[CANCER RESEARCH 62, 3485–3492, June 15, 2002]
Evaluation of Microsatellite Instability and Immunohistochemistry for the Prediction of Germ-Line MSH2 and MLH1 Mutations in Hereditary Nonpolyposis Colon Cancer Families1 Siobhan S. Wahlberg, James Schmeits, George Thomas, Massimo Loda, Judy Garber, Sapna Syngal, Richard D. Kolodner,2 and Edward Fox Dana Farber Cancer Institute [S. S. W., G. T., M. L., J. G., S. S., E. F.], Brigham and Women’s Hospital [G. T., M. L., S. S.], Harvard Medical School, Boston, Massachusetts 02115, and Ludwig Institute of Cancer Research [S. S. W., J. S., R. D. K.], Department of Medicine [R. D. K.] and Cancer Center [R. D. K.], University of California San Diego School of Medicine, La Jolla, California 92903
ABSTRACT Forty-eight hereditary nonpolyposis colorectal carcinoma (HNPCC) families for which a tumor sample was available were evaluated for the presence of germ-line mutations in MSH2 and MLH1, tumor microsatellite instability (MSI), and where possible, expression of MSH2 and MLH1 in tumors by immunohistochemistry. Fourteen of 48 of the families had a germ-line mutation in either MSH2 or MLH1 that could be detected by genomic DNA sequencing, and 28 of 48 of the families had MSI-H tumors. Four additional families showed loss of expression of MSH2, and one additional family showed loss of expression of MLH1 but did not have germ-line mutations in MSH2 or MLH1 that could be detected by DNA sequencing. MSI-H, as defined using the National Cancer Institute recommended five-microsatellite panel, had a 100% sensitivity for identifying samples having MSH2 or MLH1 mutations or loss of expression. In contrast, loss of MSH2 and MLH1 expression did not identify all samples having germ-line mutations in MSH2 or MLH1, because in five cases, a mutant protein product was expressed that could be detected by IHC. A combination of the Bethesda criteria for HNPCC and an MSI-H phenotype defined the smallest number of cases having all of the germ-line MSH2 and MLH1 mutations that could be detected by DNA sequencing.
INTRODUCTION HNPCC3 is an autosomal dominant syndrome, characterized by predisposition to develop a number of cancers including CRC and endometrial, urinary, extracolonic gastrointestinal, brain, and ovarian cancers (1). Other characteristics of the syndrome include an early age of onset and development of multiple synchronous or metasynchronous cancers in some patients (2). HNPCC has been shown to be caused by inherited defects in genes encoding components of the major postreplication DNA MMR system (3, 4). Most HNPCC cases for which the genetic defect has been identified are attributable to defects in one of two genes, MSH2 and MLH1 (1, 5, 6).4 A small proportion of cases have been shown to be caused by germ-line mutations in two other MMR genes, MSH6 and PMS2, with most PMS2 mutations apparently being associated with Turcot’s syndrome (5, 7, 8).4 Initial molecular genetic studies of MMR gene defects in HNPCC have focused on missense, nonsense, frameshift, and splice site mutations (5).4 Recent studies have shown that HNPCC can also be caused by large deletion mutations and uncharacterized mutations Received 12/11/01; accepted 4/23/02. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by a grant from the Starr Foundation. R. D. K. is an inventor on both issued patents and pending patent applications covering some of the molecular diagnostic procedures discussed in this study. 2 To whom requests for reprints should be addressed, at Ludwig Institute for Cancer Research, UCSD School of Medicine-CMME3080, 9500 Gilman Drive, La Jolla, CA 92093-0660. Phone: (858) 534-7804; Fax: (858) 534-7750; E-mail:
[email protected]. 3 The abbreviations used are: HNPCC, hereditary nonpolyposis colon cancer; CRC, colorectal cancer; NCI, National Cancer Institute. 4 Internet address for database: http://www.nfdht.nl/database/mdbchoice.htm; MMR, mismatch repair; MSI, microsatellite instability; IHC, immunohistochemistry.
that lead to loss of expression or abnormal expression of some MMR genes (6, 9 –11). Carriers of a mutation in MLH1 or MSH2 have up to 70% lifetime risk of developing colorectal and other forms of cancer (12). Early detection of a germ-line alteration predicts HNPCC, allowing for preventative or early treatment for all family members. Predicting families with a high likelihood of having a diseasecausing mutation is of great importance, because mutation detection in a number of genes is time consuming and costly. Therefore, efforts have been made to find the most sensitive clinical criteria, based on family history, to be used to select families for mutation detection analyses. Selection of samples using a clinical or molecular phenotype could ideally reduce the sample number for mutation screening. Studies of families fulfilling the Amsterdam criteria, which are the most restrictive criteria for HNPCC, have identified germ-line MSH2 and MLH1 mutations with a relatively high sensitivity (⬃60%) and specificity (⬃70%; Refs. 13–15). Analysis of HNPCC cases identified by less strict criteria such as the Modified Amsterdam (16) and Bethesda criteria (2), which include extracolonic tumors, led to an increased sensitivity and a decreased specificity for the identification of germ-line MSH2 and MLH1 mutations. The Bethesda criteria, which are the least restrictive clinical criteria, predicted germ-line MSH2 and MLH1 mutations with an even higher sensitivity (⬃94%) and a lower specificity (⬃50%; Ref. 15). Early-onset colorectal cancer or cases with a family history of endometrial cancer have also been shown to be independent predictors of germ-line mutations (17, 18). These studies indicate that the restrictive criteria identify HNPCC cases with the highest likelihood of having a germ-line mutation but exclude many cases that have a germ-line mutation. In contrast, the less restrictive criteria identify almost all cases with a germ-line mutation but include many cases without a germ-line mutation. In a clinical setting where genetic testing is desirable but expensive, the established clinical criteria for HNPCC have significant disadvantages as a starting point for identifying individuals for testing. MSI, also termed the replication error phenotype, in CRC tumors from early-onset cases and cases with a family history, is a strong predictor of germ-line mutations in MMR genes such as MSH2, MLH1, and PMS2 but not MSH6 (18 –22). MSI can be detected in 90% of tumors with a germ-line MMR defect (23). This genomic instability is characterized by small deletions or insertions within simple repeat sequences, usually mono- or dinucleotide repeat sequences, in tumor DNA compared with corresponding DNA from normal tissue or blood (24). The altered size of the repeat sequences is the result of frameshift errors during DNA replication combined with the failure to repair these errors because of a DNA MMR defect (3). The loss of expression of MSH2 and MLH1 in tumors has also been suggested to be of value in predicting a germ-line mutation in MMR genes (25–27). In this method, IHC analysis of tumors using MLH1 and MSH2 antibodies has been used to detect loss of expression of MSH2 or MLH1. Neither MSI nor IHC can be used to definitively diagnose HNPCC among unselected CRC cases. This is because 15% or more of sporadic CRC cases have been found to be
3485
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
microsatellite unstable (28 –31). MSI in the majority of these cases is associated with methylation of the MLH1 promoter and silencing of the MLH1 gene (30, 32, 33). Thus, a significant proportion of unselected CRC cases will be sporadic, MSI positive cases that do not express MLH1 but do not have a germ-line MLH1 mutation in contrast to familial CRC cases with germ-line mutations in MLH1 or MSH2 (34, 35). We have studied previously the ability of the different clinical criteria for HNPCC to predict a germ-line mutation in MLH1 and MSH2 using 70 cases of familial colorectal cancer (15, 36). The families were selected by meeting at least one of several established HNPCC criteria including the Amsterdam, Modified Amsterdam, Bethesda, or HNPCC-like criteria (15). This study indicated that those clinical criteria, which identify most of the HNPCC cases having a germ-line MMR defect, also include many individuals who do not have such defects. In the present study, we have determined the tumor MSI status for 48 of 70 families used in our previous study and tested 24 of these individuals for MSH2 and MLH1 expression using IHC to
determine whether these tests could improve the selection criteria for identifying individuals for genetic testing. The results presented here demonstrate that MSI testing using the NCI 5-marker set identified all HNPCC cases where a germ-line MSH2 or MLH1 mutation was observed by DNA sequencing or where loss of expression of one of these two genes was observed. In contrast, five germ-line MSH2 or MLH1 mutations were observed that resulted in a MSI-H phenotype where both MSH2 and MLH1 proteins were still expressed in the tumors, suggesting that the IHC analysis may be a less useful indicator than MSI analysis of the presence of a germ-line mutation in either MSH2 or MLH1 in suspected HNPCC cases. MATERIALS AND METHODS Family Information. Families were identified by referral to a cancer genetics program at the Dana-Farber Cancer Institute and Harvard Medical School. The 48 families (Table 1) are a subset of the 70 families published for the sequencing of MLH1 and MSH2 (15, 36) and represent all of the cases for
Table 1 Summary of clinical, genetic, MSI, and IHC data HNPCC criteria Case
a
1755 2675 2906 357 2722 397 232 1448 4103 1251 1025 1851 1754 241 257 1642 2956 1103 3055 1446 1769 2496 245 629 1120 1102 2214 2738 1524 261 2911 230 1252 3045 2825 1239 1648 2228 1657 1264 1372 2848 2703 2248 2642 362 2851 487
AMS
M-AMS
X X X
X X X
X X
X X
X X
X X X
HNPCC-like
X X X X
X
X X
MMR gene defect
MSH2
MLH1
NCI
10
2,3,4,7 2,3,4,7 2,3,4 4 3,4 2,3,4
MLH1 IVS7-2A3G Splice MLH1 c.677G3A Splice MLH1 c.676C3T R226X MLH1 IVS9-1G3T splice MLH1 c.1810A3T K604X MLH1 IVS16⫹1G3A Splice MLH1 c.2265G3C R755S MLH1 c.2104-2105⌬AG FS MSH2 c.704-705⌬AA FS MSH2 IVS5⫹3A3T Splice MSH2 IVS5⫹3A3T Splice MSH2 IVS5⫹3A3T Splice MSH2 c.1352-1353⌬AG FS MSH2 c.1786⌬AAT N596⌬ None None None None None None None None None None None None None None None None None None None None None None None None None None None None None None None None None None
P P P ND P ND
P P ND ND P N
H H H H H H
H H H H H H
ND P N N ND ND N P P N ND P ND N P/N ND ND N P ND ND N P/N ND ND P/N ND ND ND ND P ND ND ND ND ND P/N ND ND ND P/N ND
ND P ND P P ND P ND P P ND N ND P P ND ND P/N ND ND ND P P ND ND P ND ND ND ND P ND ND ND ND ND ND ND ND ND ND ND
H H H H H H H H H H H H H H H H H H H H H H L L L L L L L S S S S S S S S S S S S S
H H H H L L H H H H H H H H H H H H H L H H L L L L L H L L L L S S S S S S S S L S
X X X X
X X
X X
X X
4 2 3,4 4 X 3,4 3,4 X 2 2,4
X X
X X X
2,3,4 4 3,4
X X
2 4 4
X X
X
3,4 3,4 3,4,7
X X X
MSI
Bethesda
3,4 2,3 2,3,4 2,3,4 7 3 2,3,4 3 3,4 3,4,7 3,4 4 3
X X X X X
IHC
a Clinical criteria and MSH2 and MLH1 mutations have been described (15, 36). No mutations in MSH6 were found in any of the indicated samples (20). AMS, Amsterdam; M-AMS, Modified Amsterdam; P, positive staining; N, negative for staining; P/N, mixture of P and N cells; ND, not determined; NCI, NCI 5 microsatellite markers (MSH-H, 2 or more unstable markers; MSI-L, 1 unstable marker; MSS, 0 unstable markers); 10, ten microsatellite marker set (MSH-H, 40% or more unstable markers; MSS, 0 unstable markers; MSI-L, intermediate between MSI-H and MSS).
3486
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
which a tumor sample was available. The families were classified according to various criteria: Amsterdam I, Modified Amsterdam, Bethesda criteria 2,3,4,7, and HNPCC-like. Five families met the HNPCC-like criteria only, 5 families met the Modified Amsterdam criteria only, whereas all others fulfilled the Amsterdam and/or Bethesda criteria, and in some cases also the Modified Amsterdam criteria. DNA Sequence Analysis. All 16 exons of MSH2 and all 19 exons of MLH1 were amplified from genomic DNA using primers that were designed to contain standard M13 forward and reverse primer sites in individual primers (Table 2). For most exons, one of the primers had an 18 base M13 (⫺21) forward sequence at its 5⬘-end, and the other primer had an 18 base M13 (⫺28) reverse sequence at its 5⬘ end. Several of the exons had only one M13-tailed primer listed because unambiguous sequence could only be generated from one end of the exon. For some of the exons, three primers are listed: in those cases, the noncoding strand primer without an M13 tail at its 5⬘ end was used routinely, and the noncoding strand primer with an M13 tail was only used if the PCR product required sequencing of its noncoding strand. Genomic DNA was isolated from blood samples as described previously. PCR was performed in 25–50 l reactions containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 1.5 mM MgCl2, 0.1 mM each dATP, dCTP, dGTP, TTP, 0.2 M of each primer, 5 ng/l DNA, and 0.05 unit/l Taq polymerase (Taq Platinum; Life Technologies, Inc., Gaithersburg, MD). Thermocycling conditions were: 94°C, 4 min, followed by 11 cycles, each with a denaturing step at 94°C for 20 s, an extension step at 72°C for 20 s, and with a 20-s annealing step that decreased 1°C/cycle (beginning at 60°C in the first cycle and decreasing to 50°C in the eleventh cycle), and the eleventh cycle was then repeated 24 times for a total of 35 cycles of PCR. Finally, the reactions were incubated for 6 min at 72°C, followed by storage at 4°C. For dye primer sequencing, an aliquot of the PCR reaction was diluted 1:10 with water. The diluted PCR product was then sequenced using an M13 Forward Big Dye Primer kit (Perkin-Elmer/Applied Biosystems, Foster City, CA) on a PE/ABI377 sequencer according to the manufacturer’s recommendations (15, 36). For dye terminator sequencing, each PRC product was purified by digestion with exonuclease I and shrimp alkaline phosphatase and then sequenced using dye terminator chemistry on PE/ABI377 and 3700 sequencers as described previously (20). Base calls were made using the instrument software and were reviewed by visual inspection. Each sequence was compared with the corresponding normal sequence using Sequencher 3.0 software (LifeCodes). All PCR products that yielded ambiguous sequences or contained heterozygous nucleotides were sequenced on both strands. Tumor Analysis. DNA for tumor analysis was extracted from microdissected tissue samples as follows. Ten m paraffin sections were cut using a new disposable blade and collected on glass slides. Slides were baked at 55°C in an oven for 3 h. Paraffin sections were deparaffinized with xylene, washed with ethanol, and rehydrated in deionized water. Lesional tissue was visualized under the microscope, microdissected with a 30-gauge 1/2-inch sterile needle, and collected in a sterile tube containing 20 l of digesting buffer (10 mM Tris, 1 mM EDTA, 1% Tween 20 and 200 g/ml proteinase K). From the same slides, normal (nontumor) tissue was microdissected using a new 30-gauge 1/2-inch sterile needle and put in a separate sterile tube containing 50 l of digesting buffer. The digestion was performed at 37°C for 36 h. The samples were then heated at 94°C for 5 min to inactivate the proteinase K and centrifuged, and the supernatant was used as template for PCR amplification and subsequent MSI analysis. Analysis of microsatellite sequences for instability was performed using a PE/ABI377 DNA sequencer. PCR reactions were performed in a PE/ABI 9600 thermocycler, and PE/ABI GeneScan software was used for data interpretation. The microsatellite loci referred to as the 5-marker set were BAT25, BAT26, D2S123, APC, and Mfd15, and the five additional microsatellite loci analyzed were BAT40, MYCL, D18S69, D18S58, and D10S197 (29, 37). Primer sequences were as described, and both fluorescently labeled and unlabeled primers were obtained from Life Technologies, Inc. PCR was performed in 10-l reactions, using 50 – 80 ng of DNA isolated from microdissected tissue samples, 0.2 M each primer (from a 2 M stock concentration), 1⫻ PE/ABI Buffer 2, 2.5 mM MgCl2, 250 M each of the 4 deoxynucleotide triphosphates (from a 2.5 mM stock; Boehringer Mannheim), and 0.25 unit of PE/ABI Amplitaq Gold. The thermocycling conditions were 1 cycle of 10 min 95°C, followed by 11 touchdown cycles consisting of 10 s 98°C; 30 s 60°C; 60°C decreasing 1°C/cycle; 1 min 70°C, followed by 37 cycles of 10 s 98°C; 1 min
58°C; 1 min 70°C, 10 min 99°C, followed by 1 cycle of 6 min 72°C. Samples were prepared for analysis by pooling 1 l of each of up to five reactions, precipitating them with ethanol, drying the pellet, and suspending it in 1.5 l of H2O. The resulting suspension was then mixed with 2.5 l of formamide, 0.4 l of blue dextran/EDTA and 0.6 l GS350 (Tamra; PE/ABI). One l of the final mixture was heated at 95°C for 2.20 min, placed on ice, and then loaded onto an acrylamide gel (FMC BioProducts, Rockland, ME) on a PE/ABI377 sequencer. IHC was performed exactly as described previously (32). The antibodies used for these studies were anti-MSH2 FE11 (Oncogene Research products, Cambridge, MA) and anti-MLH1 G168-728, (PharMingen, San Diego, CA).
RESULTS Identification of Germ-Line MSH2 and MLH1 Mutations. All 16 exons of MSH2 and 19 exons of MLH1, along with flanking intron sequences, were sequenced from genomic DNA from 70 HNPCC cases using a PE/ABI 377 sequencer and dye primer chemistry. The results of this analysis, which have been published previously (36), indicated that among these samples there were 18 germ-line mutations predicted to be pathogenic, because their predicted consequence was a protein truncation. There were also eight missense variants for which in many cases there were additional data suggesting that these variants were pathogenic mutations. None of these 70 samples was found to have a germ-line mutation in hMSH6 (20). Each of these exons containing a mutation or missense change was resequenced on both a PE/ABI377 sequencer and a PE/ABI3700 sequencer using dye terminator chemistry. There was a 100% concordance in detecting mutations using each sequencing method, indicating that the less expensive and less tedious dye terminator methods were sufficient for detecting all mutations and sequence variants in MSH2 and MLH1. Among the 48 samples for which a tumor sample was available, there were 8 MLH1 mutations and 6 MSH2 mutations among 48 samples (Table 1). One case (no. 1448) containing a MLH1 splice site mutation also had the MSH2 G322D missense change thought to be a polymorphism,4 and one case (no. 1755) containing a MLH1 splice site mutation also contained the MLH1 V716M missense change; we did not have appropriate material to allow us to determine whether the two MLH1 changes were associated with the same or different alleles. MSI Analysis of Tumors. Tumors from 48 cases were examined for MSI using both the NCI-recommended 5-marker test where MSI-H is defined as two or more markers of five showing instability and the NCI recommended 10-marker set, where MSI-H is defined as 4 or more markers of 10 showing instability (Table 1; Refs. 29, 37). This study design also allowed evaluation of the BAT26 single microsatellite test for MSI (38). Using the 5-marker test, the MSI-H phenotype was found in 28 (58%) of the 48 families tested. These included all 14 cases having a germ-line MSH2 or MLH1 mutation and all 5 cases in which loss of expression of either MSH2 or MLH1 was observed in the absence of a detected germ-line mutation. Thus, classifying tumors as MSI-H using the 5-marker test had a 100% sensitivity for detecting cases where a definitive MSH2 or MLH1 defect (germ-line mutation or loss of expression) could be demonstrated. The 10-marker test was slightly less effective for MSI analysis. Tumors from 26 cases were MSI-H using the 10-marker test, of which 25 were classified as MSI-H and 1 was classified MSI-L by the 5-marker test. However, only 12 of 14 cases with a germ-line mutation and all 5 cases where loss of expression without a germ-line mutation were MSI-H using the 10-marker test. When BAT26 was analyzed, it was found to be unstable in 23 of 28 (82%) cases that were MSI-H by the 5-marker test, it was unstable in 11 of the 14 (79%) cases with a germ-line MSH2 or MLH1 mutation, it was unstable in 15 of the 19 (79%) cases with either a germ-line MSH2 or MLH1 mutation or loss of expression of MSH2 or MLH1, and it was
3487
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
Table 2 hMLH1 and hMSH2 exon amplification primersa hMLH1 Exon 1 N-L1-X1UFC-L1-X1LRExon 2 N-L1-X2UFC-L1-X2LRExon 3 N-L1-X3UFC-L1-X3LRExon 4 N-L1-X4UFC-L1-X4LRExon 5 N-L1-X5UFC-L1-X5LRExon 6 N-L1-X6UFC-L1-X6LRExon 7 N-L1-X7UFC-L1-X7LRExon 8 N-L1-X8UFC-L1-X8LRExon 9 N-L1-X9UFC-L1-X9LRExon 10 N-L1-X10UFC-L1-X10LRExon 11 N-L1-X11UFC-L1-X11LRExon 12 N-L1-X12AUC-L1-X12ALFN-L1-X12BUFC-L1-X12BLRExon 13 N-L1-X13UFC-L1-X13LRExon 14 N-L1-X14URC-L1-X14LFExon 15 N-L1-X15URC-L1-X15LFExon 16 N-L1-X16UFC-L1-X16LRExon 17 N-L1-X17UFC-L1-X17LRExon 18 N-L1-X18UFC-L1-X18LRExon 19 N-L1-X19URC-L1-X19LFhMSH2 Exon 1 N-S2-X1UFC-S2-X1LExon 2 N-S2-X2UC-S2-X2LFExon 3 N-S2-X3UFC-Pr18784Exon 4 N-S2-X4UFC-S2-X4LExon 5 N-S2-X5UF2C-S2-X5LR2Exon 6 N-S2-X6UFC-S2-X6LRExon 7 N-S2-X7UFC-S2-X7LR-
5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CAC-TGA-GGT-GAT-TGG-CTG-AA 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-TAG-CCC-TTA-AGT-GAG-CCC-G 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TAC-ATT-AGA-GTA-GTT-GCA-GA 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CAG-AGA-AAG-GTC-CTG-ACT-C 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TTG-GAA-AAT-GAG-TAA-CAT-GAT-T 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-TGT-CAT-CAC-AGG-AGG-ATA-T 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CTT-TCC-CTT-TGG-TGA-GGT-GA 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-GAT-TAC-TCT-GAG-ACC-GC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TCT-CTT-TTC-CCC-TTG-GGA-TTA-G 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-ACA-AAG-CTT-CAA-CAA-TTT-ACT-CT 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GTT-TTA-TTT-TCA-AGT-ACT-TCT-ATG-AAT-T 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CAG-CAA-CTG-TTC-AAT-GTA-TCA-GCA-CT 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GTG-TGT-GTT-TTT-GGC-AAC 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CAT-AAC-CTT-ATC-TCC-ACC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-AGC-CAT-GAG-ACA-ATA-AAT-CCT-TG 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-GGT-TCC-CAA-ATA-ATG-TGA-TGG 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-AAG-CTT-CAG-AAT-CTC-TTT-T 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CTG-TGG-GTG-TTT-CCT-GTG 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-ACT-TTG-TGT-GAA-TGT-ACA-CCT-GTG 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-GAG-AGC-CTG-ATA-GAA-CAT-CT 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CTT-TTT-CTC-CCC-CTC-CCA-CTA 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-AAA-ATC-TGG-GCT-CTC-ACG 5⬘ 5⬘ 5⬘ 5⬘
GGG-ACC-TGT-ATA-TCT-ATA-CT TGT-AAA-ACG-ACG-GCC-AGT-GTT-TGC-TCA-GAG-GCT-GC TGT-AAA-ACG-ACG-GCC-AGT-ACA-GAA-TAA-AGG-AGG-TAG-GC CAG-GAA-ACA-GCT-ATG-ACC-GAT-GGT-TCG-TAC-AGA-TTC-CCG
5⬘ TGT-AAA-ACG-ACG-GCC-AGT-AAC-CCA-CAA-AAT-TTG-GCT-AAG 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CTT-TCT-CCA-TTT-CCA-AAA-CC 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-TGT-CTC-TAG-TTC-TGG-TGC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TGT-TGT-AGT-AGC-TCT-GCT-TG 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-ATT-TGT-CCC-AAC-TGG-TTG-TA 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TCA-GTT-GAA-ATG-TCA-GAA-GTG 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TTG-GAT-GCT-CCG-TTA-AAG-CTT-G 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CCG-GCT-GGA-AAT-TTT-ATT-TG 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-AGG-CAC-TGG-AGA-AAT-GGG-ATT 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-TCC-AGC-ACA-CAT-GCA-TGT-ACC-GA 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GTC-TGT-GAT-CTC-CGT-TTA-GA 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-AGG-TCC-TGT-CCT-AGT-CCT 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-GAC-ACC-AGT-GTA-TGT-TG 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CGG-AAT-ACA-GAG-AAA-GAA-GA
5⬘ TGT-AAA-ACG-ACG-GCC-AGT-AGG-CGG-GAA-ACA-GCT-TAG 5⬘ AAA-GGA-GCC-GCG-CCA-CAA 5⬘ GAA-GTC-CAG-CTA-ATA-CAG-TGC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CAC-ATT-TTT-ATT-TTT-CTA-CTC-TTA-A 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TAT-AAA-ATT-TTA-AAG-TAT-GTT-CAA-G 5⬘ TTT-CCT-AGG-CCT-GGA-ATC-TCC-TCT 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GTA-GGT-GAA-TCT-GTT-ATC-ACT 5⬘ CCT-TCT-AAA-AAG-TCA-CTA-TAG-T 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GAT-CCA-GTG-GTA-TAG-AAA-TCT-TCG 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-ATA-GTG-GAG-GAG-GGG-AGA-GAA-A 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TTC-ACT-AAT-GAG-CTT-GCC-ATT-C 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-GTA-TAA-TCA-TGT-GGG-TAA-C 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CTT-ACG-TGC-TTA-GTT-GAT-AA 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CAA-CCA-CCA-CCA-ACT-TTA-TGA
3488
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
Table 2 Continued Exon 8 N-S2-X8UFC-S2-X8LRExon 9 N-S2-X9UFC-S2-X9LRExon 10 N-S2-X10UFC-S2-X10LRExon 11 N-S2-X11UFC-S2-X11LRExon 12 N-S2-X12UFC-S2-X12LExon 13 N-S2-X13UFC-S2-X13LRExon 14 N-S2-X14UFC-S2-X14LExon 15 N-S2-X15UFC-S2-X15LExon 16 N-S2-X16UFC-S2-X16L-
5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TTT-GTA-TTC-TGT-AAA-ATG-ATG-AGA-TCT-TT 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-GGC-CTT-TGC-TTT-TTA-AAA-ATA-AC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GTG-GGA-GGA-AAT-ATT-TGC-TTT 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-TTG-GGG-ACA-GGG-AAC-TTA-TA 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TAG-TAG-GTA-TTT-ATG-GAA-TAC-TTT-T 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CTT-GAC-TCT-TAC-CTG-ATG-ACT 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CAT-TGC-TTC-TAG-TAC-ACA-TTT 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CAG-GTG-ACA-TTC-AGA-ACA-TTA 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-ATT-CAG-TAT-TCC-TGT-GTA-CAT 5⬘ TTA-CCC-CCA-CAA-AGC-CCA-A 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-CGA-TTA-ATC-ATC-AGT-GTA-C 5⬘ CAG-GAA-ACA-GCT-ATG-ACC-CAG-AGA-CAT-ACA-TTT-CTA-TCT-TC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-TGG-CAT-ATC-CTT-CCC-AAT-GT 5⬘ GGT-AGT-AAG-TTT-CCC-ATT-AC 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-GCT-GTC-TCT-TCT-CAT-GCT-G 5⬘ CAT-CTT-AGT-GTC-CTG-TTT-AT 5⬘ TGT-AAA-ACG-ACG-GCC-AGT-ATT-ACT-CAT-GGG-ACA-TTC-ACA 5⬘ CCA-TGG-GCA-CTG-ACA-GTT-A
a The underlined sequences are the sequences of the M13 Forward and Reverse sequencing primer sites that are present at the 5⬘ end of many of the primers. The extra AT in the hMSH2 exon 5 C-S2-X5LR2 primer was added to allow use of an Amersham sequencing primer. However, we have no reason to believe these two nucleotides are necessary for either the PCR of the use of the PE/ABI M13 primer. The two primer pairs given for hMLH1 exon 12 allow amplification of exon 12 in two overlapping fragments.
unstable in 2 of the 7 cases that were MSI-L by the 5-marker test. Analysis of the proportion of microsatellite markers found to be unstable in the tumors classified as MSI-H by the 5-marker criteria revealed that MLH1 mutant cases had a significantly higher proportion of unstable mononucleotide repeat markers compared with MSH2 mutant cases (Table 3). This difference was independent of whether all three mononucleotide repeats were analyzed or whether only BAT26 was analyzed. IHC Analysis of MLH1 and MSH2 in Tumors. IHC analysis of MSH2 and MLH1 expression was hampered by a lack of tumor material of sufficient quality. However, it was possible to perform IHC analysis on 24 tumors, of which 18 were analyzed for MLH1 staining and 22 were analyzed for MSH2 staining (Table 1). Selected examples of IHC analysis of cases expressing MLH1 or showing loss of expression of either MSH2 or MLH1 are presented in Fig. 1. There were 4 examples of loss of MSH2 expression and 1 example of loss of MLH1 expression in MSI-H cases where no germ-line mutation in MSH2 or MLH1 was present. In addition, there was 1 example of MSH2 expression in a case with a germ-line MSH2 mutation and there were 4 examples of MLH1 expression in cases where a germ-line mutations in MLH1 was found. The MSH2 mutant case that expressed MSH2 contained a single codon deletion (N596del) and would be predicted to express a full-length protein. Two of the MLH1 mutant cases that expressed MLH1 had protein-truncating mutations late in the gene (exon 16, K604X and exon 19, c.2104 –2105delAG) and might be expected to express large fragments of MLH1. The remaining two MLH1 mutant cases that expressed MLH1 had splice site mutations (c.677G3 A and IVS7-A3 G) that might be predicted to
result in skipping of exon 8; because this would not result in an in-frame deletion, these mutations would be predicted to be more deleterious to MLH1 expression. Evaluation of Clinical Criteria. Of the 48 families subject to tumor analysis (see below), all except 9 fulfilled the guidelines for at least one of the Bethesda or Amsterdam criteria (Table 1). These 9 cases were either Modified Amsterdam or HNPCC-like, and none of them had a germ-line mutation in either MSH2 or MLH1. The value of these clinical criteria for predicting the presence of germ-line mutations in MSH2 and MLH1 has been evaluated previously using the 70-sample set (15). Because this study used only 48 of the 70 families included in the first analysis, the sensitivity and specificity of each clinical criteria for predicting germ-line mutations in MSH2 and MLH1 was recalculated for comparison and found to be similar to that reported previously. Using the 48 families, the Amsterdam criteria showed the highest specificity (68%) for detecting germ-line mutations but a low sensitivity (57%), thus missing many mutations. In contrast, all of the samples with identified germ-line mutations met at least one of the Bethesda criteria. However, the specificity of the Bethesda criteria was only 29%, indicating that most of the samples meeting these latter criteria did not contain a germ-line mutation that could be detected by sequence analysis. It should be noted that of the 5 cases in which loss of expression of either MSH2 or MLH1 was observed in the absence of a detected germ-line mutation, 1 of the 5 did not meet the Bethesda criteria. The MSI status of the cases was also evaluated relative to the clinical characteristics of the cases. Eighty-six % (24 of 28) of the MSI-H families met at least one of the Bethesda criteria, and the
Table 3 Effect of mutations in hMSH2 and hMLH1 on MSI at total and mononucleotide repeat loci
All MSI-H MSI-H with a MMR defecta MSI-H with a hMLH1 mutationa MSI-H with hMSH2 mutationa
n
Total markers
Unstable markers
Total mononucleotide repeats
Unstable mononucleotide repeats
Total BAT26
Unstable BAT26
28 19 8 6
226 158 68 48
165 (73%) 117 (74%) 53 (78%) 30 (63%)
81 56 24 17
64 (79%) 46 (82%) 23 (96%) 9 (53%) P ⬍ 0.0001
28 19 8 6
23 (82%) 15 (79%) 8 (100%) 3 (50%) P ⬍ 0.03
a “Defect” refers to those cases showing either a germ-line mutation in hMSH2 or hMLH1 or loss of expression in the absence of a germline mutation. “Mutation” includes only those cases where a germ-line mutation was found. Ps refer to the comparison of the indicated hMSH2 and hMLH1 data and were calculated using the 2 test.
3489
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
Fig. 1. Immunohistochemistry analysis of MLH1 expression in tumors from 4 HNPCC families having germ-line mutations in MLH1. Immunohistochemistry for detecting MLH1 and MSH2 expression in normal and tumor sections of the indicated cases was performed as described in “Materials and Methods.” MLH1 and MSH2 staining are visualized in brown. A, tumor expressing MLH1 from case 2722 containing the c.1810A3 T K604X nonsense mutation in MLH1 exon 16. B, tumor expressing MLH1 from case 2675 containing the c.677G3 A splice site mutation in MLH1 exon 8. Note that this mutation also potentially results in the R226Q missense change. C, tumor expressing MLH1 from case 1448 containing the c.2104 –2105delAG frameshift mutation in MLH1 exon 19. D, tumor expressing MLH1 from case 1755 containing the IVS7–2A3 G splice site mutation in MLH1 exon 7. E, normal tissue expressing MLH1 from case 397 containing the IVS16 ⫹ 1G3 A splice site mutation in MLH1 intron 16. F, tumor tissue not expressing MLH1 from case 397 containing the IVS16 ⫹ 1G3 A splice site mutation in MLH1 intron 16. G, normal tissue expressing MSH2 from case 4103 containing the c.704 –705⌬AA frameshift mutation in MSH2 exon 4. H, tumor tissue not expressing MSH2 from case 4103 containing the c.704 –705⌬AA frameshift mutation in MSH2 exon 4. Note that is this latter section (H), there is some light cytoplasmic staining, but the nuclei are negative for staining.
specificity of the Bethesda criteria for identifying MSI-H cases was 63%. In contrast, 54% of the MSI-H cases met the Amsterdam criteria, and the specificity of the Amsterdam criteria for identifying MSI-H cases was 80%. Thus, similar to the situation with germ-line mutation detection, the Bethesda criteria identify most but not all MSI-H cases at the expense of including many cases that are not MSI-H, whereas the Amsterdam criteria identify a smaller proportion of the MSI-H cases but include a much smaller proportion of cases that are not MSI-H. Similar results on the relationship between MSI status and clinical criteria have also been reported in two recent studies (39, 40). DISCUSSION The aim of this study was to evaluate the value of MSI, IHC, different clinical criteria, and different sequencing methods for predicting and detecting germ-line mutations in MMR genes in families suspected of having HNPCC. To accomplish this, 48 families for which a tumor sample was available were evaluated. The results presented here support a number of conclusions: (a) Dye terminator sequencing of PCR-amplified exons using a capillary sequencer is adequate for detecting heterozygous germ-line mutations in MSH2 and MLH1. (b) The Bethesda criteria identify all cases having a germ-line MSH2 or MLH1 mutation but included many cases that did not have such defects, as well as including many cases that were not MSI-H and were unlikely to have MMR defects. (c) All of the tumors from cases having a germ-line defect in MSH2 or MLH1 or showing loss of expression of one of these genes in a tumor sample were found to be MSI-H using the NCI-recommended 5-marker test. In contrast, analyses with just the BAT26 mononucleotide repeat was inadequate to detect all of the MMR-defective cases or all of the cases found to be MSI-H by the 5-marker test. Similarly, the 10-marker test did not detect all of the MMR-defective cases.
(d) A combination of the Bethesda criteria and MSI-H defined the smallest number of cases that included all of the germ-line MSH2 or MLH1 mutations found but excluded one case where loss of MSH2 expression in the absence of a germ-line mutation was seen. (e) IHC analysis was of little use in helping to define those cases where a germ-line mutation was found because the tumors from some cases containing a pathogenic mutation nonetheless express protein that is detected by IHC. We have described an improved set of PCR primers that allow amplification and sequencing of each exon of MSH2 and MLH1 without the need for nested PCR. The results obtained here showed that dye terminator sequencing on either a PE/ABI377 or PE/ABI3700 sequencer was capable of detecting all heterozygous mutations that could be detected using dye primer sequencing. Dye primer sequencing is well suited for mutation detection in heterozygotes because of the uniformity of sequencing chromatogram peak heights obtained but suffers from the disadvantage of being tedious because of the need to perform four sequencing reactions/sample. The ability to use dye terminator sequencing offers a significant advantage because it is less tedious to perform, and using a capillary sequencer, it is less expensive because it is possible to work with smaller sequencing reaction volumes. Previous studies have shown that the majority of tumors from HNPCC patients show MSI, and similarly, a smaller proportion of sporadic cancers show MSI (18, 19, 23, 24, 29). An NCI-sponsored workshop has recommended two standard sets of microsatellite markers for use in MSI analysis (29, 37). These markers were recommended based on the observation that they each showed a high degree of instability in a subset of primarily sporadic tumors in which a high proportion of markers were unstable (29). However, these markers have not been well validated using samples in which the nature of the MMR defects has been established. The results presented here show that the NCI 5-marker test detected 100% of the samples shown to
3490
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
have either a germ-line mutation in MSH2 or MLH1 or loss of expression in the absence of a detected mutation. To our knowledge, this is the first study to actually evaluate whether the NCI 5-marker test can actually detect all HNPCC cases having a germ-line mutation in a MMR gene and demonstrate this is indeed true. An unexpected feature of the data was that the MSH2-defective tumors showed a lower level on mononucleotide repeat instability compared with the MLH1-defective tumors. The functional basis for this is unclear and could represent a mechanistic difference between MSH2 and MLH1 defects, a difference in the nature of inactivation of the second allele in the tumors, or it could be attributable to statistical variation in a small sample set. Some studies have suggested that mononucleotide repeat instability and, in particular instability of BAT26, is sufficient for detection of MSI (28, 38). However, in the present study of HNPCC cases, such criteria would have missed a significant number of cases that proved to have a defect in MSH2. Similarly, a recent study of sporadic colon tumors identified several cases where BAT26 was stable, but expression of either MSH2 or MLH1 was absent (41). Interestingly, the studies that have reported successful use of BAT26 have primarily analyzed sporadic tumors, or sporadic and HNPCC tumors from Finland, and in each of these sample sets MLH1 is the gene that is most often defective (28, 30, 33, 38, 42). Similar to this, all of the cases with a MLH1 germ-line mutation analyzed here showed BAT26 instability. IHC analysis of MMR gene expression has proven useful in the detection of MMR defects, particularly in the case of sporadic tumors where loss of MLH1 expression is often seen (29, 30, 32–35). In the present study, IHC analysis was not useful in predicting MSH2 or MLH1 gene defects. This is because there were 5 cases where significant pathogenic germ-line mutations, including protein-truncating mutations, were found but nonetheless the tumors produced presumably nonfunctional proteins that were detected by IHC. This is not surprising, however, because in at least 3 of the cases, the nature of the mutation predicted that a full length or almost full-length, but nonfunctional, protein would be produced. In the other two cases, it is possible that the mutation resulted in the production of a stable protein fragment that was detected by IHC; it is also possible that the wild type allele was still present in the tumor, and the protein fragment produced caused a dominant-negative effect. It is also reasonable to expect the presence of proteins that would be detected by IHC in cases where a germ-line missense mutation was present. Indeed, other studies have also observed that IHC can detect expression of MSH2 or MLH1 in tumors from cases where an MSH2 or MLH1 mutation was present, although such expression was only observed in cases with missense mutations (43– 45), a result that might be expected. Our observation of MSH2 and MLH1 protein expression in a significant proportion of HNPCC cases containing a germ-line MMR defect including cases containing protein-truncating mutations indicates that IHC is unlikely to be a definitive primary test for detecting the presence of germ-line mutations in MMR genes in HNPCC cases. We observed a much higher proportion of MSH-H cases that expressed both MSH2 and MLH1 even when a germ-line MSH2 or MLH1 mutation was present than observed in a recent large scale study of unselected CRC cases, where ⬃8% of MSI-H cases expressed both MSH2 and MLH1 (46). The difference between these two studies is that our study focused on suspected HNPCC cases where germ-line mutations would be found, whereas the recent large-scale study included a high proportion of sporadic CRC cases, which are primarily attributable to silencing of MLH1 and hence loss of expression of MLH1 (29, 30, 32–35). In the analysis of suspected HNPCC cases, IHC seems more useful in the analysis of MSI-H cases where no germ-line MSH2 or MLH1 mutations were found than as a primary screen for the presence of MMR defects. Five such cases were
identified in which MSH2 (4 cases) or MLH1 (1 case) expression was absent, and these would be candidates for analysis for the presence of other types of MSH2 or MLH1 defects, such as deletion mutations (6, 9 –11). A critical question regarding the molecular analysis of CRC cases suspected of being HNPCC is what strategy should be used to analyze such cases for germ-line defects in MSH2 and MLH1, the major HNPCC genes (6),4 so as to minimize the work and expense involved while maximizing the fraction of defects detected. Our studies as well as those of others indicate that the Bethesda criteria identify a greater proportion of CRC cases having a germ-line MMR defect than other clinical criteria. Thus, the Bethesda criteria perform very well in achieving the intended goal to help guide which CRC families should undergo molecular evaluation for HNPCC. The data presented here indicate that a second step of MSI analysis using the NCI 5-marker test would yield the smallest number of cases (24 from a total of 48) where all of the MSH2 and MLH1 germ-line mutations were present. Not only is MSI analysis the only molecular method that detected all cases containing a germ-line MMR defect in MSH2 and MLH1, in so much as MSI analysis is significantly less expensive than DNA sequence analysis, MSI analysis is also a cost-effective second step. In our study, IHC analysis of MSI-H cases was not definitive for identifying cases with germ-line MMR defects in MSH2 and MLH1. Because our study indicates that MSI-H cases that express MSH2 and MLH1 proteins would still need to be analyzed for germ-line mutations in both MSH2 and MLH1, if IHC analysis were performed after MSI analysis, it would only reduce the overall amount of DNA sequence analysis required by ⬃20%. In contrast, if IHC analysis is performed on only those cases that are MSI-H where DNA sequencing did not identify a germ-line mutation in MSH2 or MLH1, IHC analysis provided an indication of whether MSH2 or MLH1 contained some sort of as yet detected germ-line defect in 5 of 7 of MSI-H, mutation-negative cases where definitive IHC was possible. This would be extremely useful in guiding additional analysis, such as the use of conversion analysis or analysis for deletion mutations (6, 9 –11). In contrast, if conversion and deletion mutation analysis were applied at earlier stages in the analysis, such as before MSI testing or before sequence analysis, these methods would only detect mutations in a small proportion of cases and would not significantly improve the efficiency of mutation detection by DNA sequencing. ACKNOWLEDGMENTS We thank R. Fishel, I. Gazzoli, A. Lindblom, and J. Mueller for comments on the manuscript and J. Weger and J. Green for performing some of the reported DNA sequencing.
REFERENCES 1. Lynch, H. T., and de la Chapelle, A. Genetic susceptibility to non-polyposis colorectal cancer. J. Med. Genet., 36: 801– 818, 1999. 2. Rodriguez-Bigas, M. A., Boland, C. R., Hamilton, S. R., Henson, D. E., Jass, J. R., Khan, P. M., Lynch, H. T., Perucho, M., Smyrk, T., Sobin, L., and Srivastava, S. A National Cancer Institute workshop on hereditary nonpolyposis colorectal cancer syndrome: meeting highlights and Bethesda Guidelines. J. Natl. Cancer Inst., 89: 1758 –1762, 1997. 3. Kolodner, R. D., and Marsischky, G. T. Eukaryotic DNA mismatch repair. Curr. Opin. Genet. Dev., 9: 89 –96, 1999. 4. Jiricny, J., and Nystron-Lahti, M. Mismatch repair defects in cancer. Curr. Opin. Genet. Dev., 10: 157–161, 2000. 5. Peltomaki, P., and Vasen, H. F. Mutations predisposing to hereditary nonpolyposis colorectal cancer: database and results of a collaborative study. The International Collaborative Group on Hereditary Nonpolyposis Colorectal Cancer. Gastroenterology, 113: 1146 –1158, 1997. 6. Yan, H., Papadopoulos, N., Marra, G., Perrera, C., Jiricny, J., Boland, C. R., Lynch, H. T., Chadwick, R. B., de la Chapelle, A., Berg, K., Eshleman, J. R., Yuan, W., Markowitz, S., Laken, S. J., Lengauer, C., Kinzler, K. W., and Vogelstein, B. Conversion of diploidy to haploidy. Nature (Lond.), 403: 723–724, 2000.
3491
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
PREDICTION OF GERM-LINE MSH2 AND MLH1 MUTATIONS
7. Wijnen, J., de Leeuw, W., Vasen, H., van der Klift, H., Moller, P., Stormorken, A., Meijers-Heijboer, H., Lindhout, D., Menko, F., Vossen, S., Moslein, G., Tops, C., Brocker-Vriends, A., Wu, Y., Hofstra, R., Sijmons, R., Cornelisse, C., Morreau, H., and Fodde, R. Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat. Genet., 23: 142–144, 1999. 8. Liu, T., Yan, H., Kuismanen, S., Percesepe, A., Bisgaard, M. L., Pedroni, M., Benatti, P., Kinzler, K. W., Vogelstein, B., Ponz de Leon, M., Peltomaki, P., and Lindblom, A. The role of hPMS1 and hPMS2 in predisposing to colorectal cancer. Cancer Res., 61: 7798 –7802, 2001. 9. Wijnen, J., van der Klift, H., Vasen, H., Khan, P. M., Menko, F., Tops, C., Meijers Heijboer, H., Lindhout, D., Moller, P., and Fodde, R. MSH2 genomic deletions are a frequent cause of HNPCC. Nat. Genet., 20: 326 –328, 1998. 10. Charbonnier, F., Olschwang, S., Wang, Q., Boisson, C., Martin, C., Buisine, M. P., Puisieux, A., and Frebourg, T. MSH2 in contrast to MLH1 and MSH6 is frequently inactivated by exonic and promoter rearrangements in hereditary nonpolyposis colorectal cancer. Cancer Res., 62: 848 – 853, 2002. 11. Wang, Y., Friedl, W., Sengteller, M., Jungck, M., Filges, I., Propping, P., and Mangold, E. A modified multiplex PCR assay for detection of large deletions in MSH2 and MLH1. Hum. Mutat., 19: 279 –286, 2002. 12. Aarnio, M., Sankila, R., Pukkala, E., Salovarra, R., Aaltonen, L. A., de la Chapelle, A., Peltomaki, P., Mecklin, J-P., and Jarvinen, H. J. Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int. J. Cancer, 81: 214 –218, 1999. 13. Wahlberg, S., Liu, T., Lindblom, P., and Lindblom, A. Various mutation screening techniques in the DNA mismatch repair genes hMSH2 and hMLH1. Genet. Testing, 3: 259 –264, 1999. 14. Wijnen, J., Khan, P. M., Vasen, H. F., van der Klift, H., Mulder, A., van LeeuwenCornelisse, I., Bakker, B., Losekoot, M., Moller, P., and Fodde, R. Hereditary nonpolyposis colorectal cancer families not complying with the Amsterdam criteria show extremely low frequency of mismatch-repair-gene mutations. Am. J. Hum. Genet., 61: 329 –335, 1997. 15. Syngal, S., Fox, E. A., Eng, C., Kolodner, R. D., and Garber, J. E. Sensitivity and specificity of clinical criteria for hereditary non-polyposis colorectal cancer associated mutations in MSH2 and MLH1. J. Med. Genet., 37: 641– 645, 2000. 16. Bellacosa, A., Genuardi, M., Anti, M., Viel, A., and Ponz de Leon, M. Hereditary nonpolyposis colorectal cancer: review of clinical, molecular genetics and counseling aspects. Am. J. Med. Genet., 62: 353–364, 1996. 17. Wijnen, J., Vasen, H. F. A., Khan, M. P., Zwinderman, A. H., van der Klift, H., Mulder, A., Tops, C., Møller, P., and Fodde, R. Clinical findings with implications for genetic testing in families with clustering of colorectal cancer. N. Engl. J. Med., 339: 511–518, 1998. 18. Farrington, S. M., Lin-Goerke, J., Ling, J., Wang, Y., Burczak, J. D., Robbins, D. J., and Dunlop, M. G. Systematic analysis of hMSH2 and hMLH1 in young colon cancer patients and controls. Am. J. Hum. Genet., 63: 749 –759, 1998. 19. Lui, T., Wahlberg, S., Burek, E., Lindblom, P., Rubio, C., and Lindblom, A. Microsatellite instability as a predictor of a mutation in a DNA mismatch repair gene in familial colorectal cancer. Genes Chromosomes Cancer, 27: 17–25, 2000. 20. Kolodner, R. D., Tytell, J. D., Schmeits, J., Kane, M. F., Das Gupta, R., Weger, J., Wahlberg, S., Fox, E. A., Peel, D. J., Ziogas, A., Garber, J. E., Syngal, S., AntonCulver, H., and Li, F. P. Germline msh6 mutations in colorectal cancer families. Cancer Res., 59: 5068 –5074, 1999. 21. Akiyama, Y., Sato, H., Yamada, T., Nagasaki, H., Tsuchiya, A., Abe, R., and Yuasa, Y. Germline mutations of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res., 57: 3920 –3923, 1997. 22. Miyaki, M., Konishi, M., Tanaka, K., Kituchi-Yanoshita, R., Muraoka, M., Igari, T., Koike, M., Chiba, M., and Mori, T. Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nat. Genet., 17: 271–272, 1997. 23. Aaltonen, L. A., Peltomaki, P., Mecklin, J. P., Jarvinen, H., Jass, J. R., Green, J. S., Lynch, H. T., Watson, P., Tallqvist, G., Juhola, M., Sistonen, P., Hamilton, S. R., Kinzler, K. W., Vogelstein, B., and de la Chapelle, A. Replication errors in benign and malignant tumors from hereditary non-polyposis colorectal cancer patients. Cancer Res., 54: 1645–1648, 1994. 24. Ionov, Y., Peinado, M. A., Malkhosyan, S., Shibata, D., and Perucho, M. Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature (Lond.), 363: 558 –561, 1993. 25. Marcus, V. A., Madlensky, L., Gryfe, R., Kim, H., So, K., Miller, A., Temple, L. K. F., Hsieh, E., Hiruki, T., Narod, S., Bapat, B., Gallinger, S., and Redston, M. Immunohistochemistry for hMLH1 and hMSH2: a practical test for DNA mismatch repair-deficient tumors. Am. J. Surg. Pathol., 23: 1248 –1255, 1999. 26. Cawkwell, L., Gray, S., Murgatroyd, H., Sutherland, F., Haine, L., Longfellow, M., O’Loughlin, S., Cross, D., Kronborg, O., Fenger, C., Mapstone, N., Dixon, M., and Quirke, P. Choice of management strategy for colorectal cancer based on a diagnostic immunohistochemical test for defective mismatch repair. Gut, 45: 409 – 415, 1999. 27. Stone, J. G., Robertson, D., and Houlston, R. S. Immunohistochemistry for MSH2 and MHL1: a method for identifying mismatch repair deficient colorectal cancer. J. Clin. Pathol., 54: 484 – 487, 2001. 28. Aaltonen, L. A., Salovaara, R., Kristo, P., Canzian, F., Hemminki, A., Peltomaki, P., Chadwick, R. B., Kaariainen, H., Eskelinen, M., Jarvinen, H., Mecklin, J-P., and de
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
la Chapelle, A. Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N. Engl. J. Med., 338: 1481–1487, 1998. Dietmaier, W., Wallinger, S., Bocker, T., Kullmann, F., Fishel, R., and Ruschoff, J. Diagnostic microsatellite instability: definition and correlation with mismatch repair protein expression. Cancer Res., 57: 4749 – 4756, 1997. Thibodeau, S. N., French, A. J., Cunningham, J. M., Tester, D., Burgart, L. J., Roche, P. C., McDonnell, S. K., Schaid, D. J., Walsh Vockley, C., Michels, V. V., Farr, G. H., and O’Connell, M. J. Microsatellite instability in colorectal cancer: different mutator phenotypes and the principle involvement of hMLH1. Cancer Res., 58: 1713–1718, 1998. Cunningham, J. M., Kim, C. Y., Christensen, E. R., Tester, D. J., Parc, Y., Burgart, L. J., Halling, K. C., McDonnell, S. K., Schaid, D. J., Walsh Vockley, C., Kubly, V., Nelson, H., Michels, V. V., and Thibodeau, S. N. The frequency of hereditary defective mismatch repair in a prospective series of unselected colorectal carcinomas. Am. J. Hum. Genet., 69: 780 –790, 2001. Kane, M. F., Loda, M., Gaida, G. M., Lipman, J., Mishra, R., Goldman, H., Jessup, J. M., and Kolodner, R. D. Methylation of the hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res., 57: 808 – 811, 1997. Herman, J. G., Umar, A., Polyak, K., Graff, J. R., Ahuja, N., Issa, J. P., Markowitz, S., Willson, J. K., Hamilton, S. R., Kinzler, K. W., Kane, M. F., Kolodner, R. D., Vogelstein, B., Kunkel, T. A., and Baylin, S. B. Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc. Natl. Acad. Sci. USA, 95: 6870 – 6875, 1998. Kuismanen, S. A., Holmberg, M. T., Salovaara, R., Schweizer, P., Aaltonen, L. A., de la Chapelle, A., Nystrom-Lahti, M., and Peltomaki, P. Epigenetic phenotypes distinguish microsatellite stable and -unstable colorectal cancers. Proc. Natl. Acad. Sci. USA, 96: 12661–12666, 1999. Salahshor, S., Koelble, K., Rubio, C., and Lindblom, A. Evaluation of hMLH1 and hMSH2 immunohistochemistry in familial and sporadic colorectal cancer. Lab. Investig., 81: 535–541, 2001. Syngal, S., Fox, E., Li, C., Dovidio, M., Eng, C., Kolodner, R. D., and Garber, J. E. Interpretation of genetic test results for hereditary nonpolyposis colon cancer. J. Am. Med. Assoc., 282: 247–253, 1999. Boland, C., Thibodeau, S., Hamilton, S. R., Sidransky, D., Eshleman, J., Burt, R., Meltzer, S., Rodriguez-Bigas, M., Fodde, R., Ranzani, N., and Srivastava, S. A National Cancer Institute workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res., 58: 5248 –5257, 1998. Hoang, J. M., Cottu, P. H., Thuille, B., Salmon, R. J., Thomas, G., and Hamelin, R. BAT-26, an indicator of the replication error phenotype in colorectal cancer cell lines. Cancer Res., 57: 300 –303, 1997. Terdiman, J. P., Gum, J. R., Jr., Conrad, P. G., Miller, G. A., Weinberg, V., Crawley, S. C., Levin, T. R., Reeves, C., Schmitt, A., Hepburn, M., Sleisenger, M. H., and Kim, Y. S. Efficient detection of hereditary nonpolyposis colorectal cancer gene carriers by screening for tumor microsatellite instability before germline genetic testing. Gastroenterology, 120: 21–30, 2001. Wullenweber, H. P., Sutter, C., Autschbach, F., Willeke, F., Kienle, P., Benner, A., Bahring, J., Kadmon, M., Herfarth, C., von Knebel Doeberitz, M., and Gebert, J. Evaluation of Bethesda guidelines in relation to microsatellite instability. Dis. Colon Rectum, 44: 1281–1289, 2001. Perrin, J., Gouvernet, J., Parriaux, D., Noguchi, T., Giovannini, M. H., Giovannini, M., Delpero, J. R., Birnbaum, D., and Monges, G. MSH2 and MLH1 immunodetection and the prognosis of colon cancer. Int. J. Oncol., 19: 891– 895, 2001. Loukola, A., Eklin, K., Laiho, P., Salovaara, R., Kristo, P., Jarvinen, H., Mecklin, J. P., Launonen, V., and Aaltonen, L. A. Microsatellite marker analysis in screening for hereditary nonpolyposis colorectal cancer (HNPCC). Cancer Res., 61: 4545– 4549, 2001. Orth, K., Hung, J., Gazdar, A., Bowcock, A., Mathis, J. M., and Sambrook, J. Genetic instability in human ovarian cancer cell lines. Proc. Natl. Acad. Sci. USA, 91: 9495–9499, 1994. Thibodeau, S. N., French, A. J., Roche, P. C., Cunningham, J. M., Tester, D. J., Lindor, N. M., Moslein, G., Baker, S. M., Liskay, R. M., Burgart, L. J., Honchel, R., and Halling, K. C. Altered expression of hMSH2 and hMLH1 in tumors with microsatellite instability and genetic alterations in mismatch repair genes. Cancer Res., 56: 4836 – 4840, 1996. Dieumegard, B., Grandjouan, S., Sabourin, J. C., Le Bihan, M. L., Lefrere, I., Bellefqih, S., Pignon, J. P., Rougier, P., Lasser, P., Benard, J., Couturier, D., and Bressac-de Paillerets, B. Extensive molecular screening for hereditary non-polyposis colorectal cancer. Br. J. Cancer, 82: 871– 880, 2000. Lindor, N. M., Burgart, L. J., Leontovich, O., Goldberg, R. M., Cunningham, J. M., Sargent, D. J., Walsh-Vockley, C., Petersen, G. M., Walsh, M. D., Leggett, B. A., Young, J. P., Barker, M. A., Jass, J. R., Hopper, J., Gallinger, S., Bapat, B., Redston, M., and Thibodeau, S. N. Immunohistochemistry versus microsatellite instability testing in phenotyping colorectal tumors. J. Clin. Oncol., 20: 1043–1048, 2002.
3492
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.
Evaluation of Microsatellite Instability and Immunohistochemistry for the Prediction of Germ-Line MSH2 and MLH1 Mutations in Hereditary Nonpolyposis Colon Cancer Families Siobhan S. Wahlberg, James Schmeits, George Thomas, et al. Cancer Res 2002;62:3485-3492.
Updated version
Cited articles Citing articles
E-mail alerts Reprints and Subscriptions Permissions
Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/62/12/3485
This article cites 46 articles, 20 of which you can access for free at: http://cancerres.aacrjournals.org/content/62/12/3485.full#ref-list-1 This article has been cited by 22 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/62/12/3485.full#related-urls
Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
[email protected]. To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/62/12/3485. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.
Downloaded from cancerres.aacrjournals.org on April 9, 2019. © 2002 American Association for Cancer Research.