer tumour phenotypes [18, 19]. In our study, and consistently with previous observations [18, 20, 21], 80% of BRCA1 carriers developed triple-negative breast cancer, while BRCA2 carriers had predominantly ER-positive tumours (82%, Table 4).
|Type||Total (%)||BRCA1 carriers in each category (%)||BRCA2 carriers in each category (%)|
|TNBC||151 (39.9)||82 (80.4)||2 (11.7)|
|ER+||175 (46.2)||11 (10.8)||14 (82.4)|
|Other||50 (13.2)||9 (8.8)||1 (5.9)|
We have screened the BRCA1 and BRCA2 genes for mutations in 473 Greek breast/ovarian cancer families, resulting in a total of 153 loss-of-function mutations among this cohort. This article revises our data on the BRCA1/BRCA2 mutation spectrum in the Greek population; present data were combined with previously published results involving a total of 760 families [5, 7, 22-24], in which patients were selected using the same inclusion criteria. This is the largest study on the prevalence and spectrum of BRCA1 and BRCA2 genes in the Greek population with familial breast/ovarian cancer today.
Our cumulative mutation data on 760 breast/ovarian cancer families showed deleterious mutations in BRCA1 or BRCA2 in 212 cases (27.9%). The prevalence of BRCA1 mutations was almost six times the prevalence of BRCA2 mutations: 181 families (85%) carried a BRCA1 mutation, while 31 families (15%) carried a BRCA2 mutation. We should highlight that 75% (136/181) of all mutations in BRCA1 are located at the 3′ end of the gene (exons 20–24) due to founder effects, while mutations in BRCA2 seem to be unique and are largely scattered on its entire sequence. A graphic representation of the Greek population BRCA1 and BRCA2 mutation spectrum is given in Fig. 1.
The current mutational spectrum of the Greek population has broadened with regard to our previous report ; recurrent mutations have emerged and 17 novel mutations (9 in BRCA1 and 8 in BRCA2) have been identified. Large genomic rearrangements in BRCA1 constitute a significant proportion of deleterious alterations found in our population (24%, 44/181). In all, the BRCA1 and BRCA2 genes display a diverse mutational spectrum in the Greek population, reflecting the population's genetic heterogeneity , although influenced by strong founder effects.
Founder mutations represent 58.5% of the total in our 212 BRCA1/BRCA2 mutation positive families and 68.5% of all BRCA1 mutations, which is high and approaches findings in other populations with strong founder effects. As we have previously stated , the presence of highly prevalent founder mutations in a population facilitates screening in a time-saving and cost-effective way. Other factors affecting screening efficiency are patient selection criteria, as well as selection of first gene to be screened according to patient/family characteristics. Table 5 summarizes frequencies of key family history characteristics among carriers. The presence of ovarian cancer was a good indicator in our cohort for the existence of a causative BRCA1 mutation; the highest mutation detection rates were obtained in families with probands diagnosed with both breast and ovarian cancer (82.6%) or with ovarian cancer (51.3%). The triple-negative immunophenotype, as clearly shown in Table 4, is also strongly indicative of the presence of a mutation in BRCA1. Also, a BRCA2 mutation was found in 3 of 11 male breast cancer patients, while none of them was found to carry a mutation in BRCA1.
|Patient category (proband)||Percentage in cohort (n = 760)||Percentage of BRCA1 carriers in each category||Percentage in BRCA1 carriers (n = 181)||Percentage of BRCA2 carriers in each category||Percentage in BRCA2 carriers (n = 31)|
|Total||100 (760)||23.9 (181/760)||100 (181)||4.1 (31/760)||100 (31)|
|Family history||78.4 (597)||25.5 (152/597)||81.7 (152)||3.4 (20/597)||66.7 (20)|
|Early onset (<40 years)||44.2 (339)||28.3 (96/339)||51.6 (96)||2.7 (9/339)||30 (9)|
|Family history and early onset||36.1 (275)||29.5 (81/275)||43.5 (81)||1.4 (4/275)||13.3 (4)|
|Ovarian cancer||10 (76)||51.3 (39/76)||20.9 (39)||0||0|
|Breast and ovarian cancer||3 (23)||82.6 (19/23)||10.2 (19)||0||0|
|Male breast cancer||1.4 (11)||0||0||27.2 (3/11)||10 (3)|
In conclusion, we have shown that familial breast/ovarian cancer in Greece is not only characterized by a wide and variable spectrum of mainly BRCA1 but also BRCA2 mutations, predominated by the 58.5% overall of the six BRCA1 founders. Given the strong founder effect in our region, we suggest that the testing strategy for breast and/or ovarian cancer families of Greek ethnicity should include these founders as a first-step screen. The complete screening of both genes, however, is highly indicated for high-risk Greek breast/ovarian cancer patients. In the light of new screening technologies emerging, based on next-generation and massively parallel sequencing, it is expected that screening of BRCA1 and BRCA2 will soon become an integrated part of a more comprehensive approach focusing on some tens of cancer predisposing genes, or even the whole genome. In that way, genome-based cancer diagnostics will be feasible for a wider group of oncology patients. However, for the smaller group of families displaying disease phenotypes characteristic of highly penetrant genes, such as BRCA1 and BRCA2, the existing technology that is readily available to the clinician in local testing facilities will be of better service for the time being.
We are indebted to the patients and their families. This study was partly supported by the Greek General Secretary for Research and Technology (GSRT) Program, Research in Excellence II, funded by 75% from the European Union and the Operational Program ‘Education & Lifelong Learning’ ESPA-THALIS#266 of the Ministry of Education, Lifelong Learning & Religious Affairs.