A Prospective, Multicenter, Population-Based Study of BRAF Mutational Analysis for Lynch Syndrome Screening

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2008;6: A Prospective, Multicenter, Population-Based Study of BRAF Mutational Analysis for Lynch Syndrome Screening XAVIER BESSA,* BELEN BALLESTÉ,* MONTSERRAT ANDREU,* ANTONI CASTELLS, BEATRIZ BELLOSILLO, FRANCESC BALAGUER, SERGI CASTELLVÍ BEL, ARTEMIO PAYA, RODRIGO JOVER, CRISTINA ALENDA, LLÚCIA TITÓ, # MERCEDES MARTINEZ VILLACAMPA,** ANGELS VILELLA, ROSA M. XICOLA, ELISENDA PONS, and XAVIER LLOR, FOR THE GASTROINTESTINAL ONCOLOGY GROUP OF THE SPANISH GASTROENTEROLOGICAL ASSOCIATION *Department of Gastroenterology, Department of Pathology, Hospital del Mar, Barcelona, Catalonia, Spain; Department of Gastroenterology, Institut de Malalties Digestives, Hospital Clínic, Centro de Investigación Biomédica en Red, Institut d Investigacións Biomèdiques August Pi I Sunyer, University of Barcelona, Barcelona, Catalonia, Spain; Department of Pathology, Department of Gastroenterology, Hospital General de Alicante, Alicante, Spain; # Department of Gastroenterology, Hospital de l Esperit Sant, Santa Coloma de Gramenet, Barcelona, Catalonia, Spain; **Hospital Comarcal de l Alt Penedès, Vilafranca del Penedes, Barcelona, Catalonia, Spain; Angels Vilella, Hospital Son Llatzer, Palma de Mallorca, Spain; and the Gastroenterology Department, Hospital Germans Trias i Pujol, Universitat Autonoma de Barcelona, Catalonia, Spain Background & Aims: Mismatch repair (MMR) deficiencies are the hallmark of tumors arising in Lynch syndrome, however, in approximately 15% of sporadic colorectal cancers (CRC) these deficiencies most often are associated with somatic methylation of the MMR gene MLH1. Recently, an oncogenic mutation in the BRAF gene has been involved in sporadic CRC showing MMR deficiencies as a result of MLH1 promoter methylation. The aim of this study was to evaluate the contribution of BRAF V600E mutation analysis in the identification of MSH2/MLH1 gene mutation carriers in newly diagnosed CRC patients. Methods: BRAF V600E mutation was analyzed in CRC patients with MMR deficiencies (microsatellite instability and/or lack of MLH1/MSH2 protein expression) in the EPICOLON population-based study. The effectiveness and efficiency of different strategies were evaluated with respect to the presence of MSH2/ MLH1 germline mutations. Results: MMR deficiencies were detected in 119 of the 1222 CRC patients with tumors showing either microsatellite instability (n 111) or loss of protein expression (n 81). BRAF mutation was detected in 22 (18.5%) of the patients. None of the patients with unambiguous germline mutation had BRAF mutation. Regardless of the strategy used to identify MSH2/MLH1 gene carriers, the introduction of BRAF analysis in these patients slightly improves their effectiveness. The introduction of BRAF mutation analysis as a step before germline genetic testing in patients with MMR deficiencies achieved a significant reduction in costs per mutation detected. Conclusions: Detection of BRAF V600E mutation could simplify and improve the cost effectiveness of genetic testing for hereditary nonpolyposis colorectal cancer, especially in patients whose family history is incomplete or unknown. Hereditary nonpolyposis colorectal cancer (HNPCC), also known as Lynch syndrome, is the most common type of hereditary colorectal cancer (CRC) and is caused by germline mutations in DNA mismatch repair (MMR) genes. 1,2 Although 7 genes have been associated with HNPCC (MSH2, MLH1, MSH6, PMS1, PMS2, MLH3, and EXO 1), mutations in only 3 of them currently are considered to cause Lynch syndrome: MLH1, MSH2, and MSH6. 3 These tumors are characterized by the presence of high levels of microsatellite instability (MSI) 4 ; however, the presence of MSI may be observed in up to 10% to 15% of sporadic CRCs. In sporadic CRCs the MSI phenotype is secondary to an epigenetic silencing of the promoter region of the MLH1 MMR gene. 5 8 Identification of patients with HNPCC has important clinical implications; in fact, in the past few years, it has been shown that CRC surveillance of HNPCC gene carriers is effective and considerably less costly than no colorectal surveillance. 9,10 Nevertheless, one of the most important limitations for the widespread implementation of genetic studies is their cost, not only in terms of laboratory expenses but also the need for timeconsuming interpretation of the sequence tracings. Although the most effective strategy for HNPCC diagnosis is the compilation of a thorough family history of CRC and fulfillment of the Amsterdam criteria for HNPCC, 11 about 20% of patients with germline mutations in MMR genes do not fulfill these strict criteria. 9,12 Recently, a second HNPCC workshop proposed a new set of recommendations, the revised Bethesda guidelines, for the identification of individuals with HNPCC who should be tested for microsatellite instability and subsequently by genetic testing. 13,14 Because microsatellite instability is a hallmark of tumors arising in the HNPCC syndrome, international criteria have been developed to combine clinical and molecular features to help in identifying patients with a high risk of HNPCC. 13 On the other hand, it is well known that most mutations in either MSH2 or MLH1 genes result in abnormal MSH2 or MLH1 protein expression on pathologic examination. As a consequence, immunostaining for these proteins in CRC specimens highly correlate with the presence of microsatellite instability 15,16 and may be another potential strategy for detecting HNPCC gene carriers. In fact, Abbreviations used in this paper: CRC, colorectal cancer; HNPCC, hereditary nonpolyposis colorectal cancer; MMR, mismatch repair; MSI, microsatellite instability; MSI-H, high-level microsatellite instability; MSI-L, low-level microsatellite instability by the AGA Institute /08/$34.00 doi: /j.cgh

2 February 2008 BRAF MUTATIONAL IN LYNCH SYNDROME SCREENING 207 protein immunostaining often is easier to perform in a clinical setting and the use of this technique appears to be an efficient HNPCC screening strategy. Recently a multicenter study published by our group 17 showed that the revised Bethesda guidelines provide a useful approach for identifying patients at risk for suffering HNPCC, and in patients who fulfill these criteria, both MSI testing and immunostaining are equivalent and highly effective strategies to further select those patients who should be tested for MSH2/ MLH1 germline mutations. The BRAF gene, 1 of 3 human isoforms of RAF, mediates cellular responses to grown signals through the RAS-RAF-MAP kinase pathway. A BRAF gene mutation, V600E, has been identified recently in 70% of malignant melanomas and in 5% to 15% of CRCs. 18,19 The BRAF V600E mutation has been identified in colorectal tumors showing MMR deficiencies associated with the epigenetic silencing of the MLH1 gene instead of germline mutations of MMR genes The results of these studies showed that patients with germline mutations in MMR genes never have a mutation in the BRAF gene. There is no such relationship between the BRAF mutation and other methylated genes, including MSH The incidence of BRAF mutations in unstable CRC tumors has been evaluated in selected groups of high-risk patients, and in 2 population-based samples of CRC patients. 27,28 However, these studies present some limitations that prevent us from drawing any definite conclusions about regarding the role of BRAF in CRC patients. Based on these data it has been suggested that the BRAF V600E mutation is characteristic of sporadic colorectal tumors with MSI. These data reinforce the concept that BRAF is not involved in the colorectal tumorigenesis of HNPCC, and suggest a potential use of BRAF in genetic testing for HNPCC as an exclusion criterion or as a molecular marker of sporadic cancer. Based on these data, it has been hypothesized that the determination of the BRAF V600E mutation as a prior step to further genetic studies in patients with putative MMR deficiencies (either microsatellite instability or lack of MLH1 protein immunostaining) will allow us to detect patients with MLH1 promoter hypermethylation, thus avoiding unnecessary genetic studies The aim of this study was to evaluate the contribution of the BRAF V600E mutation analysis in identifying MSH2/ MLH1 gene mutation carriers in newly diagnosed CRC patients by means of a prospective, multicenter, populationbased study in which both efficacy and efficiency were evaluated. Moreover, clinical-pathologic characteristics associated with the presence of the BRAF V600E mutation also were ascertained. Materials and Methods All patients with newly diagnosed CRC in 25 Spanish hospitals between November 1, 2000, and October 31, 2001, formed the basis of this study. These patients were part of the EPICOLON study, a prospective, multicenter, nationwide study that aimed to establish the incidence and characteristics of inherited and familial forms of CRC in Spain. 29 The study was approved by the institutional ethics committee of each participating hospital, and written informed consent was obtained from all patients. Demographic, clinical, and tumor-related characteristics of probands, as well as a detailed family history, were obtained using a pre-established questionnaire. Personal parameters at baseline included date and place of birth, sex, smoking history, personal history of neoplasm (ie, HNPCC-related tumors and colorectal adenomas), date of CRC diagnosis, presenting symptoms, serum carcinoembryonic antigen concentration, presence of synchronous colorectal neoplasms, and medications. Tumorrelated parameters included location, size, pathologic TNM stage, degree of differentiation, presence of lymphocytic infiltration, and mucinous carcinoma type (defined by the presence of 50% mucinous carcinoma cells). Pedigrees were traced backward and laterally as far as possible, or at least up to second-degree relatives, in terms of cancer history. Age at cancer diagnosis, type, location, and tumor stage of the neoplasm and current status were recorded for each affected family member. As a check on overreporting of disease, all patients were asked about the occurrence of stroke in their family to assess possible recall bias. Furthermore, an effort was made to verify reported neoplasm in relatives by obtaining medical records and pathology confirmation. The definition of HNPCC was based on Amsterdam II criteria. 11 HNPCC-associated neoplasms included CRC, cancer of the endometrium, small bowel, ureter and renal pelvis, ovary, stomach, and hepatobiliary system according to the revised Bethesda guidelines. 14 Analysis of Tumor Microsatellite Instability and MSH2/MLH1 Protein Expression Microsatellite instability testing using the 5-marker panel proposed by the National Cancer Institute and immunostaining for DNA mismatch repair proteins (MLH1 and MSH2) were performed in all cases regardless of age, personal or family history, and tumor characteristics. 30,31 Tumors with greater than 30% unstable markers were classified as high-level MSI (MSI-H) and those with 30% or less unstable markers were classified as low-level MSI (MSI-L). If none of the markers showed instability, tumors were classified as stable. Finally, immunohistochemistry for MSH6 was evaluated in HNPCC tumors (fulfillment of the Amsterdam I and II criteria), MSI-H tumors with normal MLH1 and MSH2 expression, and MSI-L tumors. Patients found to have tumors with MSI and/or lack of protein expression of either MSH2 or MLH1 underwent germline genetic testing for MSH2 and MLH1 genes by both multiple ligation probe amplification analysis and sequencing. Methodologic techniques are described elsewhere. 17,30 With this approach, 11 (0.9%) germline mutations in either MSH2 (7 cases) or MLH1 (4 cases) were detected. 17 Eight mutations were considered deleterious according to sequencing results, published data, and existing mutation databases. On the other hand, however, 3 of these DNA changes corresponded to missense mutations of uncertain significance. In fact, 1 tumor did not show MSI, 2 tumors showed contradictory immunostaining results, and 1 patient did not fulfill the revised Bethesda guidelines. This analysis was performed considering only unambiguous proven pathogenic variants. BRAF V600E Mutation Analysis The BRAF V600E mutation analysis was performed in all patients with MMR deficiencies, either MSI and/or lack of MLH1/MSH2 protein expression.

3 208 BESSA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 2 Table 1. Characteristics of CRC Patients With Mismatch Repair Deficiencies (n 119) Age 50, y 11 (9.2%) Sex, male/female 47/72 Right site location a 73% (61.3%) TNM stage (I/II/III/IV) b 7/67/32/13 Undifferentiated tumor 18 (15.1%) Lymphocytic infiltration 18 (15.1%) Mucinous carcinoma c 26 (21.8%) Synchronous CRC 8 (6.7%) Fulfillment of guidelines Revised Bethesda guidelines d 44 (37.0%) Amsterdam I criteria 10 (8.4%) Extended Amsterdam II criteria 11 (9.2%) Microsatellite instability MSI-H 83 (69.7%) MSI-L 28 (23.5%) Loss of protein expression MSH2 protein expression 21 (21%) MLH1 protein expression 60 (50.7%) BRAF mutation 22 (18.5%) Germline unambiguos mutation MLH1 3 MSH2 5 NOTE. Data are numbers with percentages in parentheses. a Proximal to splenic flexure. b According to TNM classification (American Joint Committee on Cancer). c Composed of 50% mucinous carcinoma cells. d For an explanation of Amsterdam and extended Amsterdam II criteria, see Vasen et al 10 ; and for revised Bethesda guidelines, see Umar et al. 12 Analysis of the BRAF V600E was performed by direct sequencing using tumor DNA. Primers were designed using the Primer Express software (Applied Biosystems, Foster City, CA) and were as follows: forward: 5=-CTCTTACCTAAACTCT- TCATAATGCTTGC -3= and reverse: 5=-CAGCATCTCAGGGC- CAAAAA -3=. Sequencing was performed with BigDye v3.1 (Applied Biosystems) following the manufacturer s instructions and analyzed on an ABI3100 Sequencer (Applied Biosystems). Statistical Analysis Associations between the BRAF V600E mutation and clinical, pathologic, or molecular features were evaluated using the Fisher exact or Pearson chi-squared tests, as appropriate. A P value of less than.05 was considered to indicate a statistically significant difference. Performance characteristics of screening strategies based on the BRAF V600E mutation analysis, either directly or through previous selection of patients according to the revised Bethesda guidelines, were calculated with respect to the diagnosis of HNPCC associated with MSH2/MLH1 germline mutations. Statistical analyses were performed using SPSS version 12.0 (SPSS Inc, Chicago, IL). A cost-minimization analysis was performed to determine the efficiency of BRAF mutational analysis in detecting all unambiguous germline mutations. For this analysis, costs were calculated according to our hospital s current billing. Costs of MSI, immunostaining (both MSH2 and MLH1), the BRAF V600E mutation, and genetic testing were established at 100 (136 US$), 200 (272 US$), 100 (136 US$), and 2400 (3259 US$), respectively, or 1200 (1630 US$) per gene. Results During the study period, 1222 patients with pathologically confirmed colorectal adenocarcinoma were diagnosed and included in the EPICOLON project. Of these, 111 (9.1%) showed microsatellite instability, 83 (6.8%) had MSI-H and 28 (2.3%) had MSI-L, respectively. In addition, 81 (6.6%) had a loss of protein expression in either MSH2 (21 cases) or MLH1 (60 cases). None of the patients with MSI-L had a lack of protein expression of MMR genes. On the other hand, 10 patients had MSI-H without a lack of protein expression and 8 microsatellite-stable tumors had a loss of protein expression. Accordingly, 119 patients had MMR deficiencies, in either MSI and/or protein immunostaining, and formed the basis of this study. Baseline patient characteristics are summarized in Table 1. The BRAF V600E mutation was observed in 20 (24.1%) of the patients with MSI-H and in 18 of the 60 patients (30.0%) with loss of MLH1 protein expression on pathologic examination. No patients with loss of MSH2 protein expression showed the BRAF V600E mutation and only 2 (7.1%) of the MSI-L patients had the BRAF mutation. Overall, the BRAF V600E mutation was detected in 22 (18.5%) of the 119 patients with MMR deficiencies. Only 1 patient with an MSI-H tumor did not express MSH6 protein, and the BRAF V600E mutation was not detected in this patient. Regarding molecular diagnosis of HNPCC, none of the patients with unambiguous germinal mutations in DNA MMR genes had the mutation (Table 2). Table 2. Characteristics of Patients With MSH2/MLH1 Germline Mutations Fullfillment of guidelines IHC expression of protein Case Amsterdam II criteria Revised Bethesda guidelines MSI status MSH2 MLH1 Gene BRAF mutant 1 Yes Yes Unstable Present Absent MLH1 Wild-type 2 Yes Yes Unstable Present Absent MLH1 Wild-type 3 Yes Yes Unstable Absent Present MSH2 Wild-type 4 Yes Yes Unstable Absent Present MLH1 Wild-type 5 No Yes Unstable Present Absent MSH2 Wild-type 6 No Yes Unstable Absent Present MSH2 Wild-type 7 No Yes Unstable Absent Present MSH2 Wild-type 8 No Yes Unstable Absent Present MSH2 Wild-type

4 February 2008 BRAF MUTATIONAL IN LYNCH SYNDROME SCREENING 209 Table 3. Performance Characteristics of Different Strategies for Identifying MSH2/MLH1 Gene Mutation Carriers Unambiguous germline mutations (n 8) Strategy Sensitivity Specificity Positive predictive value Negative predictive value Overall accuracy Presence of MSI-H Presence of MSI-H and BRAF analysis a a Loss of protein expression Loss of protein expression and BRAF a a analysis b Fulfillment of revised Bethesda guidelines Fulfillment of revised Bethesda guidelines and presence of MSI-H Presence of MSI-H and BRAF analysis Loss of protein expression Loss of protein expression and BRAF analysis NOTE. Results expressed as a percentage. a P.01 (McNemar test) with respect to the corresponding paired strategy without BRAF V600E mutation analysis. b BRAF analysis was performed only in patients with loss of MLH1 protein expression. Performance Characteristics and Efficiency of Screening Strategies Based on the BRAF V600E Mutation Analysis Performance characteristics of screening strategies for identifying MSH2/MLH1 gene carriers are shown in Table 3. Regardless of the strategy used for identifying MSH2/MLH1 gene mutation, the introduction of the BRAF V600E analysis in these patients improved their effectiveness. In this regard, the introduction of the BRAF V600E mutation analysis as a step before germline genetic testing on all CRCs without previous selection according to Bethesda guidelines significantly increased specificity and overall accuracy. Efficiency for the introduction of the BRAF V600E analysis into genetic testing for unambiguous germline mutations in MLH1 and MSH2 genes was evaluated in a cost-minimization analysis. Clinical selection of patients according to the revised Bethesda guidelines combined with MSI analysis or protein immunostaining has a high cost (13,488 and 11,825 per detected mutation, respectively). Inclusion of the BRAF V600E mutation analysis as a step before germline testing in MSI CRC patients allowed us to avoid germline testing in only 4 patients, those with BRAF mutation, achieving a small reduction in costs (a savings of 788 per detected mutation) (Figure 1). If the presence of mismatch repair deficiencies has been evaluated by protein immunostaining, the inclusion of the BRAF V600E mutation in patients with a loss of MLH1 protein expression also detects 4 patients with BRAF mutations, also leading to a small reduction in costs (a savings of 375 per detected mutation) (Figure 1). On the other hand, if the family history is not available or a universal approach to screening all CRC patients regardless of their personal and family history was used, the direct determination of MSI or protein immunostaining in all CRC patients had a cost of 40,175 and 42,700 per detected mutation, respectively. The introduction of BRAF analysis to all MSI CRC patients enabled us to detect 20 patients with the BRAF V600E mutation, avoiding Figure 1. Costs of BRAF V600E analysis per MMR detected mutation according to the revised Bethesda guidelines.

5 210 BESSA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 2 Figure 2. Costs of BRAF V600E analysis per MMC detected mutation for all CRC patients. any other unnecessary genetic tests in those patients. This strategy achieved a significant reduction in costs (a savings of 4963 per detected mutation) (Figure 2). If the presence of mismatch repair deficiencies had been evaluated by protein immunostaining, the inclusion of the BRAF V600E mutation to patients with a loss of MLH1 protein expression detected 18 patients with the BRAF V600E mutation, also leading to a substantial reduction in costs per detected mutation (a savings of 1688 per detected mutation) (Figure 2). Clinical-Pathologic Characteristics Associated With the Presence of the BRAF V600E Mutation in Mismatch Repair Deficient Tumors Regarding the individual characteristics associated with the BRAF mutation, only an older age at diagnosis was related to the BRAF V600E mutation (P.002). None of the other evaluated characteristics (synchronous or metachronous CRC and personal history of previous CRC) were related to the presence of a BRAF mutation. In terms of tumor characteristics, only right-side tumor location was related significantly (P.001) to the BRAF V600E mutation. Neither lymphocytic infiltration, nor mucinous carcinoma type, nor poor histologic grade were associated with the BRAF mutation (Table 4). It is noteworthy that none of the patients with the BRAF V600E mutation fulfilled the Amsterdam I or extended Amsterdam II criteria for HNPCC. Only 4 (11%) of the patients with at least one of the criteria of the revised Bethesda guidelines had a BRAF mutation. Finally, concerning a family history of cancer, only a family history of HNPCC-related cancer in a first-degree relative younger than 50 years was related negatively (P.04) to the BRAF V600E mutation (Table 4). Discussion We attempted to determine the efficacy and efficiency of introduction of the BRAF V600E mutation analysis in the algorithm for the identification of MLH1 and MSH2 gene carriers in unselected CRC patients. This study confirms the usefulness of the BRAF V600E mutation gene analysis in the screening strategies for the detection of HNPCC patients in a population-based approach. It has been established that genetic testing is necessary for identifying individuals with HNPCC because an exhaustive surveillance and preventive program for relatives at risk was effective and allows us to optimize its surveillance and is considerably less costly than no CRC surveillance. 9,10 Overall, 2 strategies have been suggested for MLH1/MSH2 germline mutation detection in CRC patients. The first one, suggested in the second Bethesda conference and confirmed by our group, 17 is based on clinical selection of all CRC patients depending on whether they fulfill the revised Bethesda guidelines before molecular studies. The second strategy suggests that routine molecular screening, by determining the MSI status, would be applied to all CRC patients, regardless of the personal and family history, in whom it would not otherwise have been detected. 32 The application of molecular screening to all CRC patients allows us to detect patients with known germline mutation who do not fulfill the Bethesda guidelines, as suggested by a previous study. 32 Moreover, based on the fact that patients with MSI 33 or lack of MLH1/MSH2 protein expression 34 have a more favorable prognosis, regardless of the stage at diagnosis, and do not benefit from adjuvant chemotherapy with fluorouracil, the authors 33,34 suggest that there is a need to determine the MSI status in all CRC patients. In the past few years, a mutation of the BRAF gene V600E has been identified in patients with CRC, and recent data provided evidence that this mutation is not involved in tumors from HNPCC patients with germline mutations in MLH1 and MSH2, and often is mutated in association with methylation of the MLH1 gene promoter Because of its absence in patients with MLH1/MSH2 germline mutation, a potential use in the molecular diagnostics of HNPCC as a step before germline genetic testing has been suggested. 21,24 Recently, Bettstetter et al, 35 using a reliable and quantitative MLH1 methylation analysis technique to define valid MLH1 methylation cut-off values for HNPCC diagnostics, showed that BRAF V600E mutation virtually did not occur in HNPCC patients carrying pathogenic germline MLH1 mutations and, vice versa, BRAF V600E mutation occurred specifically in a subset of sporadic MSI-H CRC. These data suggest that detection of BRAF V600E mutation is a highly specific molecular marker to exclude

6 February 2008 BRAF MUTATIONAL IN LYNCH SYNDROME SCREENING 211 Table 4. Individual, Tumor-Related and Family-Related Characteristics Associated With the Presence of BRAF Mutations in Patients With Mismatch Repair Deficiencies (MSI and/or Abnormal Immunohistochemistry for MLH1/MSH2) (n 119) BRAF mutant (n 22) BRAF wild-type (n 97) P Age 50 y Male sex 7 (31.8%) 40 (41.2%).47 Personal history of any neoplasia 1 (4.5%) 16 (16.5%).19 Personal history of colorectal cancer 0 (0%) 2 (2.1%).49 Synchronous or metachronous CCR 2 (9.1%) 8 (8.2%) 1.0 TNM stage (I/II/III/IV) a 0/14/6/2 7/53/26/11.63 CEA level 4 ng/ml 8 (36.4%) 33 (34.0%).79 Right site tumor location b 20 (90.9%) 53 (54.6%).001 Undifferentiated tumor 5 (22.7%) 13 (13.4%).42 Lymphocytic infiltration 5 (22.7%) 13 (13.4%).33 Mucinous carcinoma c 5 (22.7%) 21 (21.6%) 1.0 Fulfilment of guidelines d Revised Bethesda guidelines 4 (18.2%) 40 (41.2%).052 Amsterdam I 0 (0%) 10 (10.3).20 Extended Amsterdam II criteria 0 (0%) 11 (11.3%).21 First-degree relatives with CRC 1 (4.5%) 20 (20.6%).12 First-degree relatives with CRC 50 y 0 (0%) 12 (12.4%).12 First-degree relatives with endometrial cancer 1 (4.5%) 5 (5.2%) 1.0 First-degree relatives with endometrial cancer 50 y 0 (0%) 3 (3.1%) 1.0 Family history of HNPCC-related cancer in first-degree relative 2 (9.1%) 24 (24.7%).15 Family history of HNPCC-related cancer in first-degree relative 50 y 0 (0%) 17 (17.5%).04 Family history of HNPCC-related cancer (excluding CRC) in first-degree relative 1 (4.5%) 4 (4.1%) 1.0 Family history of HNPCC-related cancer (excluding CRC) in first-degree relative 50 y 0 (0%) 3 (3.1%) 1.0 NOTE. Data are numbers with percentages in parentheses. CEA, carcinoembryonic antigen. a According to TNM classification (American Joint Committee on Cancer, 5th ed). b Proximal to splenic flexure. c Composed of 50% mucinous carcinoma cells. d For an explanation of Amsterdam and extended Amsterdam II criteria, see Vasen et al 10 ; and for revised Bethesda guidelines, see Umar et al. 12 HNPCC and it should be used in parallel with quantitative MLH1 methylation analysis. The results of our study confirm that, regardless of the aforementioned screening strategy used, the introduction of the BRAF V600E mutation analysis in patients with mismatch repair deficiencies as a step before germline genetic testing improves the efficacy and efficiency of identifying MLH1 and MSH2 gene carriers in CRC patients. The strength of our study lies in the fact that it was performed on a general population basis and our series involves one of the most complete populations of CRC patients in which MSI and protein immunostaining were performed in a parallel and blinded fashion. It should be noted that our study was a population-based study evaluating the incidence of the BRAF V600E mutation in patients with MMR deficiencies based on protein immunostaining as the method of choice to direct the screening for germline mutations 2,15,16 and current data suggest that nearly all HNPCC cases involving MSH2 and MSH6 arise as a consequence of germline mutation Therefore, according to our study (no BRAF V600E mutation in patients with a lack of MSH2 or MSH6 protein expression), direct germline testing should be offered only to patients with a lack of MSH2 or MSH6 protein expression on pathologic examination. More importantly, to improve the effectiveness of genetic studies in tumors that do not stain for MLH1, it is necessary to determine the BRAF status to avoid unnecessary genetic studies (Figures 1 and 2). Based on the results of this study as well as in other reports, 21,24,35 we propose an algorithm for MSH2/MLH1 genetic testing in patients with CRC (Figure 3). Because HNPCC is caused mainly by germline mutations in MLH1 and MSH2 genes, 1,2 and because MSH6 protein expression was not evaluated in all patients in our series, MSH6 immunostaining was not incorporated in the algorithm. Similarly, PMS2 analysis was not included either. On the other hand, if MMR deficiencies are still looked for in any given CRC, protein immunostaining followed by BRAF analysis in patients with a lack of MLH1 protein expression continues to be the most effective strategy. In the past decade, it has been shown extensively that a subset of CRCs show promoter methylation in multiples genes, including p16, p14, MGMT, and MLH1. This phenomenon has been referred to as the CpG island methylator phenotype. Transcriptional inactivation by cytosine methylation at promoter CpG island of tumor-suppressor genes is thought to be an important mechanism in human carcinogenesis and it has been shown to occur early in CRC carcinogenesis. The very high frequency of BRAF mutations in both CRC with MSI and/or extensive DNA methylation, and in the colorectal polyps with serrated morphology, strongly supports the development of sporadic MSI-H CRC through an alternative, so-called serrated pathway. 36 This study aimed to establish the relationship between a complete personal and familial tumor history with the presence of BRAF mutations in MSI CRC patients. The strength of this study lies mainly in the large number of patients included in a prospective manner and the comprehensive information gathered with respect to familial characteristics. Concerning per-

7 212 BESSA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 2 Figure 3. Proposal for MSH2/MLH1 genetic testing in patients with colorectal cancer. sonal, familial, and tumor-related characteristics associated with the presence of BRAF mutation, we observed a slightly lower incidence of the BRAF V600E mutation in MSI CRC in relation to other studies, one of them being a single-center study including a low number of MSI CRC patients 28 and the other 29 with a preselection according to tumor location and an age-limiting cut-off at inclusion. Our results are based on one of the largest multicenter population-based series that recruited all newly diagnosed CRC patients, regardless of tumor location and age at diagnosis, and with the highest number of patients with MMR deficiencies, probably more accurately reflecting the real incidence of the BRAF V600E mutation in MSI CRC patients. In this study, the BRAF V600E mutation was associated strongly only with an older age at diagnosis and proximal tumor location. Moreover, in our study the BRAF V600E mutation was related negatively only to the presence of a family history of HNPCC-related cancer in a first-degree relative younger than 50 years, and although it was not statistically significant it was probably because of the low number of evaluated patients with a family history of CRC. These results match with a noninherited epigenetic silencing of MLH1 promoter hypermethylation. As expected, no MSI CRC patients fulfilling Amsterdam I or extended Amsterdam II criteria for HNPCC had BRAF mutations. The aforementioned previous population studies have shown contradictory results in relation to individual and tumor-related characteristics. Li et al 28 showed an association between the BRAF V600E mutation and lymphocytic infiltration, proximal tumor location, poor differentiation, and mucinous histology on pathologic examination. In contrast, a larger population-based study 29 only showed that MSI CRC with the BRAF V600E mutation was associated with an older age at diagnosis and when they evaluated only family history of CRC, they observed a relationship between BRAF mutations and family history of colon cancer, although this relationship was only with microsatellite-stable tumors. In summary, this large population-based study showed that the BRAF V600E mutation analysis as a step before germline testing could simplify and improve the efficacy and efficiency of HNPCC gene testing, especially if the screening strategy is extended to any patient with CRC. Appendix Gastrointestinal Oncology Group of the Spanish Gastroenterological Association Investigators and Study Organization. Writing Committee and Working Group: Antoni Castells (co-chair) and Montserrat Andreu (co-chair). Participants: Hospital 12 de Octubre (Madrid, Spain): Juan Diego Morillas (local coordinator), Raquel Muñoz, Marisa Manzano, Francisco Colina, Jose Díaz, Carolina Ibarrola, Guadalupe López, and Alberto Ibáñez; Hospital Clínic (Barcelona, Spain): Antoni Castells (local coordinator), Virgínia Piñol, Sergi Castellví-Bel, Francisco Rodríguez-Moranta, Francesc Balaguer, Antonio Soriano, Rosa Cuadrado, Maria Pellisé, Rosa Miquel, J. Ignasi Elizalde, and Josep M. Piqué; Hospital Clínico Universitario (Zaragoza, Spain): Ángel Lanas (local coordinator), Javier Alcedo, and Javier Ortego; Hospital Cristal-Piñor, Complexo Hospitalario de Ourense (Ourense, Spain): Joaquin Cubiella (local coordinator), Maria Soledad Díez, Mercedes Salgado, Eloy Sánchez, and Mariano Vega; Hospital del Mar (Barcelona, Spain): Montserrat Andreu (local coordinator), Xavier Bessa, Agustín Panadés, Mar Iglesias, Felipe Bory, Beatriz Belosillo, and Agustín Seoane; Hospital Donosti (San Sebastián, Spain): Luis Bujanda (local coordinator), Juan Ignacio Arenas, Isabel Montalvo, Julio Torrado, and Ángel Cosme; Hospital General Universitario de Alicante (Alicante, Spain): Artemio

8 February 2008 BRAF MUTATIONAL IN LYNCH SYNDROME SCREENING 213 Payá (local coordinator), Rodrigo Jover, Juan Carlos Penalva, and Cristina Alenda; Hospital General de Granollers (Granollers, Spain): Joaquim Rigau (local coordinator), Ángel Serrano, and Anna Giménez; Hospital General de Vic (Vic, Spain): Joan Saló (local coordinator), Eduard Batiste-Alentorn, Josefina Autonell, and Ramon Barniol; Hospital General Universitario de Guadalajara (Guadalajara, Spain): Ana María García (local coordinator), Fernando Carballo, Antonio Bienvenido, Eduardo Sanz, Fernando González, and Jaime Sánchez; Hospital General Universitario de Valencia (Valencia, Spain): Enrique Medina (local coordinator), Jaime Cuquerella, Pilar Canelles, Miguel Martorell, José Ángel García, Francisco Quiles, and Elisa Orti; Hospital do Meixoeiro (Vigo, Spain): Juan Clofent (local coordinator), Jaime Seoane, Antoni Tardío, and Eugenia Sanchez; Hospital San Eloy (Baracaldo, Spain): Luis Bujanda (local coordinator), Carmen Muñoz, María del Mar Ramírez, and Araceli Sánchez; Hospital Universitari Germans Trias i Pujol (Badalona, Spain): Xavier Llor (local coordinator), Rosa M. Xicola, Marta Piñol, Mercè Rosinach, Anna Roca, Elisenda Pons, José M. Hernández, and Miquel A. Gassull; Hospital Universitari Mútua de Terrassa (Terrassa, Spain): Fernando Fernández- Bañares (local coordinator), Josep M. Viver, Antonio Salas, Jorge Espinós, Montserrat Forné, and Maria Esteve; Hospital Universitari Arnau de Vilanova (Lleida, Spain): Josep M. Reñé (local coordinator), Carmen Piñol, Juan Buenestado, and Joan Viñas; Hospital Universitario de Canarias (Tenerife, Spain): Enrique Quintero (local coordinator), David Nicolás, Adolfo Parra, and Antonio Martín; Hospital Universitario La Fe (Valencia, Spain): Lidia Argüello (local coordinator), Vicente Pons, Virginia Pertejo, and Teresa Sala; Hospital Universitario Reina Sofía (Córdoba, Spain): Antonio Naranjo (local coordinator), María del Valle García, Patricia López, Fernando López, Rosa Ortega, Javier Briceño, and Javier Padillo; and Fundació Hospital Son Llatzer (Palma de Mallorca, Spain): Àngels Vilella (local coordinator), Carlos Dolz, and Hernan Andreu. References 1. Rustgi AK. Hereditary gastrointestinal polyposis and nonpolyposis syndromes. N Engl J Med 1994;331: Giarddiello FM, Brensinger JD, Petersen GM. AGA technical review on hereditary colorrectal cancer and genetic testing. Gastroenterology 2001;121: Lagerstedt Robinson K, Liu T, Vandrovcova J, et al. Lynch syndrome (hereditary nonpolyposis colorectal cancer) diagnostics. J Natl Cancer Inst 2007;99: Boland CR, Thibodeau SN, Hamilton SR, et al. 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 1998;58: Miyakura Y, Sugano K, Konishi F, et al. Extensive methylation of MLH1 promoter region predominates in proximal colon cancer with microsatellite instability. Gastroenterology 2001;121: Herman JG, Umar A, Polyak K, et al. Incidence and functional consequences of MLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA1998;95: Cunningham JM, Christensen ER, Tester DJ, et al. Hypermethylation of the HMLH1 promoter in colon cancer with microsatellite instability. Cancer Res 1998;58: Toyota M, Ahuja N, Ohe-Toyota M, et al. CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci USA1999; 96: Reyes CM, Allen BA, Terdiman JP, et al. Comparison of selection strategies for genetic testing of patients with hereditary nonpolyposis colorectal carcinoma: effectiveness and cost-effectiveness. Cancer 2002;95: Vasen HF, Van Ballegooijen M, Buskens E, et al. A cost-effectiveness analysis of colorectal screening of hereditary nonpolyposis colorectal carcinoma gene carriers. Cancer 1998;82: Vasen HF, Watson P, Mecklin JP, et al. New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology 1999;116: Umar A, Risinger JJ, Hawk ET, et al. Testing guidelines for hereditary non-polyposis colorectal cancer. Nat Rev Cancer 2004;4: Rodriguez-Bigas MA, 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: Umar A, Boland CR, Terdiman JP, et al. Revised Bethesda guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst 2004; 96: Terdiman JP, Gum JR Jr, Conrad PG, et al. Efficient detection of hereditary nonpolyposis colorectal cancer gene carriers by screening for tumor microsatellite instability before germline genetic testing. Gastroenterology 2001;120: Thibodeau SN, French AJ, Cunningham JM, et al. Altered expression of MSH2 and MLH1 in tumors with microsatellite instability and genetic alterations in mismatch repair genes. Cancer Res 1996;56: Pinol V, Castells A, Andreu M, et al. Gastrointestinal oncology group of the Spanish Gastroenterological Association. Accuracy of revised Bethesda guidelines, microsatellite instability and immunohistochemistry for the identification of patients with nonpolyposis colorectal cancer. JAMA 2005;293: Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417: Rajagopalan H, Bardelli A, Lengauer C, et al. Tumorigenesis: RAF/RAS oncogenes and mismatch-repair status. Nature 2002; 418: Wang L, Cunningham JM, Winters JL, et al. BRAF mutations in colon cancer are not likely attributable to defective DNA mismatch repair. Cancer Res 2003;63: Domingo E, Laiho P, Ollikainen M, et al. BRAF screening as a low-cost effective strategy for simplifying HNPCC genetic testing. J Med Genet 2004;41: Mc Givern A, Wynter CVA, Whitehall VLJ, et al. Promoter hypermethylation frequency and BRAF mutations distinguish hereditary non-polyposis colon cancer from sporadic MSI-H colon cancer. Familial Cancer 2004;3: Lumbomierski N, Plotz G, Wormek M, et al. BRAF mutations in colorectal carcinoma suggest two entities of microsatellite-unstable tumors. Cancer 2005;104: Domingo W, Niessen RC, Oliveira C, et al. BRAF V600E is not involved in the colorectal tumorigenesis of HNPCC in patients with functional MLH1 and MSH2 genes. Oncogene 2005;24: Young J, Barker MA, Simms LA, et al. Evidence for BRAF mutation and variable levels of microsatellite instability in a syndrome of familial colorectal cancer. Clin Gastroenterol Hepatol 2005; 3: Deng G, Bell I, Crawley S, et al. BRAF mutation is frequently present in sporadic colorectal cancer with methylated hmlh1, but not in hereditary nonpolyposis colorectal cancer. Clin Cancer Res 2004;10: Li WQ, Kawakami K, Ruszkiewicz A, et al. BRAF mutations are associated with distinctive clinical, pathological and molecular

9 214 BESSA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 2 features of colorectal cancer independently of microsatellite instability status. Mol Cancer 2006;5: Samowitz W, Sweeney C, Herrick J, et al. Poor survival associated with the BRAF V600E mutation in microsatellite-stable colon cancers. Cancer Res 2005;65: Pinol V, Andreu M, Castells A, et al. Gastrointestinal Oncology Group of the Spanish Gastroenterological Association. Frequency of hereditary nonpolyposis colorectal cancer and other colorectal cancer familial forms in Spain: a multicentre, prospective, nationwide study. Eur J Gastroenterol Hepatol 2004;10: Llor X, Pons W, Xicola RM, et al, Gastrointestinal Oncology Group of the Spanish Gastroenterological Association. Differential features of colorectal cancers fulfilling Amsterdam criteria without involvement of the mutator pathway. Clin Cancer Res 2005;11: Xicola RM, Llor X, Pons E, et al, Gastrointestinal Oncology Group of the Spanish Gastroenterological Association. Performance of different microsatellite marker panels for detection of mismatch repair-deficient colorectal tumors. J Natl Cancer Inst 2007;99: Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med 2005;352: Jover R, Zapater P, Castells A, et al, Gastrointestinal Oncology Group of the Spanish Gastroenterological Association. Mismatch repair status in the prediction of benefit from adjuvant fluorouracil chemotherapy in colorectal cancer. Gut 2006;55: Lanza G, Gafa R, Santini A, et al. Immunohistochemical test for MLH1 and MSH2 expression predicts clinical outcome in stage II and III colorectal cancer patients. J Clin Oncol 2006;24: Bettstetter M, Dechant S, Ruemmele P, et al. Distinction of hereditary nonpolyposis colorectal cancer and sporadic microsatellite-unstable colorectal cancer through quantification of MLH1 methylation by real-time PCR. Clin Cancer Res 2007;13: Jass JR. CRC a multipathway disease. Crit Rev Oncog 2006;12: Address requests for reprints to: Xavier Bessa Caserras, MD, Gastroenterology Department, Hospital del Mar, Passeig Maritim 25-29, 08003Barcelona,Catalonia,Spain @imas.imim.es;fax: (34) Supported by grants from Fondo de Investigación Sanitaria (PI051875, PI061384) and Instituto de Salud Carlos III (C03/02). X. B. and B. B. contributed equally to this work. Current affiliation for E.P. and X.L.: Digestive Diseases and Nutrition Section, University of Illinois Chicago, Chicago, Illinois. For a listing of the members of the Gastrointestinal Oncology Group of the Spanish Gastroenterological Association, see the Appendix.