Does the occurrence of certain rare cancers indicate ... - Springer Link

Familial Cancer 2: 15–25, 2003.  2003 Kluwer Academic Publishers. Printed in the Netherlands.

Does the occurrence of certain rare cancers indicate an inherited cancer susceptibility? Sara Levene1, Gillian Scott1, Patricia Price2, Jeremy Sanderson3, Helen Evans4, Claire Taylor5, Sylvia Bass5, Cathryn Lewis6 and Shirley Hodgson6 1 Genetics Centre, Guy’s & St Thomas’s Hospitals Trust, Guy’s Hospital, London, UK; 2 Academic Unit of Clinical Radiation Oncology, Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Manchester, UK; 3 Department of Medicine, Guy’s, King’s & St Thomas’s School of Medicine, King’s College London, St Thomas’s Hospital, London, UK; 4 Thames Cancer Registry, London, UK; 5 Mutation Detection Facility, Cancer Research UK, St James’s University Hospital, Leeds, UK; 6 Division of Medical and Molecular Genetics, Guy’s, King’s & St Thomas’s School of Medicine, King’s College London, Guy’s Hospital, London, UK Received 29 July 2002; accepted in revised form 15 January 2003

Key words: BRCA1, BRCA2, cancer susceptibility, familial, HNPCC, rare cancers

Abstract We sought to determine whether rare cancers indicate an increased risk of inherited cancer susceptibility. We ascertained 77 individuals with rare cancers which occur with increased relative risk in carriers of germline BRCA1/BRCA2 (fallopian, young-onset pancreatic) or HNPCC (biliary, small intestinal, urothelial, gallbladder, young-onset pancreatic) mutations. Individuals with two primary neoplasms (7), or with a first- or two seconddegree relatives with breast/ovarian cancer were tested for BRCA1/BRCA2 mutations (18); those with two primary HNPCC cancers or one first degree relative with an HNPCC-related cancer were tested for mutations in MLH1/MSH2 (19). Of these 77 individuals with cancer (19 fallopian, 8 gallbladder, 17 biliary, 17 pancreatic, 11 urothelial, 5 small intestinal), 39 (50.6%) had at least one first degree relative with cancer (excluding lung and skin); two conformed to Bethesda HNPCC criteria. No definitely pathogenic germline MLH1 and MSH2 mutations were found in 19 individuals, although 2 MSH2 variants were detected. A family history of breast/ovarian, HNPCC or colon cancer in a first degree relative was found in 40% of fallopian, 20% of biliary, 35% of pancreatic, 27% of urothelial and 20% of small bowel cancer patients. A BRCA1 frameshift mutation was detected in a woman with fallopian (54 y) and breast (39 y) cancers, and a BRCA2 nonsense mutation in a woman with biliary (48 y) and breast (45 y) cancers. This study supports the premise that the occurrence of rare (especially double primary) cancers does indicate an increased cancer susceptibility, although the numbers of cases ascertained were too small to draw firm conclusions.

Introduction Germline mutations in the BRCA1 and BRCA2 genes can cause an inherited susceptibility to breast and ovarian cancer, with a penetrance of approximately 80% and 25–60% for breast and ovarian cancer, respectively, but they also confer smaller increased relative risks of certain other cancers, such as pancreatic, prostate, male breast (BRCA2) and fallopian tube cancer [1–4] (Table 1). Fallopian tube cancer is rare, but documented to occur in individuals with BRCA1 and BRCA2 mutations [5–8], and is histologically and behaviourally

similar to epithelial ovarian cancer. The development of fallopian cancer may therefore indicate an increased risk of a germline mutation in one of these genes in the affected individual. In hereditary non-polyposis colorectal cancer (HNPCC), caused by germline mutations in DNA mismatch repair genes [9], mutation carriers have a lifetime risk of 80% for colorectal cancer (CRC), higher in males than in females, and about 60% for endometrial cancer in females, but also increased risks of gastric, small intestinal, biliary tract, pancreatic and urothelial cancers. The estimated proportion of small

Correspondence to: Dr Shirley Hodgson, Division of Medical and Molecular Genetics, GKT School of Medicine, 8th Floor Guy’s Tower, Guy’s Hospital, London SE1 9RT, UK. Tel +44-207-9555000, x5647; Fax +44-207-9554644; E-mail [email protected]

16

S. Levene et al.

Table 1. Population incidences of cancers studied and relative risks in HNPCC or BRCA1 and BRCA2. Cancer site

Gall bladder Pancreas Urothelial Small intestine Stomach Ovary

1997 incidence in SE England (rate/100,000) [25] Men

Women

00.5 11.1

00.7 07.5

00.9 17.0 0–

00.6 06.7 16.0

RR in HNPCC

RR in BRCA1

9 [50] 7.6; 75 [50, 52] 25; > 100 [52, 53] 4.1; 6.9; 10.5 [49, 51, 52] 3.5; 13 [50, 53]

intestinal cancers that are caused by HNPCC ranges from 8 to 50%, and of urothelial cancers from 7 to 15% (see Table 1). The relative risks of small intestinal and pancreatic cancer are also increased in familial adenomatous polyposis (FAP) in which there is a very high risk of colorectal adenomas and cancer [10]. Since all these cancers are uncommon, the occurrence of such a tumour in an individual may indicate that they have an increased chance of having an inherited susceptibility to this and related cancers. Thus, a study of patients ascertained with small intestinal cancer from the Hawaii Tumour registry found that a substantial proportion of affected individuals had synchronous and metachronous cancers of the tumour spectrum of HNPCC (28% affected men), suggesting an inherited susceptibility to these cancers [11]. A study from the Thames Cancer Registry of individuals with two primary cancers, one of which was colorectal cancer, showed an excess of small intestinal cancer after colorectal cancer, and an excess of colorectal cancer after small intestinal cancer [12]. A family history of breast and pancreatic cancer is strongly predictive of the presence of germline mutations in BRCA1 and BRCA2, and breast cancer with a family history of fallopian tube or biliary tract cancer even more so [13]. Although the penetrance of these disorders is low, in terms of the development of these rare cancers, the relative risks are high because the population incidences are low. HNPCC was originally diagnosed on the basis of the presence of strict family history criteria known as the ‘Amsterdam’ criteria [14], which included only colorectal cancer cases. However, new criteria, the ‘modified Amsterdam criteria’, were subsequently introduced which include affected individuals with endometrial, small intestinal, ureteric or renal pelvis cancer [15, 16], in 3 relatives affected over 2 generations, and at least one diagnosis before 50 years of age. The ‘Bethesda’ criteria [17–19] were also developed, to give an indication as to the type of family history in which an HNPCC germline mutation was likely to be found in the presence of microsatellite instability in the tumour of an affected individual [20, 21]. These criteria included cancers of the endometrium, ovary, stomach, hepato-biliary tract, small intestinal and urothelium as significant. One of the Bethesda criteria used is the

RR in BRCA2

5 [51] 3.5 [51]

5.9 [54] 15.5; 29 [54, 55]

2.6 [51] 5.2; 18 [54, 55]

presence of two primary cancers from the HNPCC spectrum in a single individual, regardless of family history (Table 2). Germline mutations in BRCA2 were detected in 7% of an unselected small series of pancreatic cancer cases [22] and in 4.5% of cases of fallopian tube cancer [6]. In addition, a recent study [23] reported BRCA2 mutations in 8% (3/38) of patients with pancreatic adenocarcinoma who were considered to be at high or intermediate risk of familial predisposition on the basis of their family history (3% of the total sample of pancreatic cancer cases). However, there are no systematic studies of inherited aspects of rare cancers, and it is not known what proportion of individuals with such rare cancers have an inherited cancer susceptibility. We therefore set out to ascertain unselected cases of these rare cancers and to determine whether they had a stronger family history of cancers than would be expected, ie whether they were predictive of the likelihood of the presence of HNPCC or a mutation in BRCA1/BRCA2 and whether molecular analysis of a subset of these cases demonstrated the presence of such germline mutations. We chose to ascertain individuals affected with the relatively rare cancers known to occur at increased relative risk in HNPCC (biliary tract, small intestinal, urothelial and gallbladder cancer). Individuals with pancreatic cancer diagnosed before the age of 50 years were also included, since early onset is characteristic of cancers developing in individuals with an inherited cancer susceptibility [24], and only around 3% of pancreatic cancers are diagnosed before the age of 50 years [25]. In addition, we included cancers occurring with increased relative risk in germline carriers of BRCA1 and BRCA2 mutations (pancreatic cancer diagnosed below the age of 50 years, and fallopian tube cancer).

Methods Two main methods were used to ascertain individuals eligible for the study. Firstly, the Thames Cancer Registry (TCR) regularly provided a list of details for patients diagnosed with one of the specific rare cancers being studied, during the period 1997–1999. The

Do rare cancers indicate an inherited cancer susceptibility?

17

Table 2. Amsterdam and Bethesda criteria for diagnosis of HNPCC. Amsterdam criteria [14] At • • • • •

least three relatives with CRC plus all the following: One of the affected relatives is a first degree relative of the other two Two or more successive generations affected One or more affected relatives diagnosed below 50 years Familial adenomatous polyposis excluded Tumours verified by pathological examination

‘Modified’ Amsterdam criteria [56] Three or more relatives with HNPCC-associated cancer (CRC or cancer of the endometrium, small intestine, ureter or renal pelvis) plus all the following: • One of the affected relatives is a first degree relative of the other two • Two or more successive generations affected • One or more affected relatives diagnosed below 50 years • Familial adenomatous polyposis excluded in cases of CRC • Tumours verified by pathological examination Bethesda guidelines [17] Classification of patients with colorectal tumours who should be tested for microsatellite instability: B1 Individuals with cancer who meet the Amsterdam criteria B2 Individuals with two HNPCC-related tumours, including synchronous and metachronous CRC or associated extracolonic cancer (endometrial, ovarian, gastric, hepatobiliary or small intestinal cancer; transitional cell carcinoma of the renal pelvis or ureter). B3 Individuals with CRC and a first degree relative with CRC, HNPCC-related extracolonic cancer or a colorectal adenoma, one of the cancers being diagnosed below 45 years, or the adenoma diagnosed below 40 years. B4 Individuals diagnosed below 45 years with CRC or endometrial cancer. B5 Individuals diagnosed below 45 years with right-sided CRC whose histopathology shows an undifferentiated pattern (solid or cribri form, defined as poorly differentiated or undifferentiated carcinoma composed of irregular, solid sheets of large eosinophilic cells and containing small gland-like spaces). B6 Individuals diagnosed below 45 years with signet ring cell type CRC (more than 50% signet ring cells). B7 Individuals diagnosed below 40 years with colorectal adenomas. CRC – colorectal cancer; HNPCC – hereditary nonpolyposis colorectal cancer.

researchers then contacted the GP or hospital doctor listed by the TCR to confirm that it would be appropriate to contact the patient. Secondly, contact was made and maintained with around 80 surgeons, oncologists and geneticists. These doctors were sent a monthly reminder about the aims of the study, and asked to inform the researchers of any patients they had seen in the preceding month who would be suitable for the study. After the clinician had given permission to contact the patient, a letter and information leaflet were sent inviting the patient to participate in the study. If the patient indicated that they were interested in participating, they were contacted by a researcher to arrange a meeting. At the meeting, a detailed family history was taken and if the patient had a family history of other cancers he or she was given the opportunity to give a small blood sample for DNA extraction. Confirmation of cancers in relatives was obtained wherever possible (in most cases). Informed consent was obtained for all the blood samples taken and, as specified by the MREC approval for the study, it was explained to the patients that all blood samples would be anonymised and no personal results would be available from the research. Patients were given the opportunity to receive general results of the study that might be relevant to them, and they could also nominate a next-of-kin to whom this infor-

mation could be sent if they were already deceased by the end of the study. Patients whose family histories conformed to the following criteria had blood samples taken for anonymous testing for both BRCA1 and BRCA2 mutations (in breast cancer families) or for MSH2 or MLH1 mutations in families with a history of HNPCC-associated cancers. Individuals with two primary cancers including breast or ovarian, or with a first degree relative or two second degree paternal relatives with breast or ovarian cancer were tested for BRCA1 and BRCA2 mutations. HNPCC testing (for mutations in MLH1 and MSH2) was performed in families with two primary HNPCC spectrum cancers, or one first degree relative with an HNPCC-related cancer (such as uterine, colorectal, young onset pancreatic, as defined by the Bethesda criteria) (Table 3). Mutation analysis The screening methods used and the regions of the respective genes examined are shown in Table 4. All DNA sequence changes detected by mutation screening were confirmed by direct DNA sequencing. PCR PCR amplification was carried out using Amplitaq Gold polymerase (Applied Biosystems) in the manufacturer’s

18

S. Levene et al.

Table 4. Screening methods used and regions examined for the genes studied.

rate of 0.9 ml/min and buffer B gradient increase of 2%/min for 4 min. Melting temperatures for DHPLC analysis were determined by running a wild type sample at increasing temperatures until a reduction in retention time of 1 min was observed (details available from [email protected]). PCR products were prepared for DHPLC by denaturing at 95 °C for 5 min and then cooling to 65 °C to allow heteroduplex formation. Mutation detection was by visual inspection of chromatograms. The mutation detection sensitivity of these assays is estimated to be approximately 95% [33, 34].

Gene

Exons

Screening method

Protein truncation test (PTT) analysis

MSH2

1–16

FSSCP and/or DHPLC

MLH1

1–11, 13–19, 12 (3′ end) 12 (5′ end)

FSSCP and/or DHPLC Direct DNA sequencing

PTT utilised the TnT T7 Quick Coupled Transcription/ Translation System (Promega) according to the manufacturer’s protocol, but scaled down to 5 µl. 35S-labelled, in vitro synthesised products were separated by discontinuous SDS-PAGE on 12.5% acrylamide gels [35] and detected by autoradiography. This has high sensitivity [36].

Table 3. Cases tested for mutations. Total tested

MLH1, MSH2

BRCA1/ BRCA2

Fallopian tube Biliary tract and gallbladder Pancreas Urothelial Small intestine

09 05 08 01 02

06 05 05 01 02

08 03 07 00 00

Total

25

19

18

BRCA1 2–10, 12–24, 11 (5′ and 3′ ends) DHPLC 11 PTT BRCA2 2–10, 12–27, 11 (5′ and 3′ ends) DHPLC 11 PTT and/or DHPLC

buffer; 2.5–3.0 mM MgCl2; 200 µM each dNTP and 0.5 pmol/µl each primer. For DNA sequencing, PTT and DHPLC, PCR primers were unlabelled; for FSSCP both primers were 5′ labelled with FAM, HEX or NED. PCR primers were mostly as described previously [26–31] or are listed in Table 5 (details available from [email protected]). PCR was checked for yield and specificity by agarose gel electrophoresis [32]. Fluorescent single strand conformation polymorphism (FSSCP) analysis Fluorescent PCR products were diluted with water (1 in 10 to 1 in 40, depending on yield) and 1–2 µl of diluted product mixed with 0.5 µl of ROX-500 size standards (Applied Biosystems) and 10.5 µl of HiDi Formamide (Applied Biosystems). Samples were denatured at 95 °C for 2 min and snap-cooled on ice. FSSCP analysis utilised a 3100 Genetic Analyser (Applied Biosystems) at 18 °C and 30 °C using 5% Genescan polymer containing 10% glycerol and 1×TBE. Data were analysed using Genescan 3.7.1 and Genotyper 2.5 software (Applied Biosystems); mutation detection was by visual inspection of electropherogram traces.

DNA sequencing Unincorporated primers and dNTPs were removed from PCR products using shrimp alkaline phosphatase and exonuclease I (Amersham-Pharmacia). PCR primers were used for all sequencing reactions apart from MSH2 exons 2 and 5, where alternative primers were used (Table 5). Sequencing reactions utilised the ABI PRISM BigDye Terminator Cycle Sequencing Kit versions 1 and 3 (Applied Biosystems) with electrophoresis of the products on 310 and 3100 Genetic Analysers (Applied Biosystems). Data analysis was carried out using Sequence Analysis 3.0 (Applied Biosystems) and SeqMan (DNAStar) software and by visual inspection of electropherograms.

Statistical methods The small sample sizes for each cancer site preclude detailed statistical testing. For each site, the numbers of first and second degree relatives affected with cancer were compared using Fisher’s exact test. Differences in age of diagnosis by family history were tested using the non-parametric Wilcoxon rank sum test.

Denaturing high performance liquid chromatography (DHPLC) analysis

Results

DHPLC utilised a Transgenomic WAVE Nucleic Acid Fragment Analysis system and DNASep column (Transgenomic, Crewe, UK), buffer A 0.1 M triethylammonium acetate (TEAA) and buffer B 0.1 M TEAA and 25% acetonitrile. Analysis utilised a flow

We ascertained 83 cases of cancer comprising 19 cases of fallopian cancer, 8 of gallbladder and 17 of biliary tract cancer (considered together), 17 of pancreatic cancer diagnosed before the age of 50 years, 11 of urothelial and 11 of small intestinal cancer. Data on the


19

Table 5. PCR and sequencing primers designed for this study. Gene

Exon

Primer

Sequence

BRCA1

06 06 07 09 09 10 10 11 11 11 11 17 17 18 18 23 23 24 24

BRCA1 ex6F BRCA1 ex6R BRCA1 7R BRCA1 ex9F BRCA1 ex9R BRCA1 ex10F BRCA1 ex10R BRCA1 ex11 5′F BRCA1 ex115′R BRCA1 ex113′F BRCA1 ex113′R BRCA1 ex17F BRCA1 ex17R BRCA1ex18F BRCA1ex18R BRCA1ex23F BRCA1ex23R BRCA1ex24F BRCA1ex24R

TCACTTGCTGAGTGTGTTTCTC TCCTGAGTTTTCATGGACAGC GAAGAAGAAGAAAACAAATGG TTGATTTATTTTTTGGGGGG TGCACATACATCCCTGAACC CTTGGTCATTTGACAGTTCTGC TGGGTTGTAAAGGTCCCAAA GTTGATTTCCACCTCCAAGG AATAAACTGCTGTTCTCATGCTG TGCAAATACAAACACCCAGG TGCTCCCCAAAAGCATAAAC ATTCTGAGCTGTGTGCTAGAGG TCGCCTCATGTGGTTTTATG GGACAGCACTTCCTGATTTTG TGTTAAAGGGAGGAGGGGAG TGGGTGACAGAGCAAGACC TGTGCTACTCAAGCACCAGG AGGACCCTGGAGTCGATTG AAGAGTGAGAGGAGCTCCCAG

BRCA2

10 10 10 10 10 10 11 11 11 11 11

BRCA2 BRCA2 BRCA2 BRCA2 BRCA2 BRCA2 BRCA2 BRCA2 BRCA2 BRCA2 BRCA2

ACTGTTTCTATGAGAAAGGTTGTG CCATCTGGGCTCCATTTAG CTGTGAATGGTCTCAACTAACC ATGTATTTCCAGTCCACTTTCAG CACCTAAAGAGACTTTCAATGC AACACAGAAGGAATCGTCATC GAAACAGCAAAAAGTCCTGC AGTATCTTGTTTTTCGGAGAG CTTGGAAAATAACATCTGAGGG GGGAAAAGAACAGGCTTCAC TGTTAGCATACCAAGTCTACTGAATAA

MSH2

01 01 01 02 05

MSH2 MSH2 MSH2 MSH2 MSH2

MLH1

02 02 03 03 04 04 05 05 06 06 07 07 08 08 09 09 10 10 11 11 12 12 12 12 13 13 14 14 15

MLH1 ex2F MLH1 ex2R MLH1 ex3F MLH1 ex3R MLH1 ex4F MLH1 ex4R MLH1 ex5F MLH1ex5R MLH1 ex6F MLH1 ex6R MLH1ex7F MLH1 ex7R MLH1 ex8F MLH1 ex8R MLH1 ex9F MLH1 ex9R MLH1 ex10F MLH1 ex10R MLH1 ex11F MLH1 ex11R MLH1 12AF MLH1 12AR MLH1 ex12bF MLH1 ex12bR MLH1 ex13F MLH1 ex13R MLH1 ex14F MLH1 ex14R MLH1 ex15F

10 5′F 10 5′R 10 mid F 10 mid R 10 3′F 10 3′R 11DbF 11DaR 11HR ex11 3′F ex11 3′R

ex1F ex1R ex1seqF 2seqF 5seqR

GGTGTGGTCGCCGTGG ACTCTCTGAGGCGGGAAAGG GAGAAGCCGACCACCACAGT TTTAAGGAGCAAAGAATCTG ACCTGAAAAAGGTTAAGGGC TTGTTATCATTGCTTGGCTC CCATCTGCAAAAGCCTAGTT GGAATTCAAAGAGATTTGG GACAATGTCATCACAGGAGG TAACCTTTCCCTTTGGTGAG ACACTGGTGTTGAGACAGGA TTCTCTTTTCCCCTTGGGAT AGGATATCTTGGGACCTCC TTTTCAAGTACTTCTATGAATTTACAA TTTCACCATCTAGCTCAGCA TGTGTGTTTTTGGCAACTCT CATAAAACAAAACCATCCCC CAGCCATGAGACAATAAATCC TCTGAAGCATAAAACAAGCCT GCCCTCATTTCACAAAGTTAG GTCCCATAAAATTCCCTGTG CCCTCAGGACAGTTTTGAAC CTTTCAAAGAGGAGAGCCTG CCCCTCCCACTATCTAAGGT CAAAGGCCCCAGAGAAGTA ACTTCTTATTCTGAGTCTCTCCAC AAATGCATCAAGCTTCTGTT GGAGATGGTTAAATCCACAA GAGGTAGGCTGTACTTTTCCCAA ACCCACAAAATTTGGCTAAG CAACATGACTGCTTTCTCCA CTCTAGTTCTGGTGCCTGGT GCTCTCCTTAGCTTTTGTGC GCATGAATTCAGCTTTTCCT

20

S. Levene et al.

Table 5. Continued. Gene

Exon

Primer

MLH1 (Continued)

15 16 16 17 17 18 18

MLH1 MLH1 MLH1 MLH1 MLH1 MLH1 MLH1

Sequence ex15R ex16F ex16R ex17F ex17R ex18F ex18R

CCTCCATATGCAAATCATACA ATTCAGGCTTCATTTGGATG CCGGCTGGAAATTTTATTTG TGGAGAAATGGGATTTGTTT CAGATCAAAGGGTGGTCATT CTGTGATCTCCGTTTAGAATGA TCCCGAAATTTTAGAGATGG

Primers were designed using Primer3 or by visual inspection of DNA sequence. Accession numbers for sequences used for primer design were as follows: BRCA1 L78833 and AR004688; BRCA2 U43746 and X95160; MSH2 U41206, U41207 and U41210; MLH1 AC006583 and AC011816.

numbers of first and second degree relatives affected with cancer and the types of cancer diagnosed are detailed in Table 6. We excluded six of the cases of small intestinal cancer from the study, as further investigation revealed that they were of neuroendocrine origin, and not adenocarcinomas. Neuroendocrine small intestinal cancers are thought to have a different aetiology to carcinomas [37]. Family history data are given in Table 7. The number of first and second degree relatives with any cancer, excluding lung and skin cancers (because of their strong environmental aetiology) and unknown diagnoses, per case was 38/19 (2 per case) for fallopian, 21/25 (0.84 per case) for biliary/gall bladder, 36/17 (2.1 per case) for pancreatic, 17/11 (1.5 per case) for urothelial and

6/5 (1.2 per case) for small intestinal cancer (Table 6). 14/19 (74%) of fallopian, 6/25 (24%) of biliary tract, 9/17 (53%) of pancreatic cancer cases diagnosed below 50 years, 6/11 (55%) of urothelial and 4/5 (80%) of small intestinal cases had a family history of cancer in one or more first degree relatives. In addition, five of the cases in the current study had two first degree relatives with cancer (6.5%). Seven individuals (9.1%) had more than one primary tumour (Table 8). None of the family histories conformed to the standard ‘Amsterdam criteria’ for the diagnosis of HNPCC. However, some did comply with the Bethesda (see Bethesda 2 criteria), or the modified Amsterdam criteria (Table 2). We defined families with a family history of one colorectal cancer (CRC)

Table 6. Numbers of first and second degree relatives with cancer. Type/site of cancer in 1° or 2° relative

Fallopian (n = 19)

Biliary tract/ gall bladder (n = 25)

Pancreas diagnosed below 50 years (n = 17)

Urothelial

Small intestine

(n = 11)

(n = 5)

1°

2°

1°

2°

1°

2°

1°

2°

1°

2°

Breast Ovary Colorectal Pancreas Stomach Testis Prostate Kidney Uterus Cervix Liver Lymphoma Oesophagus Thyroid Brain Leukaemia Melanoma Intestine Throat Tongue Bronchus Bladder Spine

03 01 04 0– 01 0– 03 0– 02 01 0– 01 0– 01 01 01 0– 0– 0– 0– 0– 0– 0–

05 01 03 01 03 0– 01 0– 02 01 0– 01 0– 0– 0– 0– 0– 0– 01 0– 0– 0– 0–

– – 1 – 3 1 1 – 1 – 1 – – – – – – – – – – – –

2 – 2 2 – – – 1 – – – – 2 – – – 1 – – – – – –

02 0– 02 01 03 0– 0– 01 01 0– 0– 0– 01 01 0– 0– 0– 0– 0– 0– 0– 0– 0–

04 0– 01 01 05 0– 02 0– 02 0– 0– 01 0– 0– 01 03 0– 01 02 01 0– 0– 0–

02 0– 01 0– 01 0– 01 0– 0– 0– 01 0– 0– 0– 02 01 0– 0– 01 0– 01 0– 0–

3 1 – – 1 – 1 – – – 1 – – – – – – – – – – – –

– – – – – – 1 – 1 – – – – – – 1 – – – – – – –

2 – – 1 – – – – – – – – – – – – – – – – – 1 1

Total

19

19

8

9

12

24

11

7

3

5


21

Table 7. Characteristics of family histories.

Total no. families No. families with first degree relatives affected Total first degree and second degree relatives excluding lung, skin and unknown Proband with two primaries Modified Amsterdam Bethesda (modified) Bethesda-like BRCA BRCA-like

Fallopian tube

Biliary tract and gallbladder

Pancreas

Urothelial

Small intestine

19 14

25 06

17 09

11 06

5 4

38 03 01 02 02 01 02

21 04

36 00

17 00

6 0

01 00 00 01

01 00 03 00

00 01 01 00

0 1 0 0

diagnosed at any age and one other HNPCC spectrum cancer (endometrial, pancreatic, gastric, ovary) conforming to the Bethesda criteria as ‘Bethesda positive’, and those with a family history of two cancers from the HNPCC spectrum but not conforming to the Bethesda criteria as ‘Bethesda like’. One family (fallopian cancer) conformed to the modified Amsterdam criteria for HNPCC, and four (one pancreatic, one biliary tract, two fallopian) to the Bethesda criteria (Table 5). Individuals with a family history of two breast or ovarian cancers diagnosed at any age in first or second degree relatives, two paternal second degree relatives with breast cancer, one first degree relative with breast cancer diagnosed below the age of 60 years, or one first degree relative with ovarian cancer, were defined as ‘breast/ovarian cancer families’. Three families with pancreatic cancer, one with fallopian cancer and one with urothelial cancer conformed to these criteria (Table 7). The age at diagnosis was tested for correlation with family history in fallopian and pancreatic cancer probands as these had the most extensive family history. Age at diagnosis was significantly lower in individuals with fallopian tube cancer with a family history of breast cancer in a first or second degree relative (n = 7) than in those without (n = 12) (median 45 years compared with 63 years, P = 0.006). In individuals with pancreatic cancer, no significant difference was seen between the ages at diagnosis of those cases with and without a family history of either breast or colorectal cancer (13 families without a family history of breast

cancer, mean age at diagnosis 48.5 years, 4 families with a family history, mean age at diagnosis 45 years, P = 0.65). Twelve families with no family history of CRC had a mean age at diagnosis of pancreatic cancer of 44.8 years, compared with 43.2 years for the 5 families with a family history of CRC (P = 0.63). Results of mutation screening Mutation screening was performed on genomic DNA from 25 individuals (Table 3). No known pathogenic germline mutations in MLH1 and MSH2 were detected in 19 individuals tested. However, a hitherto undescribed MSH2 missense variant, L414P, was detected in a man diagnosed aged 33 years with duodenal cancer, whose mother was diagnosed with uterine cancer at age 42 years and a retroperitoneal lymph node tumour at 49 years, but no other affected first or second degree relatives. This mutation is caused by a T>C transition at coding nucleotide 1241; its pathogenic significance is unknown. A second variant in MSH2 was detected in a man diagnosed with two primary cancers (prostate cancer diagnosed at 71 years and biliary tract cancer diagnosed at 78 years). He had five brothers and sisters and one child, but none of his first degree relatives had been diagnosed with cancer, and his parents died at 71 years and 86 years of age. The mutation is an A>G transition at coding nucleotide 380, which results in the missense mutation N127S, a conservative change considered to be of doubtful pathogenicity [38].

Table 8. Individuals diagnosed with two cancers. First cancer (age diagnosed)

Second cancer (age diagnosed)

Mutation found

Breast (45 y) Breast (39 y) Gallbladder (52 y) Gallbladder (74 y) Fallopian (45 y) Prostate (71 y) Fallopian (69 y)

Biliary (48 y) Fallopian (54 y) Uterine (55 y) Ovarian (78 y) Caecum (67 y) Biliary (78 y) Neuro-endocrine lung (69 y)

BRCA2 9382C>T 0 BRCA1 1823-826delAGAA 0 0 MLH1 1852A>G, 1853A>C In 80 y (type ?) Cervix (∆ 50s) MSH2 380A>G 0 Gynaecological cancer (∆ 33 y); ovarian cancer (∆ 72 y)

∆ = age at diagnosis.

Cancer family history in first degree relatives

Cancer family history in second degree relatives 0 Stomach (∆ 70s) Pancreas (∆ 74 y) 0 0 0 0

22

S. Levene et al.

A variant in MLH1, a K618A missense change causing 1852A>G at the amino acid level (see Table 9) was detected in a woman who was diagnosed with gallbladder cancer at 74 years and ovarian cancer at 78 years, and had no family history of cancer in first degree relatives. The pathogenic status of this polymorphism is not clear. One BRCA1 mutation (frameshift 1823→1826 del AGAA in exon 11) was detected in a woman with both fallopian and breast cancers and no family history in first degree relatives, and one BRCA2 mutation (R3182X in exon 25, caused by the nucleotide change 9382C>T, was detected in a woman with both biliary tract and breast cancer. She had two second degree relatives with breast cancer, bilateral ovarian cancer in a cousin, and a further four first and second degree relatives with cancer. Both BRCA1/BRCA2 mutations have been reported multiple times on the BIC database [39]. No other mutations were detected in BRCA1, BRCA2, MLH1 or MSH2 in our series. Variants detected in all four genes are shown in Tables 9 and 10.

Discussion We have set out to determine whether there is any evidence to suggest that the occurrence of a rare cancer, which occurs with increased relative risk in certain cancer predisposition syndromes, is a predictor of germline mutation in such a gene. We compared the occurrence of cancer in first degree relatives of cases with that found in controls in other studies. The Collaborative Group on Hormonal Factors in Breast Cancer [40] found that 7% of controls reported a first degree relative with breast cancer (4.4% had a mother with breast cancer, and 3.1% one or more affected sisters) compared with 13% of cases (7.9% a mother or 5.6% a sister affected). The number of first degree relatives with breast cancer was not significantly raised in any of our cancer sites. In fallopian cancer, 3/19 (16%) cases had a first degree relative with breast cancer (P = 0.1). Similar studies of controls have shown that 6–8% report a family history of CRC [41, 42]. None of the rare cancer sites in the present study showed a frequency of CRC in first degree relatives in excess of these figures (4/19 fallopian cases, 2/17 pancreatic cases).

Table 9. Rare or novel variants detected in HNPCC genes. Gene

Sample

Variant

MLH1

Numerous

IVS5+79A>G

24

1852A>G, 1853A>C

74

1959G>T

89

IVS11-8T>A

05

1241T>C

05

IVS8-8G>T

05

1737A>G

11

380A>G

MSH2

Effect on protein sequence

Pathological status

Comment

References

Polymorphism

Heterozygosity 48% in this study

This study

K618A missense

Not clear; reported in literature and in ICG database variously as pathogenic and nonpathogenic

Protein displays reduced binding to PMS2. Over-represented in colorectal cancer patients

[57–61]

L653L silent

Polymorphism Present in 9/93 control DNA samples

ICG-HNPCC database

Polymorphism

L414P missense

[62]

Not known

This study

Polymorphism

This study

K579K silent

Polymorphism

[63]

N127S missense

Not known; available evidence more suggestive of polymorphism

[64]

Table 10. Rare or novel variants detected in BRCA genes. Gene

Sample

Variant

Effect on protein

Pathological status sequence

Reference

BRCA2

71

1151C>T

S384F missense

Presumed polymorphism

BIC

87

IVS2+62T>G

89

2228T>C

89

IVS18+16C>G

19

5744C>T

19

8657A>C

Polymorphism

BIC

Polymorphism

BIC

Polymorphism

This study

T1915M missense

Polymorphism

[65]

E2856A missense

Presumed polymorphism

BIC

H743H silent

BIC – Breast Cancer Information Core Mutation Database: http://www.nhgri.nih.gov/Intramural_research/Lab\-transfer/Bic).

Do rare cancers indicate an inherited cancer susceptibility? Our small sample size within each rare cancer site precludes testing for family history of most cancers; however, we pooled all cancer sites to obtain the frequency of first degree relatives with cancer. Parents and siblings were included in this calculation, but offspring were excluded due to their young age in most pedigrees. There was no statistically significant difference between sites in the proportion of first degree relatives affected with cancer, which varied from 23% for relatives of fallopian and pancreatic cases to 14% for relatives of biliary tract and gall bladder cancer. The age at diagnosis of cancer in individuals with an inherited cancer susceptibility is characteristically younger than for sporadic cancers. It is therefore of interest that the cases of fallopian cancer with a family history of breast cancer had a significantly younger age at diagnosis than those without such a family history. This is additional evidence in support of the suggestion that a proportion of such cancers may arise in individuals with an hereditary cancer susceptibility. The fact that cases of pancreatic cancer did not demonstrate an earlier age at diagnosis in cases in the context of a family history of colorectal or breast cancer may reflect the fact that we were already selecting for early age at diagnosis and did not include the majority of pancreatic cancers, diagnosed after the age of 50 (3% of pancreatic cancers are diagnosed before the age of 50). There were too few individuals from the other cancer groups to test age of diagnosis. From within this cohort of patients, we ascertained 6 individuals with 2 primary cancers and one with a subsequent lung tumour (7/77 cases) (9.1%), of whom two had germline mutations in BRCA1 or BRCA2, and two had germline variants in MSH2 or MLH1 (see Table 8). This is well in excess of the approximately 5% of all cases of cancer recorded as having more than one separate primary cancer (excluding basal cell carcinoma of the skin) [12]. Breast cancer occurred at a very young age in the fallopian cancer case, who had a germline BRCA1 mutation but no family history of breast/ovarian cancer. A BRCA2 mutation was detected in another individual with two primary cancers (biliary tract and breast; both diagnosed below 50 years) who had a family history of breast cancer in second and third degree relatives. The fact that both the known pathogenic germline mutations in BRCA1 and BRCA2 were in individuals with two primary cancers is of interest. In addition, one of the variants in MSH2, and the K618A MLH1 variant, were detected in individuals with two primary cancers. However, the pathogenicity of these variants could not be confirmed with certainty because the nature of the ethical consent for the study precluded going back to the families for further family studies, or tests on the tumours. We have previously shown [43, 44] that individuals with two primary cancers have an earlier age at diagnosis of the first cancer than the general population and that there is familial clustering of cancer type, indicating that individuals with two primary cancers have an increased

23

chance of having an inherited cancer predisposition. An increased risk of germline BRCA1/2 mutations has been found in women with more than one cancer, including one breast cancer [45, 46], and there is evidence for an increased risk of HNPCC in individuals with both endometrial and colorectal cancer [47]. Epidemiological studies [48, 49] support the contention that individuals with two primary cancers have increased genetic risks. Previous studies have demonstrated an increased frequency of mutations, suggesting an increased likelihood of BRCA1/2 mutations in fallopian cancer cases [7, 8]. In this study, we did not analyse other candidate genes such as PTEN, p16, ATM or p53, since the yield of mutations was likely to be small, but this may have led to an underestimation of the proportion of cases due to germline cancer-predisposing mutations. The small numbers of cancers ascertained for individual sites were too small for firm conclusions to be drawn. However, the fact that 40% of fallopian cancers, 20% of biliary tract, 35% of pancreatic, 27% of urothelial, and 20% of small bowel cancers had a family history of HNPCC, breast/ovarian cancer or of CRC in a first degree relative suggests that a substantial proportion of such cancers arise in individuals with an inherited cancer susceptibility, and that the clinician should be alerted to this possibility. The proportion of cases with a family history suggesting the presence of familial breast cancer or of HNPCC is higher than would be expected in an unselected group of cases of breast or gastrointestinal cancer, where the proportion would be expected to be of the order of 2–5%. However, these estimates are approximate and our sample is too small to draw any definitive conclusion. In conclusion, this small study has demonstrated an increased family history of cancer in first degree relatives and a young age at diagnosis in individuals with rare cancers from the HNPCC or BRCA1/2 spectrum, indicating that the occurrence of such rare cancers in an individual should alert clinicians to the possibility that they might carry a germline mutation in a cancerpredisposing gene, although numbers are too small for firm conclusions to be drawn. In our sample, the germline mutations detected in BRCA1 and BRCA2 were in individuals who had developed two primary cancers.

Acknowledgements We are most grateful to Professor Peter Harper, Dr Alison Jones and Dr Mark Wilkinson for referring patients for the study, to Dr Jan Bell from the Thames Cancer Registry, and to Jan Ball, Adrienne Knight and Elizabeth Manners for secretarial assistance. This work was funded by the Guy’s and St Thomas’s Hospitals Charitable Foundation.

24 References 01. Durocher F, Tonin P, Shattuck-Eidens D et al. Mutation analysis of the BRCA1 gene in 23 families with cases of cancer of the breast, ovary, and multiple other sites. J Med Genet 1996; 33: 814–9. 02. Easton DF, Ford D, Bishop DT. Breast and ovarian cancer incidence in BRCA1 mutation carriers. Breast Cancer Linkage Consortium. Am J Hum Genet 1995; 56: 265–71. 03. Thompson D, Easton DF, the Breast Cancer Linkage Consortium. Cancer incidence in BRCA1 mutation carriers. J Natl Cancer Inst 2002; 94: 1358–65. 04. Stratton MR. Recent advances in understanding of genetic susceptibility to breast cancer. Hum Mol Genet 1996; 5: 1515–9. 05. Woolas R, Jacobs I, Davies AP et al. What is the true incidence of primary fallopian tube carcinoma? Int J Gynecol Cancer 1994; 4: 384–8. 06. Zweemer RP, van Diest PJ, Verheijen RH et al. Molecular evidence linking primary cancer of the fallopian tube to BRCA1 germline mutations. Gynecol Oncol 2000; 76: 45–50. 07. Aziz S, Kuperstein G, Rosen B et al. A genetic epidemiological study of carcinoma of the fallopian tube. Gynecol Oncol 2001; 80: 337–40. 08. Paley PJ, Swisher EM, Garcia RL et al. Occult cancer of the fallopian tube in BRCA-1 germline mutation carriers at prophylactic oophorectomy: a case for recommending hysterectomy at surgical prophylaxis. Gynecol Oncol 2001; 80: 176–80. 09. Lynch HT. Hereditary nonpolyposis colorectal cancer (HNPCC). Cytogenet Cell Genet 1999; 86: 130–5. 10. Giardiello FM, Offerhaus GJ, Lee DH et al. Increased risk of thyroid and pancreatic carcinoma in familial adenomatous polyposis. Gut 1993; 34: 1394–6. 11. Stemmerman GN, Goodman MT, Nomura MY. Adenocarcinoma of the proximal small intestine: a marker for familial and multicentric cancer? Cancer 1992; 70: 2766–71. 12. Evans HS, Moller H, Robinson D et al. The risk of subsequent primary cancers after colorectal cancer in Southeast England. Gut 2002; 50: 647–52. 13. Tonin P, Weber B, Offit K et al. Frequency of recurrent BRCA1 and BRCA2 mutations in Ashkenazi Jewish breast cancer families. Nat Med 1996; 2: 1179–83. 14. Vasen HF, Mecklin JP, Meera Khan P, Lynch HT. The International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC). Dis Colon Rectum 1991; 34: 424–5. 15. Akiyama Y, Sato H, Yamada T et al. Germ-line mutation of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res 1997; 57: 3920–3. 16. Peltomaki P, Lothe RA, Aaltonen LA et al. Microsatellite instability is associated with tumors that characterize the hereditary non-polyposis colorectal carcinoma syndrome. Cancer Res 1993; 53: 5853–5. 17. Rodriguez-Bigas M, Boland CR, Hamilton SR et al. A National Cancer Institute Workshop on Hereditary Nonpolyposis Colorectal Cancer Syndrome: meeting highlights and Bethesda guidelines. J Natl Cancer Inst 1997; 89: 1758–62. 18. Cravo ML, Fidalgo PO, Lage PA et al. Validation and simplification of Bethesda guidelines for identifying apparently sporadic forms of colorectal carcinoma with microsatellite instability. Cancer 1999; 85: 77985. 19. Raedle J, Trojan J, Brieger A et al. Bethesda guidelines: relation to microsatellite instability and MLH1 promoter methylation in patients with colorectal cancer. Ann Intern Med 2001; 135: 566–76. 20. Wijnen JT, Vasen HF, Khan PM et al. Clinical findings with implications for genetic testing in families with clustering of colorectal cancer. N Engl J Med 1998; 339: 511–8. 21. Syngal S, Fox EA, Li C et al. Interpretation of genetic test results for hereditary nonpolyposis colorectal cancer: implications for clinical predisposition testing. JAMA 1999; 282: 247–53.

S. Levene et al. 22. Goggins M, Schutte M, Lu J et al. Germline BRCA2 mutations with apparently sporadic pancreatic carcinomas. Cancer Res 1996; 56: 5360–4. 23. Lal G, Liu G, Schmocker B et al. Inherited predisposition to pancreatic adenocarcinoma: role of family history and germ-line p16, BRCA1, and BRCA2 mutations. Cancer Res 2002; 60: 409–16. 24. Schottenfeld D. Multiple primary cancer. In Schottenfeld D, Fraumeni JF (eds): Cancer Epidemiology and Prevention, 2nd ed. New York: Oxford University Press, 1996: 1370–87. 25. Thames Cancer Registry. Cancer in South East England 1997. London, UK: Thames Cancer Registry, 2000. 26. Friedman LS, Ostermeyer EA, Szabo CI et al. Confirmation of BRCA1 by analysis of germline mutations linked to breast and ovarian cancer in ten families. Nat Genet 1994; 8: 399–404. 27. Kolodner RD, Hall NR, Lipford J et al. Structure of the human MSH2 locus and analysis of two Muir-Torre kindreds for msh2 mutations. Genomics 1994; 24: 516–26. 28. Hogervorst FB, Cornelis RS, Bout M et al. Rapid detection of BRCA1 mutations by the protein truncation test. Nat Genet 1995; 10: 208–12. 29. Lancaster JM, Wooster R, Mangion J et al. BRCA2 mutations in primary breast and ovarian cancers. Nat Genet 1996; 13: 238–40. 30. Beck NE, Tomlinson IP, Homfray T et al. Use of SSCP analysis to identify germline mutations in HNPCC families fulfilling the Amsterdam criteria. Hum Genet 1997; 99: 219–24. 31. Wagner TM, Hirtenlehner K, Shen P et al. Global sequence diversity of BRCA2: analysis of 71 breast cancer families and 95 control individuals of worldwide populations. Hum Mol Genet 1999; 8: 413-23. 32. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press, 1989: Chapter 6. 33. Ellis LA, Taylor CF, Taylor GR. A comparison of fluorescent SSCP and denaturing HPLC for high throughput mutation scanning. Hum Mutat 2000; 15: 556–64. 34. Taylor GR, Cuthbert-Heavens D, Smith A et al. Diagnostic mutation scanning using microtitre array capillary electrophoresis fluorescent SSCP. Eur J Hum Genet 2002; 10 (Suppl 1): 1032. 35. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press, 1989: Chapter 18. 36. Den Dunnen JT, van Ommen GJB. The protein truncation test: a review. Hum Mutat 1999; 14: 95–102. 37. Gill SS, Heuman DM, Mihas AA. Small intestinal neoplasms. J Clin Gastroenterol 2001; 33: 267–82. 38. Samowitz WS, Curtin K, Lin HH et al. The colon cancer burden of genetically defined hereditary nonpolyposis colon cancer. Gastroenterology 2001; 121: 830–8. 39. Gayther SA, Russell P, Harrington P et al. The contribution of germline BRCA1 and BRCA2 mutations to familial ovarian cancer: no evidence for other ovarian cancer-susceptibility genes. Am J Hum Genet 1999; 65: 1021–9. 40. Collaborative Group on Hormonal Factors in Breast Cancer. Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 women with breast cancer and 101,986 women without the disease. Lancet 2001; 358: 1389–99. 41. Boutron MC, Faivre J, Quipourt V et al. Family history of colorectal tumours and implications for the adenoma-carcinoma sequence: a case control study. Gut 1995; 37: 830–4. 42. House W, Sharp D, Sheridan E. Identifying and screening patients at high risk of colorectal cancer in general practice. J Med Screen 1999; 6: 205–8. 43. Adami HO, Bergkvist L, Krusemo U, Person I. Breast cancer as a risk factor for other primary malignant diseases. A nationwide cohort study. J Natl Cancer Inst 1984; 73: 1049–55. 44. Evans HS, Lewis CM, Robinson D et al. Incidence of multiple


45.

46.

47.

48.

49.

50.

51. 52.

53. 54.

55.

primary cancers in a cohort of women diagnosed with breast cancer in southeast England. Br J Cancer 2001; 84: 435–40. Shih HA, Nathanson KL, Seal S et al. BRCA1 and BRCA2 mutations in breast cancer families with multiple primary cancers. Clin Cancer Res 2000; 6: 4259–64. Schorge JO, Mahoney NM, Miller DS et al. Germline BRCA1–2 mutations in non-Ashkenazi families with double primary breast and ovarian cancer. Gynecol Oncol 2001; 83: 383–7. Planck M, Rambech E, Moslein G et al. High frequency of microsatellite instability and loss of mismatch-repair protein expression in patients with double primary tumors of the endometrium and colorectum. Cancer 2002; 94: 2502–10. Pal T, Flanders T, Mitchell-Lehman M et al. Genetic implications of double primary cancers of the colorectum and endometrium. J Med Genet 1998; 35: 978–84. Neugut AI, Ahsan H, Robinson E. Pancreas cancer as a second primary malignancy. A population-based study. Cancer 1995; 76: 589–92. Aarnio M, Sankila R, Pukkala E et al. Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer 1999; 81: 214–8. Breast Cancer Linkage Consortium. Cancer risks in BRCA2 mutation carriers. J Natl Cancer Inst 1999; 91: 1310–6. Vasen HF, Wijnen JT, Menko FH et al. Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis. Gastroenterology 1996; 110: 1020–7. Watson P, Lynch HT. Extracolonic cancer in hereditary nonpolyposis colorectal cancer. Cancer 1993; 71: 677–85. Johannsson O, Loman N, Moller T et al. Incidence of malignant tumours in relatives of BRCA1 and BRCA2 germline mutation carriers. Eur J Cancer 1999; 35: 1248–57. Eerola H, Pukkala E, Pyrhonen S et al. Risk of cancer in BRCA1 and BRCA2 mutation-positive and -negative breast cancer families (Finland). Cancer Causes Control 2001; 12: 739–46.

25

56. Vasen HF, Watson P, Mecklin JP, Lynch HT. New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative Group on HNPCC. Gastroenterology 1999; 116: 1453–6. 57. Tannergard P, Lipford J, Kolodner R et al. Mutation screening in the hMLH1 gene in Swedish hereditary nonpolyposis colon cancer families. Cancer Res 1995; 55: 6092–6. 58. Guerrette S, Acharya S, Fishel R. The interaction of the Human MutL homologues in hereditary nonpolyposis colon cancer. J Biol Chem 1999; 274: 6336–41. 59. Mauillon J, Michel P, Limacher JM et al. Identification of novel germline hMLH1 mutations including a 22kb Alu-mediated deletion in patients with familial colorectal cancer. Cancer Res 1996; 56: 5728–33. 60. Liu T, Tannergard P, Hackman P et al. Missense mutations in hMLK1 associated with colorectal cancer. Hum Genet 1999; 105: 437–41. 61. Farrington S, Lin-Goerke J, Ling J et al. Systematic analysis of hMSH2 and hMLH1 in young colon cancer patients and controls. Am J Hum Genet 1998; 63: 749–59. 62. Buersetdde JM, Alday P, Torhorst J et al. Detection of new mutations in six out of 10 Swiss HNPCC families by genomic sequencing of the hMSH2 and hMLH1 genes. J Med Genet 1995; 32: 909–12. 63. Wijnen J, Vasen H, Khan M et al. Seven new mutations in hMSH2, an HNPCC gene, identified by denaturing gradient-gel electrophoresis. Am J Hum Genet 1995; 56: 1060–6. 64. Samowitz W, Curtin K, Lin H et al. The colon cancer burden of genetically defined hereditary nonpolyposis colon cancer. Gastroenterology 2001; 121: 830–8. 65. Takehashi H, Chiu HC, Bandera C et al. Mutations of the BRCA2 gene in ovarian carcinomas. Cancer Res 1996; 56: 2738–41.