Somatic Mutations In Mismatch Repair Genes Research Articles

A tumor focused approach to resolving the etiology of DNA mismatch repair deficient tumors classified as suspected Lynch syndrome

Routine screening of tumors for DNA mismatch repair (MMR) deficiency (dMMR) in colorectal (CRC), endometrial (EC) and sebaceous skin (SST) tumors leads to a significant proportion of unresolved cases classified as suspected Lynch syndrome (SLS). SLS cases (n = 135) were recruited from Family Cancer Clinics across Australia and New Zealand. Targeted panel sequencing was performed on tumor (n = 137; 80×CRCs, 33×ECs and 24xSSTs) and matched blood-derived DNA to assess for microsatellite instability status, tumor mutation burden, COSMIC tumor mutational signatures and to identify germline and somatic MMR gene variants. MMR immunohistochemistry (IHC) and MLH1 promoter methylation were repeated. In total, 86.9% of the 137 SLS tumors could be resolved into established subtypes. For 22.6% of these resolved SLS cases, primary MLH1 epimutations (2.2%) as well as previously undetected germline MMR pathogenic variants (1.5%), tumor MLH1 methylation (13.1%) or false positive dMMR IHC (5.8%) results were identified. Double somatic MMR gene mutations were the major cause of dMMR identified across each tumor type (73.9% of resolved cases, 64.2% overall, 70% of CRC, 45.5% of ECs and 70.8% of SSTs). The unresolved SLS tumors (13.1%) comprised tumors with only a single somatic (7.3%) or no somatic (5.8%) MMR gene mutations. A tumor-focused testing approach reclassified 86.9% of SLS into Lynch syndrome, sporadic dMMR or MMR-proficient cases. These findings support the incorporation of tumor sequencing and alternate MLH1 methylation assays into clinical diagnostics to reduce the number of SLS patients and provide more appropriate surveillance and screening recommendations.

Molecular characterization of gastric adenocarcinoma diagnosed in patients previously treated for Hodgkin lymphoma or testicular cancer.

IntroductionThe risk of developing gastric cancer is increased in patients treated with radiotherapy for Hodgkin lymphoma (HL) or testicular cancer (TC). This study aims to assess if gastric adenocarcinoma after treatment for HL/TC (t-GC) is molecularly different from gastric adenocarcinoma in the general population.MethodsPatients were diagnosed with t-GC ≥5 years after treatment for HL/TC. Four molecular subtypes were identified using immunohistochemical and molecular analyses: Epstein-Barr virus (EBV), mismatch repair (MMR) deficiency or microsatellite instability (MSI), aberrant p53 staining as surrogate for chromosomal instability (sCIN), and a surrogate for genomic stability (sGS) without these aberrations. Results were compared with gastric cancer in the general population (p-GC) described in literature.ResultsMolecular subtyping of 90 t-GCs resulted in 3% EBV, 8% MSI, 36% sCIN and 53% sGS. 3/6 of MSI t-GCs had MLH1 promoter methylation and 2/6 were explained by double somatic mutations in MMR genes. T-GCs were more frequently sGS than p-GCs (53% vs. 38%, p = 0.04). T-GC was more frequently sGS in HL/TC patients diagnosed before 1990, than after 1990 (63% vs. 38%, p = 0.03). T-GCs located in the antrum, an area that receives high irradiation doses, were more frequently sGS (61% vs. 28% in p-GCs, p = 0.02).ConclusionOur results demonstrate that t-GCs are more frequently of the sGS subtype than p-GCs. An association of t-GC of the sGS subtype with prior anticancer treatment is suggested by the high frequency in HL/TC patients who were treated before 1990, a time period in which HL/TC treatments were more extensive.

Germline and Tumor Sequencing as a Diagnostic Tool To Resolve Suspected Lynch Syndrome

Patients in whom mismatch repair (MMR)-deficient cancer develops in the absence of pathogenic variants of germline MMR genes or somatic hypermethylation of the MLH1 gene promoter are classified as having suspected Lynch syndrome (SLS). Germline whole-genome sequencing (WGS) and targeted and genome-wide tumor sequencing were applied to identify the underlying cause of tumor MMR deficiency in SLS. Germline WGS was performed on samples from 14 cancer-affected patients with SLS, including two sets of first-degree relatives. MMR genes were assessed for germline pathogenic variants, including complex structural rearrangements and noncoding variants. Tumor tissue was assessed for somatic MMR gene mutations using targeted, whole-exome sequencing or WGS. Germline WGS identified pathogenic MMR variants in 3 of the 14 cases (21.4%), including a 9.5-megabase inversion disrupting MSH2 in a mother and daughter. Excluding these 3 MMR carriers, tumor sequencing identified at least two somatic MMR gene mutations in 8 of 11 tumors tested (72.7%). In a second mother-daughter pair, a somatic cause of tumor MMR deficiency was supported by the presence of double somatic MSH2 mutations in their respective tumors. More than 70% of SLS cases had double somatic MMR mutations in the absence of germline pathogenic variants in the MMR or other DNA repair-related genes on WGS, and, therefore, were confidently assigned a noninherited cause of tumor MMR deficiency.

Molecular Basis of Mismatch Repair Protein Deficiency in Tumors from Lynch Suspected Cases with Negative Germline Test Results.

Some 10–50% of Lynch-suspected cases with abnormal immunohistochemical (IHC) staining remain without any identifiable germline mutation of DNA mismatch repair (MMR) genes. MMR proteins form heterodimeric complexes, giving rise to distinct IHC patterns when mutant. Potential reasons for not finding a germline mutation include involvement of an MMR gene not predicted by the IHC pattern, epigenetic mechanism of predisposition, primary mutation in another DNA repair or replication-associated gene, and double somatic MMR gene mutations. We addressed these possibilities by germline and tumor studies in 60 Lynch-suspected cases ascertained through diagnostics (n = 55) or research (n = 5). All cases had abnormal MMR protein staining in tumors but no point mutation or large rearrangement of the suspected MMR genes in the germline. In diagnostic practice, MSH2/MSH6 (MutS Homolog 2/MutS Homolog 6) deficiency prompts MSH2 mutation screening; in our study, 3/11 index individuals (27%) with this IHC pattern revealed pathogenic germline mutations in MSH6. Individuals with isolated absence of MSH6 are routinely screened for MSH6 mutations alone; we found a predisposing mutation in MSH2 in 1/7 such cases (14%). Somatic deletion of the MSH2-MSH6 region, joint loss of MSH6 and MSH3 (MutS Homolog 3) proteins, and hindered MSH2/MSH6 dimerization offered explanations to misleading IHC patterns. Constitutional epimutation hypothesis was pursued in the MSH2 and/or MSH6-deficient cases plus 38 cases with MLH1 (MutL Homolog 1)-deficient tumors; a primary MLH1 epimutation was identified in one case with an MLH1-deficient tumor. We conclude that both MSH2 and MSH6 should be screened in MSH2/6- and MSH6-deficient cases. In MLH1-deficient cases, constitutional epimutations of MLH1 warrant consideration.

Identification of Germline Mismatch Repair Gene Mutations in Lung Cancer Patients With Paired Tumor-Normal Next Generation Sequencing: A Retrospective Study.

Background: Paired tumor-normal targeted next-generation sequencing (NGS) is primarily used to identify actionable somatic mutations, but can also detect germline variants including pathogenic germline mutations in DNA mismatch repair (MMR) genes that underlie Lynch syndrome. In the present study we examined paired NGS data from lung cancer patients to identify germline mutations in MMR genes. As lung cancer is not one of the recognized Lynch syndrome-associated neoplasms, we also investigated whether these lung cancer cases are due to Lynch syndrome or are instead sporadic cancers occurring in Lynch syndrome patients.Methods: A retrospective study of 1,179 lung cancer patients with available paired NGS data was performed to identify germline mutations in the MMR genes MLH1, MSH2, MSH6, and PMS2, and evaluate tumor mutation burden (TMB). Microsatellite instability (MSI) testing was done on select cases with MMR gene mutations by either NGS or PCR/capillary electrophoresis approach. Immunohistochemistry (IHC) for MMR proteins was performed in select patients.Results: Pathogenic or likely-pathogenic germline mutations in PMS2, MSH2, or MSH6 were detected in 0.5% (6/1,179) of lung cancer patients; three of the patients had a family history of colon or gastric cancer. The median age at diagnosis of these cases was 68.5 years old. None of these six patients exhibited MSI or loss of MMR protein expression. Among them, no second hit somatic mutations in MMR genes (including single-nucleotide variants, small insertions or deletions and copy number alterations) were detected, and the median TMB was 4.5 muts/MB. Subsequent genetic testing of family members identified new Lynch syndrome cases in two first-degree relatives.Conclusion: These data imply that lung cancers in Lynch syndrome patients are unrelated to the underlying Lynch syndrome diagnosis and occur spontaneously. Nonetheless, paired tumor-normal NGS can identify germline mutations to help reveal Lynch syndrome in cancer patients. This has important implications for cancer screening and risk reduction in these patients and their families.

Abstract 3394: Actionable fusion kinases in microsatellite instability-high colorectal cancers

Abstract Colorectal cancer (CRC) patients who are microsatellite instability-high (MSI-H) due to mismatch repair (MMR) deficiency have poor prognosis after relapse by consensus molecular subtyping. In response to reports of some MSI-H CRC resistance to immune checkpoint blockade and immune-related adverse events, we sought to identify a new therapeutic target. MSI testing was done on ~2,800 resected colorectal tumors; next-generation sequencing was done on 149 MSI-H tumors showing MSI in both BAT25 and BAT26 loci. MSI-H CRCs are conventionally divided into hereditary and sporadic categories; in this study whole-exome sequencing, transcriptome sequencing, and methylation array revealed that sporadic MSI-H CRCs are further classifiable into somatic MMR gene mutation (Lynch-like) or MLH1 promoter methylated (MM) types. Methylation of other MMR genes (MSH2/MSH6/PMS2) was not detected. The mutational properties of sporadic MSI-H CRCs harboring somatic MMR gene mutations more closely resembled those of Lynch syndrome-associated MSI-H CRCs than those of MM CRCs, which were more likely to develop in the right-sided colon and harbor more insertions/deletions, more recurrent mutations of BRAF/RNF43, and fewer recurrent mutations of KRAS/APC than Lynch syndrome-associated/Lynch-like CRCs (P < 0.001, Fisher's exact test). Loss of heterozygosity of the locus encompassing TGFBR2, MLH1, and CTNNB1 was specific to Lynch syndrome-associated/Lynch-like MSI-H CRCs; this loss appears to be instrumental in the inactivation of TGFBR2/MLH1. Pathway-level analyses of MSI-H CRCs identified DNA damage-sensing and histone H3 methylation-associated pathways as well as Wnt signaling and RAS/RAF/mitogen-activated protein kinase pathways. One mutational signature, Mutational Signature A, accounted for a median percentage 2.3 times larger of mutations in tumors with mutated PMS2 than in tumors with intact PMS2 (P = 1.5 × 10-5, Mann-Whitney U test). Although it is believed that fusion kinases are found in less than 1% of all CRCs, clinically actionable fusion kinases were detected in 11/19 (58%) MM MSI-H CRCs lacking KRAS/BRAF oncogenes. A 3T3 transformation assay confirmed the tumorigenicity of identified fusion kinases; small-molecule inhibitors that suppress the activity of the kinases identified in fusion products significantly attenuated malignant transformation of 3T3 cells in a concentration-dependent manner. Patients with MSI-H CRCs harboring fusion kinases had worse post-relapse prognoses than did patients with MSI-H CRCs harboring KRAS oncogenes. These findings promise to advance MSI-H CRC precision medicine. Citation Format: Kazuhito Sato, Masahito Kawazu, Yoko Yamamoto, Toshihide Ueno, Shinya Kojima, Genta Nagae, Manabu Soda, Takafumi Oga, Shinji Kohsaka, Eirin Sai, Yoshihiro Yamashita, Shoichi Hazama, Hisae Iinuma, Hiroyuki Aburatani, Hiroyuki Mano. Actionable fusion kinases in microsatellite instability-high colorectal cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3394.

PO-467 Unclarified cases of microsatellite instability analyses for lynch syndrome diagnostics

IntroductionLynch syndrome (LS) is an autosomal dominantly inherited form of colorectal cancer (CRC) and is implicated in 2%–4% of CRC cases. LS develops from a mutation in one allele of one of the DNA mismatch repair (MMR) genes, most commonly MLH1 and MSH2, or less frequently MSH6 and PMS2. Loss of functional MMR proteins leads to defects in DNA repair and, subsequently, high DNA microsatellite instability (MSI-High). LS diagnostics currently consists of an analysis of MMR protein expression by means of Immunohistochemistry (IHC) and a molecular analysis to detect MSI. During our 10 year experience in performing LS diagnostics we have encountered few cases where IHC and MSI analyses do not match. This study is to clarify these inconsistent molecular alterations which will contribute to our understanding of LS diagnostics.Material and methodsOf 2335 LS cases 27 (1.1%) showed discrepant IHC and MSI results. To clarify this, IHC and MSI analyses were repeated if possible with different antibodies and different MSI markers (mononucleotide instead of dinucleotide markers). Using MS-MLPA MLH1 hypermethylation, BRAF V600E mutation status and LOH of the MMR genes was analysed. Finally, germline and somatic mutations in MMR genes were analysed by using NGS. Protein modelling was also performed to visualise structural protein changes.Results and discussionsAmong these 27 cases, due to the decline of patients for further testing, 4 cases were excluded from further analyses. Of the remaining cases in 2 cases using a different antibody explained the unexpected results and in 2 cases switching from dinucleotide to mononucleotide MSI markers. In respectively 8 and 4 cases somatic and germline mutations explained the IHC and MSI results. Finally, in 2 cases protein modelling of identified mutations explained the presence of protein staining in MSI-high cases. The rest cases cannot be further studied due to the lack of tissues.ConclusionBased on these results, we have concluded: 1. Mononucleotide markers are more sensitive to detect microsatellite instability. 2. Protein modelling can explain presence of protein staining for mutations that do not cause a major conformational change. 3. Somatic sequencing is a valuable addition to Lynch syndrome diagnostics.

Abstract 4266: Double somatic mutations as a cause of tumor mismatch repair-deficiency in population-based colorectal and endometrial cancer with Lynch-like syndrome

Abstract Background: Tumour mismatch repair (MMR) deficiency, determined by immunohistochemical (IHC) loss of MMR protein expression, is used diagnostically to identify individuals with Lynch syndrome. A high proportion of colorectal cancers (CRCs) and endometrial cancers (ECs) that demonstrate tumor MMR-deficiency are categorised as having “Lynch-like syndrome” due to the absence of tumor MLH1 methylation or germline MMR gene mutations after standard screening approaches. The aim of this study was to investigate somatic causes of tumor MMR-deficiency in patients with Lynch-like syndrome. Methods: Population-based participants with incident MMR-deficient colorectal (n=193; ACCFR and MCCS) or endometrial cancer (n=197; ANECS and MCCS) were categorised as either Lynch syndrome, MLH1 methylated or Lynch-like after screening for germline MMR gene mutations and for tumor MLH1 gene promoter hypermethylation. Lynch-like tumors were tested for somatic MMR gene mutations (point mutations and loss of heterozygosity) using AmpliSeq-Ion Proton custom capture sequencing and for MSH2 or MSH6 gene promoter methylation. Overall survival for molecularly defined subgroups of Lynch-like CRCs were compared to Lynch syndrome related CRCs using cox regression models to estimate hazard ratios (HR) and 95% confidence intervals adjusting for age at CRC diagnosis, sex, AJCC stage and grade. Results: Lynch-like tumors comprised 32% (63/193) and 23% (45/197) of the MMR-deficient CRCs and ECs, respectively compared with 27% and 15% for Lynch syndrome and 41% and 62% for MLH1 methylated CRCs and ECs, respectively. Two somatic mutations were identified in the MMR gene indicated by the pattern of MMR IHC loss of expression were identified in 36.7% (18/49) and 47.8% (11/23) of the Lynch-like CRCs and ECs tested. The proportion of tumors with double somatic alterations was highest for both CRC and EC tumors showing MSH2-deficiency (40% and 64.3%). The mean age at diagnosis for the Lynch-like CRCs with double somatic mutations was 49.7 ± 15.8years which was not significantly different from the Lynch syndrome CRCs (n=52; 45.4 ± 11.3years; p=0.2) but was significantly different compared with the MLH1 methylated CRCs (n=83; 70 ± 8.9years; p=0.0001). The adjusted HR for double somatic Lynch-like CRCs was 2.58 (95% CI, 0.77-8.67) compared with Lynch syndrome CRCs (p=0.1). No evidence of tumor MSH2 or MSH6 gene promoter methylation was identified in either MSH2-deficient or MSH6-deficient Lynch-like CRCs or ECs tested (n=34 and n=12, respectively). Conclusions: Double somatic mutations in the MMR genes represent a significant proportion of the unexplained Lynch-like MMR-deficient subtype for both population-based CRC and EC. Triaging strategies used to identify Lynch syndrome for both CRC and EC should include tumor testing for somatic mutations in the MMR genes. Note: This abstract was not presented at the meeting. Citation Format: Daniel D. Buchanan, Mark Clendenning, Harindra Jayasekara, Jihoon E. Joo, Ee M. Wong, Melissa C. Southey, Rhiannon J. Walters, Bernard J. Pope, Aung K. Win, John L. Hopper, Mark A. Jenkins, Roger L. Milne, Graham G. Giles, Dallas R. English, Finlay A. Macrae, Amanda B. Spurdle, Ingrid M. Winship, Christophe Rosty. Double somatic mutations as a cause of tumor mismatch repair-deficiency in population-based colorectal and endometrial cancer with Lynch-like syndrome [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4266. doi:10.1158/1538-7445.AM2017-4266

Clinical characterization of pancreatic ductal adenocarcinomas (PDAC) with mismatch repair (MMR) gene mutations.

e15791 Background: Tumors with mismatch repair-deficiency (MMRD) have a high mutational burden and have good responses to immunotherapy (Le, NEJM, 2015). We describe the natural course, clinicopathological, and genomic status of MMRD PDAC patients (pts) at Memorial Sloan Kettering Cancer Center (MSKCC). Methods: MSKCC institutional registry and ICD billing database queried from 2006-2016 for PDAC pts with genetically confirmed mutations in mismatch repair (MMR) genes. Mutation # determined via MSK-IMPACT, a targeted tumor next generation sequencing (NGS) test (Cheng, J Mol Diagn, 2015). Results: 5/607 (0.8%) PDAC pts had Lynch syndrome (LS) (confirmed germline mutations) (Table 1). Of the 5 LS pts, all had > 10 mutations in NGS, with 4 of 5 having > 50 mutations. 4 of 5 (80%) are alive at last follow-up (survival 30-314 months). N=4 had extensive personal/family history of cancer. Of N=3 who had resected disease, all 3 had recurrence at 11, 49 and 311 months, and all are alive (survival: 69-314 months). Of N= 2 pts that had unresectable tumors, one passed away at 30 months while the other is on checkpoint inhibitor trial and is alive at 30 months. In contrast, 7/607 (1.1%) PDAC pts had somatic mutations in MMR genes with an average of 5.7 mutations in NGS, with 4/7 having <5 mutations. 4/7 (57%) are deceased at last follow-up (survival: 10-42 months). Conclusions: All cases with germline mutations in the MMR genes, with one exception, had high mutation #. All cases with somatic mutations in the MMR genes had low mutation #. Germline mutations in MMR genes and high mutational burden may predict for a prognostically favorable subgroup of PDAC pts with high susceptibility to immune oncology agents. [Table: see text]

Double somatic mutations in mismatch repair genes are frequent in colorectal cancer after Hodgkin's lymphoma treatment

ObjectiveHodgkin's lymphoma survivors who were treated with infradiaphragmatic radiotherapy or procarbazine-containing chemotherapy have a fivefold increased risk of developing colorectal cancer (CRC). This study aims to provide insight into the...

Somatic MMR gene mutations as a cause for MSI-H sebaceous neoplasms in Muir-Torre syndrome-like patients.

Sebaceous neoplasms are a major clinical feature of Muir-Torre syndrome (MTS) associated with visceral malignancies, especially colorectal and endometrial tumors. The diagnosis of MTS relies largely on the microsatellite instability (MSI) phenotype in tumors, suggesting germline mutations in DNA mismatch repair (MMR) genes responsible for the inherited disease. We hypothesized that in some MSI-H sebaceous tumors, acquired rather than inherited mutations in MMR genes could be involved. Using next-generation sequencing, we screened MMR gene mutations in 18 MSI-H sebaceous tumors. We found mutations in 17 samples (94%). Indeed, 12/17 (71%) were shown to carry acquired somatic mutations and among 12 samples, seven were shown to be associated with additional somatic alterations like loss of heterozygosity or multiple mutations, suggesting somatic second hits. Our findings strongly suggest that somatic MMR deficiency is responsible for a proportion of MSI-H sebaceous tumors.

Colon and Endometrial Cancers With Mismatch Repair Deficiency Can Arise From Somatic, Rather Than Germline, Mutations

Patients with Lynch syndrome carry germline mutations in single alleles of genes encoding the mismatch repair (MMR) proteins MLH1, MSH2, MSH6, and PMS2; when the second allele becomes mutated, cancer can develop. Increased screening for Lynch syndrome has identified patients with tumors that have deficiency in MMR, but no germline mutations in genes encoding MMR proteins. We investigated whether tumors with deficient MMR had acquired somatic mutations in patients without germline mutations in MMR genes using next-generation sequencing. We analyzed blood and tumor samples from 32 patients with colorectal or endometrial cancer who participated in Lynch syndrome screening studies in Ohio and were found to have tumors with MMR deficiency (based on microsatellite instability and/or absence of MMR proteins in immunohistochemical analysis, without hypermethylation of MLH1), but no germline mutations in MMR genes. Tumor DNA was sequenced for MLH1, MSH2, MSH6, PMS2, EPCAM, POLE, and POLD1 with ColoSeq and mutation frequencies were established. Twenty-two of 32 patients (69%) were found to have 2 somatic (tumor) mutations in MMR genes encoding proteins that were lost from tumor samples, based on immunohistochemistry. Of the 10 remaining tumors 3 had one somatic mutation in a MMR gene, with possible loss of heterozygosity that could lead to MMR deficiency, 6 were found to be false-positive results (19%), and 1 had only one mutation in a MMR gene and remained unexplained. All of the tumors found to have somatic MMR mutations were of the hypermutated phenotype (>12 mutations/megabase); 6 had mutation frequencies >200/megabase, and 5 of these had somatic mutations in POLE, which encodes a DNA polymerase. Some patients are found to have tumors with MMR defects during screening for Lynch syndrome, yet have no identifiable germline mutations in MMR genes. We found that almost 70% of these patients acquire somatic mutations in MMR genes, leading to a hypermutated phenotype of tumor cells. Patients with colon or endometrial cancers with MMR deficiency not explained by germline mutations might undergo analysis for tumor mutations in MMR genes to guide future surveillance guidelines.

Somatic mutations in mismatch repair genes in sporadic gastric carcinomas are not a cause but a consequence of the mutator phenotype

In hereditary nonpolyposis colorectal cancer (HNPCC), patients' mismatch repair (MMR) gene mutations cause MMR deficiency, leading to microsatellite instability (MSI-H). MSI-H is also found in a substantial fraction of sporadic gastric carcinomas (SGC), mainly due to MLH1 promoter hypermethylation, although somatic mutations in MMR genes have been described. We aimed to investigate which MMR defects are present in SGC. Twenty-nine MSI-H SGC investigated previously for MLH1 promoter hypermethylation were screened for somatic mutations in MLH1, MSH2, MSH6, MLH3, and MBD4 by denaturing gradient gel electrophoresis and sequencing. Five truncating mutations (three in MSH6, one in MLH3, and one in MBD4) and one missense mutation ( MLH1) were identified. Of these, three truncating mutations were in MSI-H cases that lack MLH1 hypermethylation. As all truncating mutations were found in the coding poly-A tracts, it seems likely that they result from the MSI phenotype rather than cause it. In summary, somatic mutations in MMR genes are rare in SGC and do not explain the development of these tumors reflecting, rather than causing, the mutator phenotype. Other MMR genes are probably involved in MSI-H gastric cancer without MLH1 hypermethylation.