Clin Genet 2014 Printed in Singapore. All rights reserved
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd CLINICAL GENETICS doi: 10.1111/cge.12526
Characterization of a novel founder MSH6 mutation causing Lynch syndrome in the French Canadian population Castellsagué E., Liu J., Volenik A., Giroux S., Gagné R., Maranda B., Roussel-Jobin A., Latreille J., Laframboise R., Palma L., Kasprzak L., Marcus V.A., Breguet M., Nolet S., El-Haffaf Z., Australie K., Gologan A., Aleynikova O., Oros-Klein K., Greenwood C., Mes-Masson A.M., Provencher D., Tischkowitz M., Chong G., Rousseau F., Foulkes W.D. Characterization of a novel founder MSH6 mutation causing Lynch syndrome in the French Canadian population. Clin Genet 2014. © John Wiley & Sons A/S. Published by John Wiley & Sons Ltd, 2014 We identified an MSH6 mutation (c.10C>T, p.Gln4*) causing Lynch syndrome (LS) in 11 French Canadian (FC) families from the Canadian province of Quebec. We aimed to investigate the molecular and clinical implications of this mutation among FC carriers and to assess its putative founder origin. We studied 11 probands and 27 family members. Additionally 6433 newborns, 187 colorectal cancer (CRC) cases, 381 endometrial cancer (EC) cases and 179 additional controls, all of them from Quebec, were used. Found in approximately 1 of 400 newborns, the mutation is one of the most common LS mutations described. We have found that this mutation confers a greater risk for EC than for CRC, both in the 11 studied families and in the unselected cases: EC [odds ratio (OR) = 7.5, p < 0.0001] and CRC (OR = 2.2, p = 0.46). Haplotype analyses showed that the mutation arose in a common ancestor, probably around 430–656 years ago, coinciding with the arrival of the first French settlers. Application of the results of this study could significantly improve the molecular testing and clinical management of LS families in Quebec. Conﬂict of interest
The authors declare no conflicts of interest.
E. Castellsaguéa,b,c , J. Liud , A. Volenike , S. Girouxf , R. Gagnég , B. Marandah , A. Roussel-Jobinf , J. Latreillei , R. Laframboisej , L. Palmak , L. Kasprzakk , V.A. Marcusl , M. Breguetm , S. Noletn , Z. El-Haffafn , K. Australiei , A. Gologand , O. Aleynikovad , K. Oros-Kleino , C. Greenwoodo,p , A.M. Mes-Massonq , D. Provencherq , M. Tischkowitzr , G. Chongd , F. Rousseauf,s and W.D. Foulkesa,b,k,t a Department
of Human Genetics, McGill University, Montreal, Quebec, Canada, b Department of Medical Genetics, The Lady Davis Institute, Segal Cancer Centre, Jewish General Hospital, Montreal, Quebec, Canada, c Translational Research Laboratory, Catalan Institute of Oncology, Bellvitge Institute for Biomedical Research, Barcelona, Spain, d Department of Pathology, e Cancer Prevention Centre, Jewish General Hospital, Montreal, Quebec, Canada, f Unité de recherche en génétique humaine et moléculaire, Axe Santé des Populations et Pratiques Optimales en Santé, Centre de recherche du CHU de Québec, Montreal, Quebec, Canada, g Department of Medical Genetics, CHU of Quebec and Laval University, Quebec, Quebec, Canada, h Service of Genetics, Université de Sherbrooke, Sherbrooke, Quebec, Canada, i Faculté de médecine et des sciences de la santé, Université de Sherbrooke, CSSS Champlain Charles Le Moyne, Greenﬁeld Park, Quebec, Canada, j Génétique Médicale, CHU Laval, Quebec city, Quebec, Canada, k Department of Medical Genetics, Research Institute of the McGill University Health Centre, Montreal, Quebec,
Castellsagué et al. Canada, l Department of Pathology, McGill University, Montreal, Quebec, Canada, m Medecine Genique, n Medical Genetics Department, CHU de Montreal, Montreal, Quebec, Canada, o Lady Davis Institute, Jewish General Hospital, Montreal, Quebec, Canada, p Departments of Oncology, Epidemiology Biostatistics and Occupational Health, and Human Genetics, McGill University, Montreal, Quebec, Canada, q Centre de recherche du Centre Hospitalier de l’Université de Montréal, Institut du Cancer de Montréal, Montreal, Quebec, Canada, r Department of Medical Genetics, University of Cambridge, Cambridge, UK, s Département de biologie moléculaire, biochimie médicale et pathologie, Université Laval, Quebec City, Quebec, Canada, and t Program in Cancer Genetics, Department of Oncology and Human Genetics, McGill University, Montreal, Quebec, Canada Key words: colorectal cancer – endometrial cancer – founder mutations – French Canadians – Lynch syndrome – molecular diagnostics – MSH6 Corresponding author: Dr Ester Castellsagué, PhD, Department of Human Genetics, McGill University, Lady Davis Institute, Jewish General Hospital, 3755 Cote Ste Catherine Road, Montreal H3T 1E2, Quebec, Canada. Tel.: +1 (514) 340 8222x3100; fax: +1 (514) 340 8600; e-mail: [email protected]
Received 22 August 2014, revised and accepted for publication 8 October 2014
Lynch syndrome is an autosomal dominant cancer susceptibility syndrome caused by mutations and epimutations in the human DNA mismatch repair (MMR) genes MLH1, MSH2, MSH6 and PMS2. At least 3% of all colorectal cancer (CRC) cases are caused by Lynch syndrome (LS) and such cases are characterized by an early age of onset and an accelerated carcinogenesis. Additionally to CRC, LS is characterized by an increased risk of malignancies at certain extra-colonic sites such as the endometrium, ovary and stomach, among others (1). Whether breast cancer (BC) is part of the LS tumour spectrum is under debate (2). LS tumours require the inactivation of one of the key MMR genes usually by two ‘hits’, one inherited and one acquired later in life (3). Therefore, microsatellite instability (MSI) and immunohistochemical (IHC) staining are the routine tumour screening tests to identify those individuals that will most probably benefit from
germline mutational testing of MMR genes. Clinically, two major criteria have been used to select patients at risk of LS: Amsterdam criteria are more specific but less sensitive than Bethesda guidelines (1). Approximately two-thirds of the mutations in LS patients are distributed through MLH1 and MSH2, whereas the remaining one-third occurs in MSH6 and PMS2 (4). Although the proportion of MMR mutations varies between studies, MSH6 mutations are usually associated with a later age of cancer onset and an overall lower penetrance for LS-related tumours compared with MLH1 and MSH2 carriers (5–7). Importantly, endometrial cancer (EC) appears more frequently than CRC in women carrying MSH6 mutations (5, 8–10). A founder mutation is a mutation that occurred once in a ‘founder individual’ who passed it on to succeeding generations until the mutation become widespread, usually in a specific population. Importantly, a number
A novel MSH6 founder mutation in Quebec of founder mutations have been described in MMR genes (11) and, by testing them as a first step, it has been possible to lower the cost of molecular diagnosis in appropriate populations. To date, eight different MSH6 founder mutations have been described in the literature (12–16). The first to be identified was the c.650dupT Dutch founder mutation, detected in seven LS families (12). Two other Dutch MSH6 founder mutations (c.467C>G and c.1614_1615delinsAG) have been published so far (14). In Finland, two atypical LS families with predominance of endometrial carcinoma appeared to carry the same mutation (c.2983G>T) and haplotype analysis pointed to a common ancestral origin. The mutation was not found in 268 healthy controls and 245 CRC cases of Finnish origin (16). Cederquist et al. described two founder Swedish MSH6 mutations (c.1346T>C and c.2931C>G) occurring in two large pedigrees of late 17th and 18th centuries showing a late age of onset but a high lifetime risk of LS tumours, with a significant predominance of EC (13). Finally, two Ashkenazi Jewish (AJ) MSH6 founder mutations (c.3984_3987dup and c.3959_3962del) were evaluated in large series of controls (n = 3310), CRC (n = 2685) and EC cases (n = 337), and appeared to harbour higher risks to develop endometrial than colorectal malignancies. The mutations were calculated to arise around 585 Common Era (CE) and 685 CE, respectively, (15). Here we describe a novel MSH6 founder mutation (c.10C>T; p.Gln4*) predisposing to LS in the French Canadian (FC) population of Quebec. We have investigated 11 carrier families for genealogy, segregation of the mutation with LS tumours, MSI, IHC and loss of heterozygosity (LOH). We have assessed the frequency of this mutation in a large cohort of FC controls and in CRC and EC cases. Furthermore, we sought to assess its founder effect and calculate the age of the first MSH6 founder mutation described in the FC population of Quebec. Materials and methods Patients and samples
This study includes 11 LS families fulfilling Amsterdam or Bethesda criteria from several hospitals in the Canadian province of Quebec that tested positive for a truncating mutation in the MSH6 gene (c.10C>T). Where available, germline DNA and tumour samples were obtained from 11 probands and 27 additional family members. For the haplotype analysis, germline DNA from 179 FC controls recruited at the Jewish General Hospital (JGH) was used. In order to calculate the risk of the mutation to develop LS-related tumours we screened two different sets of unrelated FC samples: germline DNA from 381 EC cases (Notre-Dame Hospital); and formalin-fixed, paraffin-embedded (FFPE) DNA from normal tissue of 187 CRC patients (Centre Hospitalier Universitaire de Québec). To set the allelic frequency, germline DNA from 6433 FC anonymous-unlinked newborns was genotyped. Informed consent was obtained from participants and all analyses were approved by the appropriate ethics committee.
Molecular testing strategy
Familial samples were analysed in a routine manner in the molecular diagnostic units of the Pathology Departments of the JGH or Hotel-Dieu Hospital: germline DNA was tested for MSH6 mutations by High Resolution Melting (HRM) curve analysis and/or Sanger sequencing; IHC staining for MLH1, MSH2, MSH6 and PMS2 proteins, MSI analyses and LOH of the mutation were performed where tumoural material was available. FC controls and EC cases were genotyped by HRM. CRC cases were genotyped using LightCycler® 2.0 technology (Roche, Indianapolis, IN, USA). Germline DNA from the newborn controls were pooled (in 8-plex) and screened as described previously (17) by allele-specific-polymerase chain reaction (AS-PCR). Samples presenting abnormal patterns by HRM, LightCycler or AS-PCR were validated by Sanger sequencing. All primers, probes and specific conditions are available upon request. Haplotype analysis
Thirty-eight DNA samples from members of the 11 studied families were haplotyped using 14 microsatellite markers spanning 12.3 Mb upstream and 9.5 Mb downstream the mutation (Table 3; primers and conditions available upon request). One-hundred and seventy-nine FC controls were haplotyped using 10 of the microsatellite markers (Table 3). To deduce the mutation-associated haplotype, intra-familial segregation analysis was performed and alleles were phased when possible. Statistical analysis
Estimation of the mutation age was performed using dmle+ (18, 19), a Bayesian method for fine mapping of a mutation by using the linkage disequilibrium patterns. We assumed growth rates between 1.38 and 1.48-fold increase per generation. These values were derived by comparing the 1851 population of 892,061 with the 1961 FC population (5,259,211) and assuming between 4.4 and 5.5 generations during these 110 years (20 or 25 years per generation). Furthermore, we assumed that we sampled 24 of 30,809 of the mutations among FC. The numerator is based on the 23 individuals carrying the mutation (one was homozygous) and the denominator is based on the estimated allele frequency and the 2011 Quebec population. The association between MSH6 mutation and risk to develop CRC, EC and OC was estimated by calculating odds ratio (OR) according to Altman (Table 2) (20). Results Characterization of carrier families
All families presented with either Bethesda or Amsterdam clinical criteria for LS (Table 1, Table S1, Supporting Information). Notably, only two of the families fulfilled Amsterdam criteria, but all affected carriers were diagnosed with at least one tumour that has
Clinical criteria Proband Son Proband Sister Proband Nephew Proband Proband Sister Brother Proband Proband Mother Sister Sister
Son Son Proband Father Son Proband
FC-MSH6-1.2 FC-MSH6-2.1 FC-MSH6-2.2 FC-MSH6-3.1 FC-MSH6-3.3 FC-MSH6-4.1 FC-MSH6-5.1 FC-MSH6-5.2 FC-MSH6-5.3 FC-MSH6-6.1 FC-MSH6-7.1 FC-MSH6-7.2 FC-MSH6-8.1 FC-MSH6-8.2
FC-MSH6-9.2 FC-MSH6-9.3 FC-MSH6-10.1 FC-MSH6-10.2 FC-MSH6-10.3 FC-MSH6-11.1
M M M
F M F M F F F
M F M
M F F F
Carrier Carrier Carrierc
Carrier Carrier Carrier Homoz-Carrier Carrier Carrier Carrier
Carrier Carrier Carrier
Carrier Carrier Carrier Carrier
Status CRC CRC Unaffected EC BC Cervix EC Unaffected EC CRC SE EC Unaffected EC CRC Unaffected EC CRC EC TC Unaffected EC OC BC CRC Adr Unaffected Unaffected CRCd CRCd SCe NHL Unaffected CRC
Tumour 53 55 33 50 34 37 43 45 55 59 59 50 62 40 10 55 55 57 65 67 43 44 44 51 54 54 30 35 22 22 54 54 3 43
+ + + +
+ + + +
− − − −
+ + + +
No LOH NE
No LOH No LOH
Adr, adrenal tumour; BC, breast cancer; CC, cervical cancer; CHUL, Centre Hospitalier Université Laval; CHUQ, Centre Hospitalier Universitaire de Québec; CRC, colorectal cancer; EC, endometrial cancer; FAP, Familial adenomatous polyposis; HCLM, Hôpital Charles-LeMoyne; HD-CHUM, Hôtel-Dieu-Centre Hospitalier de l’Université de Montréal; IHC, immunohistochemical analysis in tumour tissue (+, expression −, lost expression); LOH, loss of heterozygosity; MGH, Montreal General Hospital; MSI, microsatellite instability; NE, not evaluable (PCR failed to amplify the DNA); NHL, non-Hodgkin lymphoma; OC, ovarian cancer; TC, thyroid cancer; SE, sebaceous epithelioma; SC, stomach cancer. a Because of homozygosity for the mutation in the germline, the tissue showed lack of staining both in neoplastic and non-neoplastic cells. b No germline MSH2 mutations found. c Carrier of an additional de novo APC mutation (c.3927_3931del, p.Glu1309Aspfs*4). d Synchronous CRC. e Not conﬁrmed by pathology report, conﬁrmed by family history.
Table 1. Clinicopathological features of c.10C>T MSH6 carriers
Castellsagué et al.
A novel MSH6 founder mutation in Quebec Table 2. Risk of CRC and EC conferred by c.10C>T MSH6 mutation Cases
1/187 (0.53%) 7/381 (1.84%)
16/6433 (0.25%) 16/6433 (0.25%)
0.4572 T, p.Gln4*) found in 11 LS families from the province of Quebec. We aimed to assess its clinical implications in the 11 affected families as well as in unselected CRC and EC cases from Quebec. We also confirmed its founder effect by finding that a common haplotype was present in all carriers among the 11 families. Additionally, we were able to calculate that the mutation appeared 430–656 years ago, probably coming with the first Quebec founders. All clinicopathological evidence pointed to the unequivocal pathogenicity of this mutation. In the carrier families we found an excess of mutation carriers among affected family members compared to unaffected (83% vs 40%, respectively). Significantly, a homozygous carrier of the mutation (FC-MSH6-7.1) developed early CRC at age 10, which is consistent with constitutional MMR deficiency (CMMR-D) (21). Other typical features of CMMR-D such as haematological or brain malignancies (21) were not detected in this patient. All available tumours from carrier individuals showed MSI and IHC loss of MSH6 protein (two of them showed additional loss of MSH2, but tested negative for mutations in this gene). Although LOH in the mutation position was not found in any of the studied tumours, the gene was probably inactivated by point mutations or deletions elsewhere in MSH6. We found that 65% of carriers were affected by a LS cancer (15/23), with a similar average age at diagnosis than the age at interview of the eight unaffected carriers, thus consistent with the incomplete penetrance characteristic of LS (22). Accordingly, we found 16 carriers of the mutation in a control population of 6433 newborns in Quebec (∼1/400). As far as we are aware, such a high prevalence of a MMR gene mutation has only been
284 284 262/286 278/286 286 276/280 262 286 288 298 286 280/286
The markers tested in a panel of French Canadian (FC) controls are shown in a box. The "/" symbol indicates that the phase of the disease haplotype cannot be established. Grey shading indicates non-recombinant haplotypes. The sizes of the minimum and maximum conserved haplotypes are shown at the bottom. Disease-associated alleles within the conserved haplotypes appear in bold. a Two phases for the homozygote carrier in family FC-MSH6-7.
132/140 132 136 140 136 132 136 136 136 132 136 136/144 276 266 270 266/268 270 274/278 284 270 294 284 270 270/284 270 264 264 262/270 264 266/272 264 264 264 264 264 264/270 225 225 225 225 225 227 225 225 225 225 225 225/241 166 218 166 218 166 218 166/174 202/218 166 218 166 202/218 166 218 166 218 166 218 166 218 166 218 166/172 214/218 Max 3.42 Mb Min 1.61 Mb Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 155 155 163 155/171 155 151/155 155 155 155 155 155 155 169 169 163/167 163 169 163/167 169 169 167 161/169 169 161/169 294 294 296 292/294 294 294/296 302 296 296 292/294 292 294/296 209 209 205 207/209 203 209/211 205 207 205 209 211 205/211 290 290 286/288 288/290 290 280/282 292 274 290 288 294 290/292 225 225 225 199/225 203 203/225 223 223 225 225 227 199/201
200 200 206 200/206 200 200/206 204/206 204/206 196 206 196 196/206
57582313 54209829 53224312 51288437 49622186 48131631 48010382 47866183 46411503 45178668 42604659 40896211 38026891 35709843
NC_000002.11 Family FC-MSH6-1 FC-MSH6-2 FC-MSH6-3 FC-MSH6-4 FC-MSH6-5 FC-MSH6-6 FC-MSH6-7a FC-MSH6-7a FC-MSH6-8 FC-MSH6-9 FC-MSH6-10 FC-MSH6-11
D2S2306 D2S2272 D2S177 D2S2186 D2S2374 Microsatellite
Table 3. Haplotypes associated with the c.10C>T MSH6 mutation
Castellsagué et al. reported in another founder mutation in PMS2 (∼1/399). However, no risk estimates to LS-associated cancers from unselected cases were provided in that study (23). Among the 11 studied families, a substantial increase in the presence of cancer was noticed among females compared to males (73% vs 27%), probably reflecting the fact that EC is the major cancer phenotype observed in female carriers of this mutation (8/11 affected carrier females developed an EC). This phenotype is a typical presenting feature of a MSH6 mutation in women, who are more commonly diagnosed with EC than CRC (5, 8–10). Consistent with this, by assessing the presence of the mutation in 381 and 187 unselected EC and CRC cases from Quebec, we found that the mutation confers a major risk to develop EC than CRC (OR = 7.5, p < 0.0001 vs OR = 2.2, p = 0.46). This would also explain why 88% of male carriers were unaffected. One of the affected carriers presented only with a BC (FC-MSH6-2.2), a tumour that does not definitely belong within the LS tumour spectrum (2). Unfortunately, no BC tissue was available. In another of the studied families (FC-MSH6-10) a pathogenic de novo mutation in APC gene [responsible for familial adenomatous polyposis (1)] was found in two MSH6 carrier individuals. Although the polyposis phenotype (>100 polyps and 2 synchronous CRC at age 22) may be dominant over the LS phenotype, we have not excluded this family from our study because MSI and loss of MSH6 staining were found in a CRC from one of the two carriers, indicating that MSH6 probably played a major role in the tumourigenesis process. We present here the 9th described MSH6 founder mutation, which, based on available data, is the most frequent founder mutation found in this gene and at the population level, it is probably one of the most prevalent MMR mutations ever reported in LS. The high prevalence among FC LS families of this mutation (∼50% of all MSH6 mutations in our database, data not shown) indicates that screening for c.10C>T should be the first step in the molecular diagnostic testing of MSH6-deficient FC tumours. We hope that the detailed clinical description of this mutation provided here will also facilitate the counselling of at-risk carriers. Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web-site.
Acknowledgements E. C. is a recipient of a Marie Curie International Outgoing Fellowship (PIOF-GA-2012-327193), co-sponsored with the Catalan Institute of Oncology-Bellvitge Institute for Biomedical Research, Barcelona, Spain. F. R. holds an MSSS/FQRS/CHUdeQuébec Research Chair in Technology Assessment and Evidence-based Laboratory Medicine. This project was also funded partly by the APOGÉE-Net/CanGèneTest Research and Knowledge Network in Genetic health Services, funded by the Canadian Institutes for Health Research, (grant number ETG-92250) (to F. R. and W. D. F.). Additional funding came from the Ministère de l’enseignement supérieur
A novel MSH6 founder mutation in Quebec de la recherche de la science et de la technologie (grant number PSR-SIIRI-846) (to W. D. F.).
References 1. Pineda M, Gonzalez S, Lazaro C, Blanco I, Capella G. Detection of genetic alterations in hereditary colorectal cancer screening. Mutat Res 2010: 693: 19–31. 2. Win AK, Lindor NM, Jenkins MA. Risk of breast cancer in Lynch syndrome: a systematic review. Breast Cancer Res 2013: 15: R27. 3. Foulkes WD. Inherited susceptibility to common cancers. N Engl J Med 2008: 359: 2143–2153. 4. Palomaki GE, McClain MR, Melillo S, Hampel HL, Thibodeau SN. EGAPP supplementary evidence review: DNA testing strategies aimed at reducing morbidity and mortality from Lynch syndrome. Genet Med 2009: 11: 42–65. 5. Hendriks YM, Wagner A, Morreau H et al. Cancer risk in hereditary nonpolyposis colorectal cancer due to MSH6 mutations: impact on counseling and surveillance. Gastroenterology 2004: 127: 17–25. 6. Plaschke J, Engel C, Kruger S et al. Lower incidence of colorectal cancer and later age of disease onset in 27 families with pathogenic MSH6 germline mutations compared with families with MLH1 or MSH2 mutations: the German Hereditary Nonpolyposis Colorectal Cancer Consortium. J Clin Oncol 2004: 22: 4486–4494. 7. Wagner A, Hendriks Y, Meijers-Heijboer EJ et al. Atypical HNPCC owing to MSH6 germline mutations: analysis of a large Dutch pedigree. J Med Genet 2001: 38: 318–322. 8. Baglietto L, Lindor NM, Dowty JG et al. Risks of Lynch syndrome cancers for MSH6 mutation carriers. J Natl Cancer Inst 2010: 102: 193–201. 9. Wijnen J, de Leeuw W, Vasen H et al. Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat Genet 1999: 23: 142–144. 10. Ramsoekh D, Wagner A, van Leerdam ME et al. Cancer risk in MLH1, MSH2 and MSH6 mutation carriers; different risk profiles may influence clinical management. Hered Cancer Clin Pract 2009: 7: 17.
11. Ponti G, Castellsagué E, Ruini C, Percesepe A, Tomasi A. Mismatch repair genes founder mutations and cancer susceptibility in Lynch syndrome. Clin Genet in press. 12. Berends MJ, Wu Y, Sijmons RH et al. Molecular and clinical characteristics of MSH6 variants: an analysis of 25 index carriers of a germline variant. Am J Hum Genet 2002: 70: 26–37. 13. Cederquist K, Emanuelsson M, Wiklund F, Golovleva I, Palmqvist R, Gronberg H. Two Swedish founder MSH6 mutations, one nonsense and one missense, conferring high cumulative risk of Lynch syndrome. Clin Genet 2005: 68: 533–541. 14. Ramsoekh D, Wagner A, van Leerdam ME et al. A high incidence of MSH6 mutations in Amsterdam criteria II-negative families tested in a diagnostic setting. Gut 2008: 57: 1539–1544. 15. Raskin L, Schwenter F, Freytsis M et al. Characterization of two Ashkenazi Jewish founder mutations in MSH6 gene causing Lynch syndrome. Clin Genet 2011: 79: 512–522. 16. Vahteristo P, Ojala S, Tamminen A et al. No MSH6 germline mutations in breast cancer families with colorectal and/or endometrial cancer. J Med Genet 2005: 42: e22. 17. Giroux S, Dube-Linteau A, Cardinal G et al. Assessment of the prevalence of the 985A>G MCAD mutation in the French-Canadian population using allele-specific PCR. Clin Genet 2007: 71: 569–575. 18. Rannala B, Reeve JP. High-resolution multipoint linkage-disequilibrium mapping in the context of a human genome sequence. Am J Hum Genet 2001: 69: 159–178. 19. Reeve JP, Rannala B. DMLE+: Bayesian linkage disequilibrium gene mapping. Bioinformatics 2002: 18: 894–895. 20. Altman DG. Practical statistics for medical research. London: Chapman and Hall, 1991. 21. Wimmer K, Kratz CP. Constitutional mismatch repair-deficiency syndrome. Haematologica 2010: 95: 699–701. 22. Stoffel E, Mukherjee B, Raymond VM et al. Calculation of risk of colorectal and endometrial cancer among patients with Lynch syndrome. Gastroenterology 2009: 137: 1621–1627. 23. Clendenning M, Senter L, Hampel H et al. A frame-shift mutation of PMS2 is a widespread cause of Lynch syndrome. J Med Genet 2008: 45: 340–345.