Prevalence of Lynch Syndrome in a Middle Eastern Population With Colorectal Cancer

Transcription

1 Original Article Prevalence of Lynch Syndrome in a Middle Eastern Population With Colorectal Cancer Abdul K. Siraj, PhD 1 ; Sarita Prabhakaran, MD 1 ; Prashant Bavi, MD 1 ; Rong Bu, MDPhD 1 ; Shaham Beg, MD 1 ; Mohsen Al Hazmi, MBBS 1 ; Maha Al-Rasheed, BSc 1 ; Mohammed Al-Assiri, MD 2 ; Rami Sairafi, MD 2 ; Fouad Al-Dayel, MD 3 ; Nasser Al-Sanea, MD 4 ; Shahab Uddin, PhD 1 ; and Khawla S. Al-Kuraya, MD, FCAP 1,5 BACKGROUND: Lynch syndrome (LS; hereditary nonpolyposis colorectal cancer) is a common cause of hereditary colorectal cancer (CRC). CRC is the most common cancer diagnosed among males in Saudi Arabia but to the authors knowledge there is a lack of data regarding the prevalence of LS in patients with CRC. There currently are no clear guidelines for the selection criteria for these patients to screen for LS. METHODS: A comprehensive molecular characterization was performed in a cohort of 807 CRC cases by immunohistochemical and microsatellite analysis using polymerase chain reaction. BRAF mutation screening, high CpG island methylator phenotype, and analysis for germline mutations were performed in 425 CRC samples. These were all high microsatellite instability (MSI-H) samples (91 cases), all low MSI samples (143 cases), and selected cases from the microsatellite stable group (191 cases) that met revised Bethesda guidelines. RESULTS: Polymerase chain reaction identified 91 MSI-H cases (11.3%) and sequencing revealed mismatch repair germline mutations in 8 CRC cases only. Of the total of 807 CRC cases, these 8 cases (0.99%) were MSI-H, met the revised Bethesda guidelines, and did not harbor BRAF mutations. CONCLUSIONS: The results of the current study confirmed cases of LS in approximately 1.0% of CRC samples and reflects the efficacy of screening among MSI-H cases that lack BRAF mutations. This comprehensive study from Saudi Arabia will help in implementing a universal screening/reflex testing strategy in a clinical setting in Saudi Arabia and in conducting a national screening program that benefits both patients and their relatives. Cancer 2015;121: VC 2015 American Cancer Society. KEYWORDS: colorectal cancer (CRC), hereditary nonpolyposis colorectal cancer (HNPCC), Lynch syndrome (LS), microsatellite instability (MSI), mismatch repair (MMR) genes, mutl homolog 1 (MLH1), muts protein homolog 2 (MSH2), muts protein homolog 6 (MSH6), postmeiotic segregation increased 2 (PMS2). INTRODUCTION The morbidity and mortality caused by colorectal cancer (CRC) is of major concern throughout the world. It is the third most common cancer and the second leading cause of cancer-related deaths around the world. 1 The incidence of CRC is rising among the Arab population. Reports from Saudi Arabia suggest it represents a more aggressive disease. 2,3 Lynch syndrome (LS), or hereditary nonpolyposis colorectal cancer (HNPCC), is responsible for 1% to 3% of all cases of CRC. Germline mutations in DNA mismatch repair (MMR) genes (ie, mutl homolog 1 [MLH1], muts protein homolog 2 [MSH2], muts protein homolog 6 [MSH6], and postmeiotic segregation increased 2 [PMS2]) cause this autosomal dominant disease. Microsatellite instability (MSI) or the immunohistochemical losses of MMR protein expression in tumors of patients are the common hallmarks of MMR deficiency in patients with LS. 4,5 Deficient MMR and MSI arise due to germline mutations in MMR genes or, more commonly, from somatic hypermethylation of CpG islands surrounding the promoter region of MLH1 and other genes, which is known as the CpG island methylator phenotype (CIMP). 6 The incidence of BRAFV600E mutations in CRC cases with sporadic MSI with epigenetic inactivation of MLH1 is observed to be significantly higher (approximately 50%) compared with an overall BRAF mutation frequency of Corresponding author: Khawla S. Al-Kuraya, MD, FCAP, Human Cancer Genomic Research, Research Center, King Faisal Specialist Hospital and Research Centre, MBC#98-16, PO Box 3354, Riyadh 11211, Saudi Arabia; Fax: (011) ; kkuraya@kfshrc.edu.sa 1 Human Cancer Genomic Research, Research Center, King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia; 2 Department of Surgery, Security Forces Hospital, Riyadh, Saudi Arabia; 3 Department of Pathology, King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia; 4 Colorectal Unit, Department of Surgery, King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia; 5 College of Medicine, Alfaisal University, Riyadh, Saudi Arabia We thank Mehar Sultana, Valorie Balde, Padmanaban Annaiyappanaidu, Khadija Abdul Rahman Alobaisi, Wael Haqqawi, and Hassan Al Dossarie for technical assistance and Zeeshan Qadri for data analysis. DOI: /cncr.29288, Received: October 7, 2014; Revised: December 30, 2014; Accepted: January 8, 2015, Published online February 24, 2015 in Wiley Online Library (wileyonlinelibrary.com) 1762 Cancer June 1, 2015

2 Lynch Syndrome in CRC in Saudi Arabia/Siraj et al only 8% to 11% among all cases of sporadic CRC. 7 The presence of a BRAF mutation favors a diagnosis of a CRC with sporadic MSI. LS tumors exhibit an increased risk of CRC and other cancers. 8 CRC develops at a young age, with an accelerated rate of carcinogenesis observed in individuals with LS. 9 Early screening for CRC in MMR mutation carriers has been proven to reduce morbidity and mortality substantially. 10 Colonoscopic surveillance in individuals with LS at a young age has been reported to decrease the incidence of CRC. 11 Surveillance in CRC MMR mutation carriers saves the health care system a considerable financial cost. 12 Thus, it is imperative to identify patients with LS and their family members who carry a MMR gene mutation in Saudi Arabia. MMR gene mutation carriers may not fulfill the Amsterdam II and revised Bethesda guidelines criteria, which diminishes their usefulness in clinical practice. 13 Version of the National Comprehensive Cancer Network guidelines endorsed by the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) recommends LS screening using either immunohistochemistry (IHC) or the MSI screening approach for all patients with CRC or those patients diagnosed at age <70yearsaswellasthosediagnosedatage70 years who meet the Bethesda guidelines criteria. This universal or reflex testing method helps in identifying patients who should undergo further genetic testing for LS regardless of family history. 14 Modern families tend to be small and the small size of these families could relate to incomplete penetrance of MMR mutations, making it difficult to identify LS. 15 Assuming an average family size of 3 to 4 individuals in the United States, the successful identification of 1 proband case of LS and screening of >5 of his or her at-risk relatives can lead to detection of 3 additional family members who carry the mutation. 4 However, Saudi families tend to differ in size and consist of at least 6 members. 16 Therefore, the screening of relatives of a confirmed LS case will most likely yield a higher frequency of affected family members with LS. The current study focused on establishing the incidence of LS in a cohort of 807 cases of CRC in Saudi Arabia by reflex testing for MSI by polymerase chain reaction (PCR) and IHC in all samples followed by comprehensive MMR mutation testing in all cases with high MSI (MSI- H) and cases with low MSI (MSI-L) along with selected high-risk cases from the microsatellite stable (MSS) group. MATERIALS AND METHODS Patient Selection A random selection of 807 patients with CRC diagnosed between 1990 and 2011 was made from King Faisal Specialist Hospital and Research Centre and Security Forces Hospital in Riyadh, Saudi Arabia. All clinical data were available for each patient. The study was approved by the Institutional Review Board at King Faisal Specialist Hospital and Research Centre and a waiver of consent was obtained for Research Advisory Council (RAC) project RAC# Selection Criteria for HNPCC Screening All MSI-H cases, all MSI-L cases, and 191 high-risk MSS cases were selected to undergo sequencing to estimate MMR gene mutations by PCR. Although family history is central to the Amsterdam II or revised Bethesda guidelines criteria, the social and cultural issues in the region did not permit us to interview all 807 individuals. We followed the revised Bethesda guidelines criteria and narrowed down the family history screening to the 8 cases with HNPCC who had definite gene mutations. Any case who was diagnosed with CRC before age 50 years or who had MSI-H-like morphology before age 60 years was considered to have met the revised Bethesda guidelines criteria. Immunohistochemistry Assessment Tissue microarray (TMA) was constructed and TMA slides from 2 TMA replica blocks were processed and stained manually, as described earlier. 17 The primary antibodies used are shown in Table 1. Protein expression for the MLH1, MSH2, MSH6, and PMS2 antigens were ascertained in 807 CRC tissue samples. Normal expression of the respective antigen was reported when there was nuclear staining in normal and cancerous colon epithelium. Abnormal gene expression resulted in absent nuclear staining in cancerous epithelium. Loss of staining in cancer with concurrent positive staining in the nuclei of normal colon epithelial cells indicated protein inactivation. Adjacent nonneoplastic colon and stromal inflammatory cells served as internal positive controls. DNA isolation DNA was isolated from paraffin-embedded tumor tissues and /or nontumor tissues using the Gentra DNA isolation kit (Qiagen, Germantown, Md) following the manufacturer s recommendations as described previously. 18,19 PCR and DNA Sequencing for 4 MMR Genes and the BRAF Gene To search for germline mutations of the 4 MMR genes (MLH1, MSH2, MSH6, and PMS2), all exons, including Cancer June 1,

3 Original Article TABLE 1. Antibodies Used For the Immunohistochemistry Assay Antibody Clone Source Antigen Retrieval Visualization System Dilution MLH1 G BD Pharmingen Dako retrieval solution (ph 9) Dako EnVision1 1:50 overnight MSH2 FE11 Oncogene/CalBiochem Dako retrieval solution (ph 9) Dako EnVision1 1:100 overnight PMS2 C-20 SCBT Dako retrieval solution (ph 9) Dako EnVision1 1:100 overnight MSH6 44 BD Transduction Laboratories Dako retrieval solution (ph 9) Dako EnVision1 1:100 overnight Abbreviations: MLH1, mutl homolog 1; MSH2, muts protein homolog 2; MSH6, muts protein homolog 6; PMS2, postmeiotic segregation increased 2. TABLE 2. Primer Sequence of the BRAF Gene Exon Forward Primer Reverse Primer BRAF Exon 15 TGCTTGCTCTGATAGGAAAATG AGCATCTCAGGGCCAAAAAT splice junctions, were amplified and sequenced using primers and DNA samples from formalin-fixed paraffin-embedded blocks as described previously. 20 Primer 3 software was used to design the primers for exon 15 of BRAF (Table 2). PCR was performed in a total volume of 25 ml using 20 ng of genomic DNA, 2.5 ml of 10X Taq buffer, 0.8 ml of MgCl2 (25 mm), 0.2 ml of dntp (25 mm), 0.2 ml of Taq polymerase (5U/mL) (all reagents were from Qiagen), 0.5 ml of each primer (5 mm), and water. The efficiency and quality of the amplified PCR products were confirmed by running the PCR products on a 2% agarose gel. The PCR products were subsequently subjected to direct sequencing with BigDye terminator V 3.1 cycle sequencing reagents and analyzed on an ABI 3130XL genetic analyzer (Applied Biosystems, Foster City, Calif). Sequencing results were compared with the reference DNA sequence by Mutation Surveyor V4.04 (SoftGenetics LLC, State College, PA). Bisulfite Modification of DNA Bisulfite modification was performed as described previously. 21 Real-Time PCR (MethyLight) for Quantitative DNA Methylation Analysis Real-time PCR (MethyLight) for the determination of CIMP was performed as described previously. 22 We used an ABI 7300 analyzer (Applied Biosystems) for quantitative real-time PCR. Nine sets of primers and probes were used to amplified promoters of 8 genes of interest (including CDKN2A, CRABP1, NEUROG1, CACNA1G, IGF2, RUNX3, MLH-1, andsocs1) andcol2a1 (the collagen 2A1 gene) to normalize for the amount of input bisulfiteconverted DNA. Primers and probes for all the genes were previously validated and published. 23,24 The percentage of TABLE 3. Clinicopathological Variables for the Patient Cohort Patients Characteristics N (%) Age Median 57.0 Range (IQR) Sex Male 430 (53.2) Female 377 (46.8) Status at last follow-up Alive 605 (75.0) Dead 182 (22.5) Unknown 20 (2.5) Histological type Adenocarcinoma 716 (88.7) Mucinous carcinoma 91 (11.3) Histological grade Well differentiated 77 (9.6) Moderately differentiated 630 (78.2) Poorly differentiated 98 (12.2) Tumor site Left 649 (80.4) Right 125 (15.5) Unknown 33 (4.1) TNM stage of disease I 94 (11.6) II 258 (32.0) III 321 (39.8) IV 100 (12.4) Unknown 34 (4.2) Abbreviation: IQR, interquartile range. methylated reference (PMR) as described 25 for each gene was calculated by dividing the GENE:COL2A1 ratio of a sample by the GENE:COL2A1 ratio of CpG methyltransferase-treated human genomic DNA (assuming it was fully methylated) and multiplying it by 100. A PMR cutoff of 4 was established to distinguish methylation positivity (PMR >4) from methylation negativity (PMR 4). The PMR cutoff of 4 as previously described was used for CDKN2A, NEUROG1, CACNA1G, RUNX3, MLH-1, and SOCS1 and a PMR cutoff of 6 was used for IGF2 and CRABP1. 29 Cases were considered to be high for CIMP if 5 of 8 CIMP-specific genes were methylated. 23,30 Microsatellite Markers and Analyses Allelic imbalances were measured by PCR as described. 31 A reference panel of 5 pairs of microsatellite primers, 1764 Cancer June 1, 2015

4 Lynch Syndrome in CRC in Saudi Arabia/Siraj et al Figure 1. Flowchart of 807 cases of colorectal cancer (CRC) studied to estimate the prevalence of Lynch syndrome. IHC indicates immunohistochemistry; MLH-1, mutl homolog 1; MSH-2, muts protein homolog 2; MSH-6, muts protein homolog 6; MSI-H, high microsatellite instability; MSI-L, low microsatellite instability; MSI-S, stable microsatellite instability; NA, not available; PMS-2, postmeiotic segregation increased 2. comprising 2 mononucleotide microsatellites (BAT24 and BAT26) and 3 dinucleotide microsatellites (DS123, D5S346, and D17S240), was used to determine tumor MSI status. 32 Multiplex PCR was performed in a total volume of 24 ml using 50 ng of genomic DNA, 2.5 ml of10xtaq buffer, 1.5 mlofmgcl 2 (24mM),10pmoloffluorescentlabeled primers, 0.05 ml of dntp(10mm), and 0.2mLof Taq polymerase (1UmL-1; all reagents were from Qiagen Inc). PCR was performed using an MJ Research PTC-200 thermocycler (Bio-Rad Laboratories, Hercules, Calif). The samples in which the novel alleles were found at one of these loci and 2 of those 5 loci were classified as MSI-L and MSI-H, respectively, whereas samples without novel alleles at any of those loci were classified as MSS. Statistical Analysis The JMP 10.0 statistical software package (SAS Institute Inc, Cary, NC) was used for data analyses. Comparisons between groups were made with the Student t test. Chisquare tests were used to examine the relation between nominal variables. Values of P<.05 were considered to be statistically significant. RESULTS Clinicopathologic Data Themedianageofthepatientsatthetimeofsurgery was 57 years (interquartile range, ( years). Detailed clinicopathological information is shown in Table 3. Microsatellite Analysis Among the cohort of 807 cases, there were 91 MSI-H cases (11.3%), 143 MSI-L cases (17.7%), and 573 MSS cases (71%). Of the 91 CRC cases with MSI-H, 10 (11.0%) were determined to be noninterpretable by IHC due to a lack of representative cancer tissue or for technical reasons (Fig. 1). Of the 81 MSI-H cases that had analyzable IHC expression data, 10 cases (12.3%) expressed MMR staining by IHC and were discordant with MSI-PCR results. The remaining 71 MSI-H cases Cancer June 1,

5 Original Article Figure 2. Tissue microarray (TMA) spots of colorectal cancer (CRC) showing loss of staining in mutl homolog 1 (MHL1), muts protein homolog 2 (MSH2), and muts protein homolog 6 (MSH6). The neoplastic cells were negative and the stromal cells demonstrated a positive nuclear reaction. (A, C, and E) TMA spot of CRC demonstrating staining for MLH1, MSH2, and MSH6, respectively. (B, D, and F) View using a 20 3 /0.70 objective lens. Inset: 403 /0.85 aperture magnified view of the same. (87.7%) were concordant with IHC because they displayed loss of protein expression for 1 of the 4 MMR genes. Of the 143 MSI-L cases, 5 (3.5%) were not interpretable by IHC due to a lack of representative cancer tissue or for technical reasons, and the remaining 138 cases 1766 Cancer June 1, 2015

6 Lynch Syndrome in CRC in Saudi Arabia/Siraj et al TABLE 4. Correlation Between MSI-PCR and Clinicopathological Parameters in CRC (N 5 807) a Total MSI-H MSI-L/S No. % No. % No. % Total no. of cases Age, y < > Sex Male Female Tumor site Left colon <.0001 Right colon Histological type Adenocarcinoma <.0001 Mucinous carcinoma TNM stage of disease I II III IV Differentiation Well Moderate Poor BRAF mutation Positive CIMP High Low and middle Abbreviations: CIMP, CpG island methylator phenotype; CRC, colorectal carcinoma; MSI-H, high microsatellite instability; MSI-L/S, low microsatellite instability or microsatellite stable; MSI-PCR, microsatellite instability polymerase chain reaction. a Data were not available for some cases for tumor site (33 cases), TNM stage of disease (34 cases), grade (2 cases), BRAF mutation (18 cases), and CIMP (315 cases). (96.5%) were determined to be microsatellite proficient and thus were concordant with IHC. Of the remaining 573 MSS cases, 66 cases (11.5%) were excluded from analysis because they were nonrepresentative TMA spots. Of the 507 MSS-PCR cases that had analyzable IHC expression data, discordance with IHC was observed in 2 cases (0.4%) that demonstrated loss of MLH1 expression and the remaining 505 cases (99.6 %) were determined to be MMR proficient by IHC. P Immunohistochemistry Complete loss of nuclear protein expression in the tumor cells for a minimum of 1 of the 4 MMR genes by IHC was noted among 73 of the 807 cases. Of the 73 MSI cases, 71 (97.2%) were concordant with MSI-PCR and the remaining 2 cases (0.3%) were discordant with MSI-PCR and observed in the subgroup of 573 MSS cases. The distribution of loss of expression of MLH1, MSH2, MSH6, and PMS2 in the 71 MSI-H CRC cases was 38 cases, 14 cases, 9 cases, and 10 cases, respectively. Of the 38 cases with loss of MLH1 expression, 9 cases had loss of MLH1 expression alone, and 28 cases had loss of both MLH1 and PMS2 expression. It is interesting to note that 1 case demonstrated loss of expression for MLH1, MSH2, and MSH6. Of the 14 CRC cases with loss of MSH2 expression, 7 demonstrated loss of MSH2 alone whereas the remaining cases demonstrated loss of both MSH2 and MSH6. Similarly, 7 cases demonstrated loss of only MSH6 expression. It is interesting to note that 2 cases were observed to have loss of both MSH6 and PMS2 expression. Loss of PMS2 expression only was observed in 10 cases (Fig. 2). Correlation of MSI-PCR and MMR Protein-IHC With Clinicopathological Parameters Tumors found to be MSI-H by PCR were found to be significantly associated with right-sided tumors (P<.0001) and mucinous histology (P<.0001) (Table 4). Loss of staining of any 1 of the 4 MMR genes was found to be significantly associated with right-sided tumors (P<.0001), mucinous histology (P ), and MSI-H tumors (P<.0001) (Table 5). Correlation of MSI-PCR With BRAF Mutations and CIMP Phenotype BRAF mutations were observed in 2.4% of cases (19 of 789 cases); the remaining 18 cases were not interpretable due to a lack of sufficient DNA or for technical reasons. Due to limited DNA availability, CIMP data were analyzed only in 492 cases and the incidence of tumors with high CIMP was 4.7% (23 of 492 cases). Four of the 19 BRAF-mutated cases also had the high CIMP phenotype. MSI-H tumors were found to be significantly associated with BRAF mutations (P ) and high CIMP phenotype (P ). Cases With LS Of the 425 samples, we identified MMR germline mutations in 8 cases and the frequency of LS in the current study cohort was 0.99% (8 of 807 cases). All 8 cases fulfilled the revised Bethesda guidelines criteria and were determined to be MSI-H by molecular testing; all of them had a BRAF wild-type, 6 were right sided, 3 were of poor grade, and 4 were diagnosed among patients aged <50 years. MMR mutations in MLH1, MSH2,andMSH6 were observed in 2 (25%), 5 (62.5%), and 1 (12.5%) cases, respectively. One MLH1 germline mutation that was observed in 2 cases was missense mutations c.1652a>c, p.asn551thr, which was observed in exon 14 and Cancer June 1,

7 Original Article TABLE 5. Correlation Between MSI-IHC and Clinicopathological Parameters in CRC (N 5 807) a Total Loss of Expression b Normal Expression c No. % No. % No. % P Total no. of cases Age, y < > Sex Male Female Tumor site Left colon <.0001 Right colon Histological type Adenocarcinoma Mucinous carcinoma TNM stage of disease I II III IV Differentiation Well <.0001 Moderate Poor MSI molecular MSI-H <.0001 MSI-S/L BRAF mutation Positive CIMP High Low and middle Abbreviations: CIMP, CpG island methylator phenotype; CRC, colorectal carcinoma; MSI-H, high microsatellite instability; MSI-IHC, microsatellite instability immunohistochemistry; MSI-L/S, low microsatellite instability or microsatellite stable. a Data were not available for some cases for tumor site (30 cases), TNM stage of disease (31 cases), BRAF mutation (5 cases), and CIMP (272 cases). b Loss of expression was observed in tumor glands accompanied by persistent staining in stromal cells and normal glands. These tumors require further testing. c Normal expression was observed in tumor glands, adjacent normal tissue, and stromal cells. demonstrated loss of expression for both MLH1 and PMS2. To our knowledge, there were 2 novel MSH2 germline mutations that were noted in 3 cases. The first was an in-frame deletion in exon 14 (c delttctac, p.thr756-tyr757 del) and the second MSH2 mutation was a frameshift mutation (c.1964.delt, p.val655fs) in exon 12 and truncated at codon 684. Of these 5 MSH2 germline mutations, 2 cases demonstrated loss of MSH2 expression alone, 2 cases had loss of both MSH2 and MSH6 expression, and the fifth case had an isolated loss of MSH6 expression and had a nonsense mutation at c 289C>T 97Q>/X in exon 2. Only 1 case demonstrated a germline mutation in the MSH6 gene (insertion in exon 6 at c.3475inst, p.tyr1159fs) and demonstrated loss of MSH6 expression (Table 6) (Fig. 3). DISCUSSION To the best of our knowledge, the current study is one of the most comprehensive studies performed to date, and the first of its kind to examine the incidence of LS in the Arab population from the Middle East. The gold standard for the diagnosis of LS is demonstration of a germline mutation in 1 of the 4 MMR genes. However, adoption of an effective screening program that can accurately identify the majority of LS cases is the subject of continuous debate in the literature, with some groups advocating selective screening whereas recent studies strongly emphasize the role of a universal screening program for all patients with CRC regardless of clinicopathological features or family history. 33 With the use of both PCR and IHC as universal screening tools followed by comprehensive sequencing of 1768 Cancer June 1, 2015

8 Lynch Syndrome in CRC in Saudi Arabia/Siraj et al TABLE 6. Details of 8 Cases of HNPCC HNPCC Gene Mutation MMR- IHC Proteins MSI M olecular Family History Age, Years Sex Exon Mutation Mutation HGMD BRAF Mutation Case 1 MLH1 MLH1 and PMS2 MSI-H Positive (patient had CRC at age 50 y; 2 firstdegree relatives had CRC at age 45 y) Case 2 MSI-H Positive (patient had CRC at age 40 y; father and brother had CRC age 45 y) Case 3 MSH2 MSH2 and MSH6 MSI-H Positive (patient had CRC at age 50 y and endometrial cancer; son had CRC at age 36 y); patient diagnosed with endometrial cancer in 2001 and CRC in 2003 Case 4 MSI-H Positive (patient s mother died of CRC at age 50 y) Case 5 MSH2 MSI-H Positive (patient also had breast cancer; son had CRC at age 42 y) Case 6 MSI-H Positive (patient had CRC at age 36 y; mother had CRC at age 50 y) Case 7 MSH6 MSI-H Positive (family history: his father, son, and sister died of colon cancer) Case 8 MSH6 MSH6 MSI-H Positive (patient had CRC at age 44 y and histologic subtype was mucinous cancer) 46 Male 14 c.1652a>c; p.n551t 39 Male 14 c.1652a>c; p.n551t Deleterious Deleterious Reported Reported 61 Female 12 c. 1964delT Deleterious Not reported 54 Male 7 c.1226_1227delag Deleterious Reported 60 Female 14 c.2262_2267 delttctac Deleterious Not reported 36 Male 12 c. 1964delT Deleterious Not reported 65 Male 2 c.289c>t; p.q97* Deleterious Reported 44 Male 6 c.3475inst Deleterious Not reported Abbreviations: CRC, colorectal carcinoma; HGMD, Human Gene Mutation Database; HNPCC, hereditary nonpolyposis colorectal cancer; MLH1, mutl homolog 1; MMR-IHC, mismatch repair immunohistochemistry; MSH2, muts protein homolog 2; MSH6, muts protein homolog 6; MSI, microsatellite instability; MSI-H, high microsatellite instability; PMS2, postmeiotic segregation increased 2. all 4 MMR genes, the incidence of LS among the patients with CRC in the current study was 0.9%. This prevalence translates to nearly 1 LS case per every 100 cases of CRC newly diagnosed in Saudi Arabia. This accounted for 9.9% of all MSI-H cases, with the most common germline MMR mutation noted in MSH2 (5 cases; 62.5%), followed by MLH1 mutations (2 cases; 25%); MSH6 germline mutations were found to be the least common (1 case; 12.5%). These results are in agreement with earlier studies from the United States, Finland, and Spain, with the most common MMR mutation observed in MSH2. Considering that the size of the average Saudi family tends to be approximately 6 individuals compared with a family size of 3 in the West, the detection rate among relatives of a Saudi LS proband would also be higher compared with the West. 16 Once the LS proband is identified and the type of MMR mutation characterized, it would also be significantly cost-effective to screen all relatives who are at risk only for that particular mutation. In addition to cancer surveillance, carriers of MMR germline mutations can benefit from cancer preventive measures including prophylactic surgeries and preimplantation genetic diagnosis, which will prevent further transmission of the germline mutations to future generations. 13,37,38 Although LS has been extensively characterized in a Western population and the current study is not novel, to the best of our knowledge, the true incidence of LS in Saudi Arabia has not been reported to date. The incidence of LS in all CRC cases in the current study was 0.9%, which is in agreement with incidences reported in earlier studies One of the limiting factors of the current study was the lack of a detailed family history for LSassociated cancers. Evaluating cases for Amsterdam II criteria posed unique difficulties. The data collected regarding CRC address one important issue: because of the lack of a modern health care system in Saudi Arabia until the late 1970s, it is impossible to depend on the information Cancer June 1,

9 Original Article Figure 3. Sequencing of traces of different mismatch repair gene mutations: (A) mutl homolog 1 (MLH1): c.1652a>c, p.asn551thr mutation; (B) muts protein homolog 2 (MSH2): c.2262_2267delttctac, p.thr756_tyr757del mutation; (C) muts protein homolog 6 (MSH2): c.289c>t, p.gln97stop mutation; (D) MSH2: c.1226_1227delag,p.gln409fs mutation; (E) MSH2: c.1964delt, p.val655fs mutation; and (F) MSH6: c.3475inst, p.thr1159fs mutation. The upper panel of each section denotes the wild-type sequence and the lower panel denotes the mutation. provided by the patients, especially if they did not know the disease from which their family member suffered or died. 39 Although family history is central to the Amsterdam II or the revised Bethesda guidelines criteria, the social and cultural issues in the region did not permit us to interview all 807 patients. This was partially overcome by performing MSI analysis by PCR followed by gene mutation testing despite the high costs involved. Furthermore, previous studies have shown discrepancies in MSI and IHC testing, resulting in discordant cases that may be classified as MSI-H by PCR but as MSS by IHC and vice versa. Future studies exploring familial CRC are being undertaken in the study laboratory to identify relevant molecular impairments using next-generation sequencing technology such as whole exome sequencing. In addition, we are also investigating the prevalence of MLH1 and MSH2 epigenetic mutations in CRC. Conclusions The current study highlights the incidence of LS among patients with CRC in Saudi Arabia. Reflex testing of all newly diagnosed CRC cases for MSI by PCR and IHC followed by MMR sequencing in only those cases demonstrating MSI-H that are BRAF wild-type will enable clinicians to initiate timely and appropriate screening, thereby facilitating prevention in patients and families at risk. This will pave the way for a national program to screen all patients with CRC for LS in Saudi Arabia. FUNDING SUPPORT No specific funding was disclosed. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. REFERENCES 1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, CA Cancer J Clin. 2009;59: Aljebreen AM. Clinico-pathological patterns of colorectal cancer in Saudi Arabia: younger with an advanced stage presentation. Saudi J Gastroenterol. 2007;13: Isbister WH. Colorectal cancer below age 40 in the Kingdom of Saudi Arabia. Aust N Z J Surg. 1992;62: Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for Lynch syndrome among patients with colorectal cancer. J Clin Oncol. 2008;26: Lynch HT, de la Chapelle A. Hereditary colorectal cancer. N Engl J Med. 2003;348: Cancer June 1, 2015

10 Lynch Syndrome in CRC in Saudi Arabia/Siraj et al 6. Toyota M, Ahuja N, Ohe-Toyota M, Herman JG, Baylin SB, Issa JP. CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U S A. 1999;96: Tol J, Nagtegaal ID, Punt CJ. BRAF mutation in metastatic colorectal cancer. N Engl J Med. 2009;361: Stoffel E, Mukherjee B, Raymond VM, et al. Calculation of risk of colorectal and endometrial cancer among patients with Lynch syndrome. Gastroenterology. 2009;137: Lynch HT, Lynch J. Lynch syndrome: genetics, natural history, genetic counseling, and prevention. J Clin Oncol. 2000;18(suppl 21): 19S-31S. 10. de Jong AE, Hendriks YM, Kleibeuker JH, et al. Decrease in mortality in Lynch syndrome families because of surveillance. Gastroenterology. 2006;130: Stupart DA, Goldberg PA, Algar U, Ramesar R. Surveillance colonoscopy improves survival in a cohort of subjects with a single mismatch repair gene mutation. Colorectal Dis. 2009;11: 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, Blanco I, Aktan-Collan K, et al; Mallorca group. Revised guidelines for the clinical management of Lynch syndrome (HNPCC): recommendations by a group of European experts. Gut. 2013;62: Benson AB 3rd, Bekaii-Saab T, Chan E, et al; National Comprehensive Cancer Network. Localized colon cancer, version : featured updates to the NCCN Guidelines. J Natl Compr Canc Netw. 2013;11: Steinke V, Engel C, Buttner R, Schackert HK, Schmiegel WH, Propping P. Hereditary nonpolyposis colorectal cancer (HNPCC)/ Lynch syndrome. Dtsch Arztebl Int. 2013;110: Abou-Auda HS. An economic assessment of the extent of medication use and wastage among families in Saudi Arabia and Arabian Gulf countries. Clin Ther. 2003;25: Bavi P, Uddin S, Bu R, et al. The biological and clinical impact of inhibition of NF-jB-initiated apoptosis in diffuse large B cell lymphoma (DLBCL). J Pathol. 2011;224: Abubaker J, Jehan Z, Bavi P, et al. Clinicopathological analysis of papillary thyroid cancer with PIK3CA alterations in a Middle Eastern population. J Clin Endocrinol Metab. 2008;93: Khoo US, Ozcelik H, Cheung AN, et al. Somatic mutations in the BRCA1 gene in Chinese sporadic breast and ovarian cancer. Oncogene. 1999;182: Bu R, Siraj AK, Bavi P, et al. Constitutional mismatch repairdeficiency syndrome is a rare cause of cancer even in a highly consanguineous population. J Cancer Ther. 2013;4: Siraj AK, Hussain AR, Al-Rasheed M, et al. Demethylation of TMS1 gene sensitizes thyroid cancer cells to TRAIL-induced apoptosis. J Clin Endocrinol Metab. 2011;96:E215-E Ogino S, Kawasaki T, Brahmandam M, et al. Precision and performance characteristics of bisulfite conversion and real-time PCR (MethyLight) for quantitative DNA methylation analysis. J Mol Diagn. 2006;8: Berg M, Hagland HR, Soreide K. Comparison of CpG island methylator phenotype (CIMP) frequency in colon cancer using different probe- and gene-specific scoring alternatives on recommended multigene panels. PLoS One. 2014;9:e Ogino S, Nosho K, Kirkner GJ, et al. CpG island methylator phenotype, microsatellite instability, BRAF mutation and clinical outcome in colon cancer. Gut. 2009;58: Eads CA, Lord RV, Wickramasinghe K, et al. Epigenetic patterns in the progression of esophageal adenocarcinoma. Cancer Res. 2001;61: Widschwendter M, Siegmund KD, Muller HM, et al. Association of breast cancer DNA methylation profiles with hormone receptor status and response to tamoxifen. Cancer Res. 2004;64: Eads CA, Danenberg KD, Kawakami K, et al. MethyLight: a highthroughput assay to measure DNA methylation. Nucleic Acids Res. 2000;28:E Eads CA, Lord RV, Kurumboor SK, et al. Fields of aberrant CpG island hypermethylation in Barrett s esophagus and associated adenocarcinoma. Cancer Res. 2000;60: Samowitz WS. The CpG island methylator phenotype in colorectal cancer. J Mol Diagn. 2007;9: Kim JH, Shin SH, Kwon HJ, Cho NY, Kang GH. Prognostic implications of CpG island hypermethylator phenotype in colorectal cancers. Virchows Arch. 2009;455: Bavi P, Jehan Z, Bu R, et al. ALK gene amplification is associated with poor prognosis in colorectal carcinoma. Br J Cancer. 2013;109: 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: Moreira L, Balaguer F, Lindor N, et al; EPICOLON Consortium. Identification of Lynch syndrome among patients with colorectal cancer. JAMA. 2012;308: Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med. 2005;352: Mecklin JP, Jarvinen HJ, Hakkiluoto A, et al. Frequency of hereditary nonpolyposis colorectal cancer. A prospective multicenter study in Finland. Dis Colon Rectum. 1995;38: Perez-Carbonell L, Ruiz-Ponte C, Guarinos C, et al. Comparison between universal molecular screening for Lynch syndrome and revised Bethesda guidelines in a large population-based cohort of patients with colorectal cancer. Gut. 2012;61: Rich TA, Liu M, Etzel CJ, et al. Comparison of attitudes regarding preimplantation genetic diagnosis among patients with hereditary cancer syndromes. Fam Cancer. 2014;13: Daina G, Ramos L, Obradors A, et al. First successful double-factor PGD for Lynch syndrome: monogenic analysis and comprehensive aneuploidy screening. Clin Genet. 2013;84: Khaliq AA. The Saudi health care system: a view from the minaret. World Health Popul. 2012;13: Cancer June 1,