Constitutional MLH1 Methylation Is a Major Contributor to Mismatch Repair–Deficient, MLH1-Methylated Colorectal Cancer in Patients Aged 55 Years and Younger

Authors:
Megan P. Hitchins Bioinformatics and Functional Genomics Center, Department of Biomedical Sciences, Cedars-Sinai Cancer, Cedars-Sinai Medical Center, Los Angeles, California
Department of Medicine (Oncology), Stanford University, Stanford, California

Search for other papers by Megan P. Hitchins in
Current site
Google Scholar
PubMed
Close
 PhD
,
Estela Dámaso Department of Medicine (Oncology), Stanford University, Stanford, California
Hereditary Cancer Program, Catalan Institute of Oncology, Institut d'Investigació Biomèdica de Bellvitge (IDIBELL), ONCOBELL Program, L’Hospitalet de Llobregat, Barcelona, Spain

Search for other papers by Estela Dámaso in
Current site
Google Scholar
PubMed
Close
 PhD
,
Rocio Alvarez Bioinformatics and Functional Genomics Center, Department of Biomedical Sciences, Cedars-Sinai Cancer, Cedars-Sinai Medical Center, Los Angeles, California

Search for other papers by Rocio Alvarez in
Current site
Google Scholar
PubMed
Close
 MS
,
Lisa Zhou Bioinformatics and Functional Genomics Center, Department of Biomedical Sciences, Cedars-Sinai Cancer, Cedars-Sinai Medical Center, Los Angeles, California

Search for other papers by Lisa Zhou in
Current site
Google Scholar
PubMed
Close
 MS
,
Yajing Hu Department of Medicine (Oncology), Stanford University, Stanford, California

Search for other papers by Yajing Hu in
Current site
Google Scholar
PubMed
Close
 PhD
,
Marcio A. Diniz Biostatistics and Bioinformatics Research Center, Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, California

Search for other papers by Marcio A. Diniz in
Current site
Google Scholar
PubMed
Close
 PhD
,
Marta Pineda Hereditary Cancer Program, Catalan Institute of Oncology, Institut d'Investigació Biomèdica de Bellvitge (IDIBELL), ONCOBELL Program, L’Hospitalet de Llobregat, Barcelona, Spain
Consortium for Biomedical Research in Cancer – CIBERONC, Carlos III Institute of Health, Madrid, Spain

Search for other papers by Marta Pineda in
Current site
Google Scholar
PubMed
Close
 PhD
,
Gabriel Capella Hereditary Cancer Program, Catalan Institute of Oncology, Institut d'Investigació Biomèdica de Bellvitge (IDIBELL), ONCOBELL Program, L’Hospitalet de Llobregat, Barcelona, Spain
Consortium for Biomedical Research in Cancer – CIBERONC, Carlos III Institute of Health, Madrid, Spain

Search for other papers by Gabriel Capella in
Current site
Google Scholar
PubMed
Close
 MD, PhD
,
Rachel Pearlman Department of Internal Medicine, Ohio State University, Columbus, Ohio
The Comprehensive Cancer Center, Ohio State University, Columbus, Ohio

Search for other papers by Rachel Pearlman in
Current site
Google Scholar
PubMed
Close
 MS, CGC
, and
Heather Hampel Department of Internal Medicine, Ohio State University, Columbus, Ohio
The Comprehensive Cancer Center, Ohio State University, Columbus, Ohio
Division of Clinical Cancer Genomics, Department of Medical Oncology and Therapeutics Research, City of Hope National Medical Center, Duarte, California

Search for other papers by Heather Hampel in
Current site
Google Scholar
PubMed
Close
 MS, CGC
Full access

Background: Most mismatch repair–deficient (MMRd) colorectal cancer (CRC) cases arise sporadically, associated with somatic MLH1 methylation, whereas approximately 20% have germline mismatch repair pathogenic variants causing Lynch syndrome (LS). Universal screening of incident CRC uses presence of MLH1 methylation in MMRd tumors to exclude sporadic cases from germline testing for LS. However, this overlooks rare cases with constitutional MLH1 methylation (epimutation), a poorly recognized mechanism for LS. We aimed to assess the frequency and age distribution of constitutional MLH1 methylation among incident CRC cases with MMRd, MLH1-methylated tumors. Methods: In retrospective population-based studies, we selected all CRC cases with MMRd, MLH1-methylated tumors, regardless of age, prior cancer, family history, or BRAF V600E status, from the Columbus-area HNPCC study (Columbus) and Ohio Colorectal Cancer Prevention Initiative (OCCPI) cohorts. Blood DNA was tested for constitutional MLH1 methylation by pyrosequencing and real-time methylation-specific PCR, then confirmed with bisulfite-sequencing. Results: Results were achieved for 95 of 98 Columbus cases and all 281 OCCPI cases. Constitutional MLH1 methylation was identified in 4 of 95 (4%) Columbus cases, ages 34, 38, 52, and 74 years, and 4 of 281 (1.4%) OCCPI cases, ages 20, 34, 50, and 55 years, with 3 showing low-level mosaic methylation. Mosaicism in blood and normal colon, plus tumor loss of heterozygosity of the unmethylated allele, demonstrated causality in 1 case with sample availability. Age stratification showed high rates of constitutional MLH1 methylation among younger patients. In the Columbus and OCCPI cohorts, respectively, these rates were 67% (2 of 3) and 25% (2 of 8) of patients aged <50 years but with half of the cases missed, and 75% (3 of 4) and 23.5% (4 of 17) of patients aged ≤55 years with most cases detected. Conclusions: Although rare overall, a significant proportion of younger patients with MLH1-methylated CRC had underlying constitutional MLH1 methylation. Routine testing for this high-risk mechanism is warranted in patients aged ≤55 years for a timely and accurate molecular diagnosis that will significantly alter their clinical management while minimizing additional testing.

Background

Approximately 15% of colorectal cancer (CRC) cases are mismatch repair (MMR)–deficient (MMRd), detected by immunohistochemistry (IHC),1 microsatellite instability (MSI),2 and/or the hypermutator phenotype.3 Most MMRd CRC cases arise sporadically in older patients, associated with acquired (somatic in origin) aberrant methylation of the MLH1 promoter in the tumor.46 Approximately 20% of MMRd CRC cases are associated with Lynch syndrome (LS), caused by a germline pathogenic variant within an MMR gene, MLH1, MSH2, MSH6, PMS2, or terminal EPCAM deletion.7 Constitutional MLH1 methylation (epimutation), characterized by promoter methylation and transcriptional inactivation of a single allele throughout normal tissues, is an alternative and poorly recognized mechanism for LS.8 Cases with constitutional MLH1 methylation have presented with early-onset and/or multiple primary tumors consistent with MLH1-LS9; hence, clinical management according to MLH1-LS guidelines has been suggested for this subgroup.10

Universal screening of all CRC for MMRd is recommended as standard of care.1116 This stratifies patients by prediction of response to systemic therapies and identifies those warranting genetic testing for LS.6,17,18 Current stepwise algorithms entail tumor testing for MMRd by IHC and/or MSI in the first tier.12 For tumors exhibiting MLH1 loss and/or MSI, second-tier testing for MLH1 methylation is recommended, and cases with MLH1 methylation are considered sporadic and hence omitted from follow-up evaluation for LS.15,16,1922 The BRAF V600E tumor mutation is sometimes used as a surrogate for MLH1 methylation, given that it correlates, albeit imperfectly, with somatic-in-origin MLH1 methylation and is rare in LS-associated CRC.2225 Cases with MMR protein loss and unmethylated MLH1, or BRAF wild-type, are eligible for LS testing. Next-generation sequencing tumor panels that detect MSI, hypermutator phenotype, and BRAF V600E may replace these stepwise algorithms.26 A potential problem with these algorithms is that they may misdiagnose cases with constitutional MLH1 methylation as common sporadic cases, thereby omitting them from appropriate follow-up testing. In these high-risk cases, the constitutional MLH1 methylation serves as the “first hit,” predisposing to tumors that exhibit MLH1 loss, MSI, and MLH1 methylation—features that overlap with common sporadic cases.8,2734 Furthermore, most cases arise de novo due to a “primary epimutation” with no apparent genetic basis; consequently, carriers tend to have no remarkable family history.8,31,35,36 Although constitutional MLH1 methylation is potentially heritable, few familial cases have been described. These have included families with autosomal dominant inheritance linked to a genetic variant and others with non-Mendelian inheritance without an associated genetic variant.34,35,3746 Thus, differentiating seemingly sporadic cases with constitutional MLH1 methylation from “true sporadic” cases with somatic MLH1 methylation poses a clinical and molecular diagnostic challenge.

Constitutional MLH1 methylation is rare in CRC overall.47 However, its frequency and age distribution among incident MMRd, MLH1-methylated CRC cases remains unclear because prior screens have been conducted primarily in single-center studies on small sample sizes. Ascertainment biases in estimations of frequency are likely, given the use of different inclusion criteria with respect to age, clinicopathologic characteristics including BRAF V600E status, and prior genetic testing. The frequency of constitutional MLH1 methylation was 3% to 9%8,30,35,41,4750 in patients ascertained via cancer clinics as fulfilling at least one of the clinical criteria in the revised Bethesda guidelines for LS evaluation,51 with MSI and MLH1 loss in their tumor and negative germline test results. Marginally higher rates of constitutional MLH1 methylation were found with inclusion of MLH1 methylation as a tumor feature, at 3.5% to 15.6%.33,47,52

In this study, we leveraged the biospecimen resources and prior clinicopathologic and molecular results from the population-based Columbus-area HNPCC study (Columbus)53,54 and Ohio Colorectal Cancer Prevention Initiative (OCCPI)55 cohorts, designed for prospective studies on universal screening for hereditary cancer, to determine the frequency and age distribution of constitutional MLH1 methylation among incident CRC cases selected solely by the tumor features of MMRd and presence of MLH1 methylation.

Methods

Patients

The patient selection strategy is shown in supplemental eFigure 1 (available with this article at www.JNCCN.org). The first case series was derived from 1,566 unselected CRC cases in the Columbus cohort (recruited 1999–2004).53,54 The second series was derived from 3,310 CRC cases in the OCCPI cohort (recruited 2013–2016).55 Eligibility criteria for inclusion in this study were patients whose CRC showed (1) MLH1 absence and/or MSI-high if IHC was missing/uninterpretable, (2) MLH1 methylation, and (3) availability of peripheral blood leukocyte (PBL) DNA (Columbus) or whole-blood DNA (OCCPI). We leveraged the existing tumor molecular pathology data from prior universal screening for MMRd and LS to select cases for inclusion, including MMR IHC, MSI status (Bethesda 5-marker panel),51 and MLH1 methylation status previously determined using methylation-specific PCR (Columbus)5,56,57 or CpG pyrosequencing58 (OCCPI) assays. All patients were deidentified before study initiation.

Screen for Constitutional MLH1 Methylation and Promoter Variants

Bisulfite conversion was performed on 1 μg of DNA using the EZ DNA Methylation-Gold Kit (Zymo Research), with approximately 50 ng input into each assay. Screening for constitutional MLH1 methylation was performed using 2 assays previously described for this purpose: (1) quantitative CpG pyrosequencing using the PyroMark Q96 system (Qiagen)31,36,59 and (2) high-sensitivity real-time semi-quantitative methylation-specific PCR (qMSP) followed by melt curve analysis (methylation-specific peak at 76.5°C ± 0.5°C) using the CFX96 Thermal Cycler system (Bio-Rad Laboratories), enabling the detection of low-level mosaicism.3436,42,60 Samples were considered methylation-positive if either assay yielded a signal above the respective limit of detection, at 2.3% for pyrosequencing and percentage of methylated reference (PMR) value at 0.1% for qMSP (supplemental eFigures 2 and 3).42,61 Presence of methylation was confirmed by clonal bisulfite sequencing (≥24 clones) across a promoter fragment encompassing 16 CpGs and single-nucleotide polymorphism (SNP) c.-93G>A (rs1800734).3436,61 This also allowed allele-specific methylation patterns to be determined in patients heterozygous for this SNP. The MLH1 CpG island was Sanger sequenced in methylation-positive cases to identify potential genetic variants associated with constitutional MLH1 methylation.34,42

Statistics

Descriptive statistics were performed using SPSS Statistics, Version 27. Based on a Bayesian approach, posterior probabilities and 95% high-density probability confidence intervals for various rates of detection were calculated as described in supplemental eAppendix 1.

Results

Detection of Constitutional MLH1 Methylation

Columbus Cohort

Among 105 of 1,566 (6.7%) eligible cases, 98 cases with MLH1 loss (n=95; 3 had missing data) and/or MSI-high (n=98) plus MLH1 methylation in their tumor had PBL DNA available (supplemental eFigure 1). Median age at diagnosis was 72 years (range, 34–98 years). Methylation testing was successful for 95 cases, with complete concordance between pyrosequencing and qMSP results. Constitutional MLH1 methylation was detected in 4 of 95 (4%) of the selected cases, aged 34, 36, 52, and 74 years at diagnosis (Figure 1). Clonal bisulfite sequencing confirmed the presence of constitutional MLH1 methylation affecting a proportion of alleles in PBL DNA from all 4 cases (supplemental eFigure 6). The levels and allelic patterns of constitutional MLH1 methylation and clinical, demographic, and molecular pathology features are summarized in Table 1. Patient Columbus-6 had a classic pattern of hemiallelic methylation (approximately half of alleles methylated), consistent with an epimutation affecting a single parental allele. Notably, the other 3 patients (Columbus-1, -2, and -65) had low-level methylation consistent with mosaicism affecting a proportion of cells (Table 1, supplemental eFigure 6). CRC was the first presentation of cancer in all 4 cases. No genetic alterations were identified within the MLH1 CpG island in the 4 carriers. Only Columbus-65 had a positive family history recorded, which included CRC and/or endometrial cancer in a first-degree relative, but precise details are unknown.

Figure 1.
Figure 1.

Blood-based detection of constitutional MLH1 methylation by CpG pyrosequencing in CRC cases with MMRd, MLH1-methylated tumors. Pyrogram traces are shown for patients with CRC in whom constitutional MLH1 methylation was detected across 5 CpG sites within the c.-241 to c.-272 sequence (GenBank accession no. NG_007109.2) of the MLH1 promoter in bisulfite-converted germline DNA. (A) MLH1 methylation was detected in PBL DNA of 4 cases from the Columbus cohort. (B) MLH1 methylation was detected in whole-blood DNA from 4 cases from the OCCPI cohort. (C) Illustrative examples of pyrograms from the hypermethylated RKO CRC cell line as positive control and an unmethylated healthy control PBL as a negative control. Methylation is detected by the presence of a peak at the cytosine (C) within each CpG site interrogated (blue bars), whereas unmethylated cytosines are detected as thymine (T) peaks within the same CpG sites, due to the conversion of unmethylated cytosines to uracils using bisulfite treatment. The percentage methylation value at each CpG site appears above each bar. The mean percentage of methylation across all 5 CpG sites is calculated and shown above each pyrogram. The yellow bar indicates a non-CpG cytosine used as a quality control measure to ensure complete bisulfite conversion to T, whereupon this yields a valid test result.

Abbreviations: Columbus, Columbus-area HNPCC study; CRC, colorectal cancer; MMRd, mismatch repair–deficient; OCCPI, Ohio Colorectal Cancer Prevention Initiative; PBL, peripheral blood leukocyte.

Citation: Journal of the National Comprehensive Cancer Network 21, 7; 10.6004/jnccn.2023.7020

Table 1.

Demographic and Clinicopathologic Features of CRC Cases With Constitutional MLH1 Methylation Identified as Likely Cause for Their MMRd, MLH1-Methylated Tumor

Table 1.

OCCPI Cohort

Eligible CRC cases included 281 of 3,310 (8.5%) with MLH1 loss (n=280, 1 equivocal was MSI-high) plus MLH1 methylation in their tumor and had whole-blood DNA available (supplemental eFigure 1).55 Median age at diagnosis was 71.7 years (range, 20–93 years). Methylation testing was successful for all 281 cases, with complete concordance between pyrosequencing and qMSP results. Constitutional MLH1 methylation was detected in 4 of 281 (1.4%) selected cases, aged 20, 34, 50, and 55 years (Figure 1). These were confirmed by clonal bisulfite sequencing (supplemental eFigure 6), as summarized in Table 1. CRC was the first presentation of cancer in all 4 cases. All 4 cases had an unremarkable family history, and no genetic alterations were identified within the MLH1 CpG island. Patients OCCPI-2, -10, and -15 had classic hemiallelic constitutional MLH1 methylation, with methylation levels of approximately 50% measured by pyrosequencing (Figure 1B) and about half of alleles found to be methylated by clonal bisulfite sequencing (supplemental eFigure 6), consistent with epimutation of one parental allele. OCCPI-1 had low-level mosaic (4.2%) constitutional MLH1 methylation in blood (Figure 1B); he was heterozygous for the c.-93G>A promoter SNP, with methylation affecting about 20% of “G” alleles (supplemental eFigure 6). Together, these data are consistent with monoallelic, mosaic constitutional MLH1 methylation affecting approximately 8% to 10% of leukocytes. Tumor and paired normal colorectal mucosa (NCM) DNA samples were retrieved for OCCPI-1. Pyrosequencing detected similarly low-level mosaic MLH1 methylation (2.6%) in NCM, therefore affecting approximately 5% NCM cells (Figure 2A). In the tumor, a high level of methylation (61%) was detected (Figure 2A), and clonal bisulfite sequencing found only “G” alleles at c.-93G>A, which were predominantly methylated (Figure 2B). Sequencing across the MLH1 promoter showed loss of heterozygosity (LoH) of the unmethylated “A” allele at c.-93G>A in the tumor (Figure 2C). These findings are consistent with constitutional MLH1 methylation of the “G” allele in a small proportion of cells as the “first hit” predisposing to tumorigenesis, accompanied by somatic LoH of the unmethylated “A” allele as the “second hit.”

Figure 2.
Figure 2.

Identification of LoH of the unmethylated allele in the tumor of patient OCCPI-1 indicates low-level mosaic constitutional MLH1 methylation predisposed to CRC development. (A) Pyrosequencing in NCM shows low-level methylation at 2.6% in the tissue of tumor origin and confirms high-level methylation in the tumor (60.8%). Legend according to Figure 1. (B) Clonal bisulfite sequencing in the tumor. Legend according to supplemental eFigure 6. Although patient OCCPI-1 was heterozygous at the c.-93G>A SNP (rs1800734), only detected “G” alleles were sequenced in the tumor, most of which were methylated. The methylation patterns on individual alleles were repetitive, suggesting clonal expansion from a small number of cells that contained methylated G alleles. (C) Sanger sequencing electropherogram across the c.-93G>A SNP site within the MLH1 promoter is shown in the tumor and paired NCM samples, with the RFU measured at the A and G peaks at the c.-93G>A SNP (indicated by arrows), shown for each allele below. The LoH index in the tumor was calculated with reference to the paired NCM sample as (A/GTumor)/(A/GNCM), whereby values <0.6 or >1.7 indicate LoH of the A or G allele, respectively. NCM has approximately equal representation of the G and A alleles, consistent with heterozygosity. The tumor shows reduced representation of the A allele (LoH index 0.23), consistent with LoH of the A allele as the “second hit” in the tumor. See supplemental eFigure 6, available with this article at JNCCN.org.

Abbreviations: CRC, colorectal cancer; LoH, loss of heterozygosity; NCM, normal colorectal mucosa; OCCPI, Ohio Colorectal Cancer Prevention Initiative; RFU, relative fluorescence units; SNP, single-nucleotide polymorphism.

Citation: Journal of the National Comprehensive Cancer Network 21, 7; 10.6004/jnccn.2023.7020

Stratification by Age

To determine the optimal age threshold at which screening yielded the highest rate of detection of constitutional MLH1 methylation with minimal cases missed while also minimizing the number of cases screened overall, we calculated the frequency of constitutional MLH1 methylation by age and within 5-year age bins for each cohort (Figure 3, Table 2).

Figure 3.
Figure 3.

Rate of detection of constitutional MLH1 methylation by age among the Columbus and OCCPI cohorts. (A) Histograms showing the number of cases with constitutional MLH1 methylation detected (orange) among CRC cases with an MMRd, MLH1-methylated tumor (blue) tested within 5-year age bins beginning at age <50 years for each cohort. (B) Line graphs showing the cumulative percentage of CRC cases tested by age (blue), beginning at age <50 years pooled, with 100% of cases being n=95 for the Columbus cohort and n=281 for the OCCPI cohort. The percentage of cases with constitutional MLH1 methylation detected among cases tested of the same age, beginning at age <50 years, are shown in orange. In the Columbus cohort the highest detection rate was 75% (3 of 4) at age threshold <55 years, with 1 older case missed. There were no cases aged 55 years at diagnosis in the Columbus cohort; therefore, the same detection rate was observed at ≤55 years. In the OCCPI cohort, the highest detection rate was 25% (2 of 8) cases <50 years, but with 2 cases missed. The youngest age threshold at which all cases were detected was age ≤55 years (<56 years), yielding a detection rate of 23.5% (4 of 17 cases). In both cohorts, screening at age ≤55 years would entail screening of a small fraction of total cases: 4% for the Columbus cohort and 6% of the OCCPI cohort.

Abbreviations: Columbus, Columbus-area HNPCC study; CRC, colorectal cancer; MMRd, mismatch repair–deficient; OCCPI, Ohio Colorectal Cancer Prevention Initiative.

Citation: Journal of the National Comprehensive Cancer Network 21, 7; 10.6004/jnccn.2023.7020

Table 2.

Detection Rate of Constitutional MLH1 Methylation as a Function of Age Group Among Patients With MMRd, MLH1-Methylated CRC

Table 2.

Columbus Cohort

Of the 4 cases with constitutional MLH1 methylation detected, 2 were aged <50 years, an age bin comprising only 3 of 95 (3%) of the cohort, yielding a positive detection rate of 2 of 3 (67%), but with 2 older cases missed. The only patient with unmethylated PBL in the <50 years bin carried the 5′ untranslated region promoter variant c.-11C>T (rs776898290).54 This variant has been identified in additional suspected LS cases and showed significantly diminished transcriptional activity in promoter reporter assays,42 but it remains a variant of uncertain significance. The highest positive detection rate for constitutional MLH1 methylation at 3 of 4 (75%) was <55 years, an age threshold that comprised only 4 of 95 (4%) of the cohort, but with 1 older case missed. To detect all 4 cases, screening of 58 of 95 (61%) cases <75 years of age yielded a detection rate of 4 of 58 (7%).

OCCPI Cohort

Constitutional MLH1 methylation was detected in 2 of 8 (25%) patients aged <50 years, an age bin comprising just 8 of 281 (2.8%) of the cohort, but with 2 older cases missed. The detection rate was 3 of 14 (21%) patients aged <55 years, an age threshold that comprised 14 of 281 (5%) of the cohort, but with 1 case (aged 55 years) missed. Increasing the age threshold to <60 years entailed screening 31 of 281 (11%) patients in the cohort, yielding a detection rate of 4 of 31 (13%), with all 4 cases detected. The youngest age threshold at which all 4 cases were detected was ≤55 years, which entailed screening of 17 of 281 (6%) of the cohort, yielding a positive detection rate of 4 of 17 (23.5%).

Posterior probabilities for various rates of detection were calculated for each cohort to assess the age limit at which an acceptable detection rate was yielded (supplemental eFigure 7 and eTable 1). Assuming testing for constitutional MLH1 methylation is considered acceptable when it yields a detection rate >10% with very high probability, the age threshold for testing implementation would be <60 years for the Columbus-area cohort (probability >0.99) and ≤55 years for the OCCPI cohort (probability 0.96).

Discussion

Current universal testing algorithms may result in patients with constitutional MLH1 methylation going underdiagnosed and underserved as the opportunity for a clinical and molecular diagnosis at first presentation of cancer is missed. For routine implementation of additional blood-based methylation testing into the existing universal screening algorithm for MMRd and LS, multiple factors would be taken into consideration, including the rate/probability of detection among those selected for screening and the number (and cost) of additional tests needed to identify each new case. Our study aimed to address the frequency of constitutional MLH1 methylation among incident MMRd/MLH1-methylated CRC cases and whether age at diagnosis could be incorporated into existing algorithms to select patients warranting referral for additional methylation testing to detect this rare high-risk defect.

Our retrospective nested study of the population-based Columbus and OCCPI cohorts selected CRC cases based solely on tumor molecular features (MLH1 loss/MSI-high and MLH1 methylation) used in universal screening algorithms; therefore, it was unbiased with respect to age at diagnosis, tumor BRAF V600E status, personal or family history of cancer, or whether prior germline genetic testing had been undertaken. With selection by these key tumor features, constitutional MLH1 methylation was detected in just 4% (4/95) of Columbus and 1.4% (4/281) of OCCPI cases. This low frequency across all ages is unsurprising, given the rarity of this defect and because MLH1 methylation is typically somatic in origin in CRC and associated with advanced age. Although the full cohorts of unselected cases were not screened, by extrapolation this makes constitutional MLH1 methylation extremely rare among unselected CRC series, at 4 of 1,566 (∼0.26%) of Columbus cases and 4 of 3,310 (∼0.12%) of OCCPI cases. This mechanism also represents a minor fraction (<10%) of LS cases, because MMR germline pathogenic variants were previously identified in 2.8% of Columbus cases and 4% of OCCPI cases.5355 Nevertheless, a key finding of this study was that after stratification by age at diagnosis, constitutional MLH1 methylation was a major contributor to incident MMRd, MLH1-methylated CRC among younger cases, which, in turn, represented a small fraction of cases overall. This shows that selection by key tumor features with age limitation would incur screening of small numbers. The highest rates of detection of patients with constitutional MLH1 methylation were 75% (3/4 tested) of those aged <55 years in the Columbus cohort and 25% (2/8 tested) of those aged <50 years in the OCCPI cohort; however, these age thresholds resulted in missed cases. In the OCCPI cohort, increasing the age of testing to ≤55 years maintained a high detection rate at 23.5% (4/17 tested), with all 4 cases with constitutional MLH1 methylation detected and minimal additional screening. The 2022 NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Genetic/Familial High-Risk Assessment: Colorectal recommend that testing for constitutional MLH1 methylation be considered in patients with CRC whose tumor exhibited MLH1 loss and MLH1 methylation if it is early onset (occurs at <50 years of age) or they have a family history.16 However, in the 2 cohorts we studied, restricting testing to patients aged <50 years would have resulted in half (4/8) of cases being missed. Furthermore, probands with constitutional MLH1 methylation often have no family history, as was found for 7 of 8 cases herein, given that epimutations tend to arise de novo. Only the oldest patient had a family history but no apparent underlying genetic variant. The finding of this 74-year-old patient with constitutional MLH1 methylation illustrates that age-limited screening will result in occasional missed cases. However, given the rarity of this defect, triaging is needed to increase detection rates for routine implementation of screening. By using age ≤55 years at diagnosis as the threshold for referral of incident cases of CRC with MMRd, MLH1-methylated tumors for blood-based methylation testing, most cases with constitutional MLH1 methylation herein would have been identified with high detection rates (23.5% Columbus, 75% OCCPI) and with minimal cases screened (4% Columbus, 6% OCCPI with MMRd/MLH1-methylated tumors). This would provide a timely and correct molecular diagnosis of this defect at first presentation of CRC at an age when intensive surveillance for MLH1-LS cancers could result in the prevention or earlier diagnosis of metachronous cancer and potential life-years saved.

The role of low-level mosaic constitutional MLH1 methylation in cancer predisposition has remained in question. An interesting observation herein was that 3 patients had low-level mosaic constitutional MLH1 methylation. Although it is plausible that this may confer lower risks for cancer than hemiallelic methylation, mosaicism was found in both the oldest (age 74 years) and the youngest (age 20 years) patients. Prior cases with low-level mosaicism have been reported with early-onset and/or multiple primary cancers.42,62 Although additional samples were unavailable from Columbus cases, for patient OCCPI-1 (age 20 years), we were able to demonstrate that methylation was present in NCM (tumor tissue of origin), albeit also low level. Furthermore, by tracing allelic representation and methylation at the c.-93G>A SNP in tumor and NCM, we found high levels of monoallelic methylation in the tumor combined with LoH of the unmethylated allele. These findings are consistent with a causal role for mosaic constitutional MLH1 methylation in early-onset tumorigenesis, whereby monoallelic methylation in the cell(s) of origin is predisposing, followed by somatic LoH of the remaining functional allele.

A limitation of this study was the lack of systematic BRAF V600E tumor testing within the cohorts. For 4 of 8 constitutional MLH1 methylation carriers with available data, all 4 were BRAF wild-type (Table 1). This is consistent with prior case reports suggesting a higher frequency of BRAF wild-type, although a few cases with BRAF V600E CRC have been reported.31,45 Nevertheless, in the absence of tumor MLH1 methylation results or in universal screening scenarios where tumor BRAF V600E testing is used in place of MLH1 methylation testing, carriers of constitutional MLH1 methylation may undergo germline genetic testing for LS (on account of MLH1 loss plus BRAF wild-type in their tumor). This will likely yield a negative/uninformative germline test result. Therefore, testing for constitutional MLH1 methylation should also be considered in patients with CRC aged ≤55 years whose tumor showed MLH1 loss and BRAF wild-type (in the absence of tumor MLH1 methylation information) if germline genetic testing returns a nonactionable result to correctly diagnose or rule out constitutional MLH1 methylation. Finally, referral for constitutional MLH1 methylation testing should also be considered in patients with a clinical history of synchronous or metachronous LS-associated cancers displaying MMRd and MLH1 methylation at any age.

Conclusions

Our findings provide evidence in support of referral for blood-based testing for constitutional MLH1 methylation among patients with first presentation of CRC at age ≤55 years whose tumor exhibits MMRd and MLH1 methylation. Given the low prevalence of CRC cases with these tumor features at younger ages, additional methylation testing will be minimal yet likely to yield high rates of detection of constitutional MLH1 methylation. Currently, CLIA-certified testing for constitutional MLH1 methylation in the United States is available at the Mayo Clinic, Rochester, Minnesota, which uses a qMSP test similar to one of the screening tests used in this study.

Acknowledgments

We acknowledge the contribution of, and dedicate this work to, Dr. Albert de la Chapelle.

References

  • 1.

    Parsons R, Li GM, Longley MJ, et al. Hypermutability and mismatch repair deficiency in RER+ tumor cells. Cell 1993;75:12271236.

  • 2.

    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;58:52485257.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Yamamoto H, Imai K. An updated review of microsatellite instability in the era of next-generation sequencing and precision medicine. Semin Oncol 2019;46:261270.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Cunningham JM, Christensen ER, Tester DJ, et al. Hypermethylation of the hMLH1 promoter in colon cancer with microsatellite instability. Cancer Res 1998;58:34553460.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Herman JG, Umar A, Polyak K, et al. Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA 1998;95:68706875.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.

    Ward R, Meagher A, Tomlinson I, et al. Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut 2001;48:821829.

  • 7.

    Peltomäki P. Update on Lynch syndrome genomics. Fam Cancer 2016;15:385393.

  • 8.

    Hitchins M, Williams R, Cheong K, et al. MLH1 germline epimutations as a factor in hereditary nonpolyposis colorectal cancer. Gastroenterology 2005;129:13921399.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Hitchins MP. The role of epigenetics in Lynch syndrome. Fam Cancer 2013;12:189205.

  • 10.

    Lynch HT, Snyder CL, Shaw TG, et al. Milestones of Lynch syndrome: 1895–2015. Nat Rev Cancer 2015;15:181194.

  • 11.

    Burt RW, Cannon JA, David DS, et al. Colorectal cancer screening. J Natl Compr Canc Netw 2013;11:15381575.

  • 12.

    Beamer LC, Grant ML, Espenschied CR, et al. Reflex immunohistochemistry and microsatellite instability testing of colorectal tumors for Lynch syndrome among US cancer programs and follow-up of abnormal results. J Clin Oncol 2012;30:10581063.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Heald B, Plesec T, Liu X, et al. Implementation of universal microsatellite instability and immunohistochemistry screening for diagnosing Lynch syndrome in a large academic medical center. J Clin Oncol 2013;31:13361340.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Schofield L, Grieu F, Amanuel B, et al. Population-based screening for Lynch syndrome in Western Australia. Int J Cancer 2014;135:10851091.

  • 15.

    Ness RM, Llor X, Abbadessa B, et al. NCCN Clinical Practice Guidelines in Oncology: Colon Cancer. Version 3.2022. Accessed November 7, 2022. To view the most recent version, visit https://www.nccn.org

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16.

    Gupta S, Weiss JM, Axell L, et al. NCCN Clinical Practice Guidelines in Oncology: Genetic/Familial High-Risk Assessment: Colorectal. Version 2.2022. Accessed July 11, 2022. To view the most recent version, visit https://www.nccn.org

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med 2015;372:25092520.

  • 18.

    Webber EM, Kauffman TL, O’Connor E, et al. Systematic review of the predictive effect of MSI status in colorectal cancer patients undergoing 5FU-based chemotherapy. BMC Cancer 2015;15:156.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.

    Pérez-Carbonell L, Alenda C, Payá A, et al. Methylation analysis of MLH1 improves the selection of patients for genetic testing in Lynch syndrome. J Mol Diagn 2010;12:498504.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Nakagawa H, Nagasaka T, Cullings HM, et al. Efficient molecular screening of Lynch syndrome by specific 3′ promoter methylation of the MLH1 or BRAF mutation in colorectal cancer with high-frequency microsatellite instability. Oncol Rep 2009;21:15771583.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Gausachs M, Mur P, Corral J, et al. MLH1 promoter hypermethylation in the analytical algorithm of Lynch syndrome: a cost-effectiveness study. Eur J Hum Genet 2012;20:762768.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    McGivern A, Wynter CV, Whitehall VL, et al. Promoter hypermethylation frequency and BRAF mutations distinguish hereditary non-polyposis colon cancer from sporadic MSI-H colon cancer. Fam Cancer 2004;3:101107.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Loughrey MB, Waring PM, Tan A, et al. Incorporation of somatic BRAF mutation testing into an algorithm for the investigation of hereditary non-polyposis colorectal cancer. Fam Cancer 2007;6:301310.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    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;41:664668.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25.

    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;10:191195.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26.

    Hampel H, Pearlman R, Beightol M, et al. Assessment of tumor sequencing as a replacement for Lynch syndrome screening and current molecular tests for patients with colorectal cancer. JAMA Oncol 2018;4:806813.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27.

    Gazzoli I, Loda M, Garber J, et al. A hereditary nonpolyposis colorectal carcinoma case associated with hypermethylation of the MLH1 gene in normal tissue and loss of heterozygosity of the unmethylated allele in the resulting microsatellite instability-high tumor. Cancer Res 2002;62:39253928.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Miyakura Y, Sugano K, Akasu T, et al. Extensive but hemiallelic methylation of the hMLH1 promoter region in early-onset sporadic colon cancers with microsatellite instability. Clin Gastroenterol Hepatol 2004;2:147156.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Suter CM, Martin DI, Ward RL. Germline epimutation of MLH1 in individuals with multiple cancers. Nat Genet 2004;36:497501.

  • 30.

    Valle L, Carbonell P, Fernandez V, et al. MLH1 germline epimutations in selected patients with early-onset non-polyposis colorectal cancer. Clin Genet 2007;71:232237.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Goel A, Nguyen TP, Leung HC, et al. De novo constitutional MLH1 epimutations confer early-onset colorectal cancer in two new sporadic Lynch syndrome cases, with derivation of the epimutation on the paternal allele in one. Int J Cancer 2011;128:869878.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32.

    Auclair J, Vaissière T, Desseigne F, et al. Intensity-dependent constitutional MLH1 promoter methylation leads to early onset of colorectal cancer by affecting both alleles. Genes Chromosomes Cancer 2011;50:178185.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33.

    Pineda M, Mur P, Iniesta MD, et al. MLH1 methylation screening is effective in identifying epimutation carriers. Eur J Hum Genet 2012;20:12561264.

  • 34.

    Hitchins MP, Rapkins RW, Kwok CT, et al. Dominantly inherited constitutional epigenetic silencing of MLH1 in a cancer-affected family is linked to a single nucleotide variant within the 5'UTR. Cancer Cell 2011;20:200213.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35.

    Hitchins MP, Wong JJ, Suthers G, et al. Inheritance of a cancer-associated MLH1 germ-line epimutation. N Engl J Med 2007;356:697705.

  • 36.

    Hitchins MP, Owens SE, Kwok CT, et al. Identification of new cases of early-onset colorectal cancer with an MLH1 epimutation in an ethnically diverse South African cohort. Clin Genet 2011;80:428434.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37.

    Kwok CT, Vogelaar IP, van Zelst-Stams WA, et al. The MLH1 c.-27C>A and c.85G>T variants are linked to dominantly inherited MLH1 epimutation and are borne on a European ancestral haplotype. Eur J Hum Genet 2014;22:617624.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38.

    Leclerc J, Flament C, Lovecchio T, et al. Diversity of genetic events associated with MLH1 promoter methylation in Lynch syndrome families with heritable constitutional epimutation. Genet Med 2018;20:15891599.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39.

    Morak M, Schackert HK, Rahner N, et al. Further evidence for heritability of an epimutation in one of 12 cases with MLH1 promoter methylation in blood cells clinically displaying HNPCC. Eur J Hum Genet 2008;16:804811.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40.

    Morak M, Koehler U, Schackert HK, et al. Biallelic MLH1 SNP cDNA expression or constitutional promoter methylation can hide genomic rearrangements causing Lynch syndrome. J Med Genet 2011;48:513519.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 41.

    Gylling A, Ridanpää M, Vierimaa O, et al. Large genomic rearrangements and germline epimutations in Lynch syndrome. Int J Cancer 2009;124:23332340.

  • 42.

    Ward RL, Dobbins T, Lindor NM, et al. Identification of constitutional MLH1 epimutations and promoter variants in colorectal cancer patients from the Colon Cancer Family Registry. Genet Med 2013;15:2535.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43.

    Cini G, Carnevali I, Quaia M, et al. Concomitant mutation and epimutation of the MLH1 gene in a Lynch syndrome family. Carcinogenesis 2015;36:452458.

  • 44.

    Morak M, Ibisler A, Keller G, et al. Comprehensive analysis of the MLH1 promoter region in 480 patients with colorectal cancer and 1150 controls reveals new variants including one with a heritable constitutional MLH1 epimutation. J Med Genet 2018;55:240248.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 45.

    Crépin M, Dieu MC, Lejeune S, et al. Evidence of constitutional MLH1 epimutation associated to transgenerational inheritance of cancer susceptibility. Hum Mutat 2012;33:180188.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 46.

    Sloane MA, Nunez AC, Packham D, et al. Mosaic epigenetic inheritance as a cause of early-onset colorectal cancer. JAMA Oncol 2015;1:953957.

  • 47.

    Castillejo A, Hernández-Illán E, Rodriguez-Soler M, et al. Prevalence of MLH1 constitutional epimutations as a cause of Lynch syndrome in unselected versus selected consecutive series of patients with colorectal cancer. J Med Genet 2015;52:498502.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 48.

    Niessen RC, Hofstra RM, Westers H, et al. Germline hypermethylation of MLH1 and EPCAM deletions are a frequent cause of Lynch syndrome. Genes Chromosomes Cancer 2009;48:737744.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 49.

    Pineda M, Mur P, Iniesta MD, et al. MLH1 methylation screening is effective in identifying epimutation carriers. Eur J Hum Genet 2012;20:12561264.

  • 50.

    Crucianelli F, Tricarico R, Turchetti D, et al. MLH1 constitutional and somatic methylation in patients with MLH1 negative tumors fulfilling the revised Bethesda criteria. Epigenetics 2014;9:14311438.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 51.

    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;96:261268.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 52.

    van Roon EH, van Puijenbroek M, Middeldorp A, et al. Early onset MSI-H colon cancer with MLH1 promoter methylation, is there a genetic predisposition? BMC Cancer 2010;10:180.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 53.

    Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med 2005;352:18511860.

  • 54.

    Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for Lynch syndrome among patients with colorectal cancer. J Clin Oncol 2008;26:57835788.

  • 55.

    Pearlman R, Frankel WL, Swanson BJ, et al. Prospective statewide study of universal screening for hereditary colorectal cancer: the Ohio Colorectal Cancer Prevention Initiative. JCO Precis Oncol 2021;5:PO.20.00525.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 56.

    Deng G, Chen A, Hong J, et al. Methylation of CpG in a small region of the hMLH1 promoter invariably correlates with the absence of gene expression. Cancer Res 1999;59:20292033.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 57.

    Deng G, Peng E, Gum J, et al. Methylation of hMLH1 promoter correlates with the gene silencing with a region-specific manner in colorectal cancer. Br J Cancer 2002;86:574579.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 58.

    Newton K, Jorgensen NM, Wallace AJ, et al. Tumour MLH1 promoter region methylation testing is an effective prescreen for Lynch syndrome (HNPCC). J Med Genet 2014;51:789796.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 59.

    Dámaso E, Castillejo A, Arias MDM, et al. Primary constitutional MLH1 epimutations: a focal epigenetic event. Br J Cancer 2018;119:978987.

  • 60.

    Hitchins MP, Ward RL. Erasure of MLH1 methylation in spermatozoa- implications for epigenetic inheritance. Nat Genet 2007;39:1289.

  • 61.

    Wong JJ, Hawkins NJ, Ward RL, et al. Methylation of the 3p22 region encompassing MLH1 is representative of the CpG island methylator phenotype in colorectal cancer. Mod Pathol 2011;24:396411.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 62.

    Dámaso E, Canet-Hermida J, Vargas-Parra G, et al. Highly sensitive MLH1 methylation analysis in blood identifies a cancer patient with low-level mosaic MLH1 epimutation. Clin Epigenetics 2019;11:171.

    • PubMed
    • Search Google Scholar
    • Export Citation

Submitted July 22, 2022; final revision received March 3, 2023; accepted for publication March 9, 2023.

Author contributions: Study design: Hitchins, Hampel. Data generation and analyses: Hitchins, Dámaso, Alvarez, Zhou, Hu, Pearlman, Hampel. Statistical analyses: Hitchins, Diniz. Results interpretation: Hitchins, Diniz, Pineda, Capella, Pearlman, Hampel. Manuscript preparation and review: All authors.

Disclosures: H. Hampel has disclosed serving on an advisory board for Invitae Genetics, Genome Medical, Natera, and Promega; serving as a consultant for 23andMe, GI OnDemand, and AIM Specialty Health; and owning stock or having an ownership interest in Genome Medical and GI OnDemand. The remaining authors have disclosed that they have not received any financial considerations from any person or organization to support the preparation, analysis, results, or discussion of this article.

Funding: Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under award numbers CA218342 (M.P. Hitchins), as well as CA67941 and CA16058 (H. Hampel); the Cedars-Sinai Medical Center Precision Health Initiative (M.P. Hitchins); Ministerio de Economía y Competitividad (EEBB-I-16-11581, E. Dámaso); and the Spanish Ministry of Science (PID2019-111254RB-I00, M. Pineda; SAF2015-68016-R, G. Capella).

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Correspondence: Megan P. Hitchins, PhD, Cedars-Sinai Medical Center, 8687 Melrose Avenue, Suite 230.84 B (East), West Hollywood, CA 90069. Email: megan.hitchins@cshs.org

Supplementary Materials

  • Collapse
  • Expand
  • Figure 1.

    Blood-based detection of constitutional MLH1 methylation by CpG pyrosequencing in CRC cases with MMRd, MLH1-methylated tumors. Pyrogram traces are shown for patients with CRC in whom constitutional MLH1 methylation was detected across 5 CpG sites within the c.-241 to c.-272 sequence (GenBank accession no. NG_007109.2) of the MLH1 promoter in bisulfite-converted germline DNA. (A) MLH1 methylation was detected in PBL DNA of 4 cases from the Columbus cohort. (B) MLH1 methylation was detected in whole-blood DNA from 4 cases from the OCCPI cohort. (C) Illustrative examples of pyrograms from the hypermethylated RKO CRC cell line as positive control and an unmethylated healthy control PBL as a negative control. Methylation is detected by the presence of a peak at the cytosine (C) within each CpG site interrogated (blue bars), whereas unmethylated cytosines are detected as thymine (T) peaks within the same CpG sites, due to the conversion of unmethylated cytosines to uracils using bisulfite treatment. The percentage methylation value at each CpG site appears above each bar. The mean percentage of methylation across all 5 CpG sites is calculated and shown above each pyrogram. The yellow bar indicates a non-CpG cytosine used as a quality control measure to ensure complete bisulfite conversion to T, whereupon this yields a valid test result.

    Abbreviations: Columbus, Columbus-area HNPCC study; CRC, colorectal cancer; MMRd, mismatch repair–deficient; OCCPI, Ohio Colorectal Cancer Prevention Initiative; PBL, peripheral blood leukocyte.

  • Figure 2.

    Identification of LoH of the unmethylated allele in the tumor of patient OCCPI-1 indicates low-level mosaic constitutional MLH1 methylation predisposed to CRC development. (A) Pyrosequencing in NCM shows low-level methylation at 2.6% in the tissue of tumor origin and confirms high-level methylation in the tumor (60.8%). Legend according to Figure 1. (B) Clonal bisulfite sequencing in the tumor. Legend according to supplemental eFigure 6. Although patient OCCPI-1 was heterozygous at the c.-93G>A SNP (rs1800734), only detected “G” alleles were sequenced in the tumor, most of which were methylated. The methylation patterns on individual alleles were repetitive, suggesting clonal expansion from a small number of cells that contained methylated G alleles. (C) Sanger sequencing electropherogram across the c.-93G>A SNP site within the MLH1 promoter is shown in the tumor and paired NCM samples, with the RFU measured at the A and G peaks at the c.-93G>A SNP (indicated by arrows), shown for each allele below. The LoH index in the tumor was calculated with reference to the paired NCM sample as (A/GTumor)/(A/GNCM), whereby values <0.6 or >1.7 indicate LoH of the A or G allele, respectively. NCM has approximately equal representation of the G and A alleles, consistent with heterozygosity. The tumor shows reduced representation of the A allele (LoH index 0.23), consistent with LoH of the A allele as the “second hit” in the tumor. See supplemental eFigure 6, available with this article at JNCCN.org.

    Abbreviations: CRC, colorectal cancer; LoH, loss of heterozygosity; NCM, normal colorectal mucosa; OCCPI, Ohio Colorectal Cancer Prevention Initiative; RFU, relative fluorescence units; SNP, single-nucleotide polymorphism.

  • Figure 3.

    Rate of detection of constitutional MLH1 methylation by age among the Columbus and OCCPI cohorts. (A) Histograms showing the number of cases with constitutional MLH1 methylation detected (orange) among CRC cases with an MMRd, MLH1-methylated tumor (blue) tested within 5-year age bins beginning at age <50 years for each cohort. (B) Line graphs showing the cumulative percentage of CRC cases tested by age (blue), beginning at age <50 years pooled, with 100% of cases being n=95 for the Columbus cohort and n=281 for the OCCPI cohort. The percentage of cases with constitutional MLH1 methylation detected among cases tested of the same age, beginning at age <50 years, are shown in orange. In the Columbus cohort the highest detection rate was 75% (3 of 4) at age threshold <55 years, with 1 older case missed. There were no cases aged 55 years at diagnosis in the Columbus cohort; therefore, the same detection rate was observed at ≤55 years. In the OCCPI cohort, the highest detection rate was 25% (2 of 8) cases <50 years, but with 2 cases missed. The youngest age threshold at which all cases were detected was age ≤55 years (<56 years), yielding a detection rate of 23.5% (4 of 17 cases). In both cohorts, screening at age ≤55 years would entail screening of a small fraction of total cases: 4% for the Columbus cohort and 6% of the OCCPI cohort.

    Abbreviations: Columbus, Columbus-area HNPCC study; CRC, colorectal cancer; MMRd, mismatch repair–deficient; OCCPI, Ohio Colorectal Cancer Prevention Initiative.

  • 1.

    Parsons R, Li GM, Longley MJ, et al. Hypermutability and mismatch repair deficiency in RER+ tumor cells. Cell 1993;75:12271236.

  • 2.

    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;58:52485257.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Yamamoto H, Imai K. An updated review of microsatellite instability in the era of next-generation sequencing and precision medicine. Semin Oncol 2019;46:261270.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Cunningham JM, Christensen ER, Tester DJ, et al. Hypermethylation of the hMLH1 promoter in colon cancer with microsatellite instability. Cancer Res 1998;58:34553460.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Herman JG, Umar A, Polyak K, et al. Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA 1998;95:68706875.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.

    Ward R, Meagher A, Tomlinson I, et al. Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut 2001;48:821829.

  • 7.

    Peltomäki P. Update on Lynch syndrome genomics. Fam Cancer 2016;15:385393.

  • 8.

    Hitchins M, Williams R, Cheong K, et al. MLH1 germline epimutations as a factor in hereditary nonpolyposis colorectal cancer. Gastroenterology 2005;129:13921399.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Hitchins MP. The role of epigenetics in Lynch syndrome. Fam Cancer 2013;12:189205.

  • 10.

    Lynch HT, Snyder CL, Shaw TG, et al. Milestones of Lynch syndrome: 1895–2015. Nat Rev Cancer 2015;15:181194.

  • 11.

    Burt RW, Cannon JA, David DS, et al. Colorectal cancer screening. J Natl Compr Canc Netw 2013;11:15381575.

  • 12.

    Beamer LC, Grant ML, Espenschied CR, et al. Reflex immunohistochemistry and microsatellite instability testing of colorectal tumors for Lynch syndrome among US cancer programs and follow-up of abnormal results. J Clin Oncol 2012;30:10581063.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Heald B, Plesec T, Liu X, et al. Implementation of universal microsatellite instability and immunohistochemistry screening for diagnosing Lynch syndrome in a large academic medical center. J Clin Oncol 2013;31:13361340.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Schofield L, Grieu F, Amanuel B, et al. Population-based screening for Lynch syndrome in Western Australia. Int J Cancer 2014;135:10851091.

  • 15.

    Ness RM, Llor X, Abbadessa B, et al. NCCN Clinical Practice Guidelines in Oncology: Colon Cancer. Version 3.2022. Accessed November 7, 2022. To view the most recent version, visit https://www.nccn.org

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16.

    Gupta S, Weiss JM, Axell L, et al. NCCN Clinical Practice Guidelines in Oncology: Genetic/Familial High-Risk Assessment: Colorectal. Version 2.2022. Accessed July 11, 2022. To view the most recent version, visit https://www.nccn.org

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med 2015;372:25092520.

  • 18.

    Webber EM, Kauffman TL, O’Connor E, et al. Systematic review of the predictive effect of MSI status in colorectal cancer patients undergoing 5FU-based chemotherapy. BMC Cancer 2015;15:156.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.

    Pérez-Carbonell L, Alenda C, Payá A, et al. Methylation analysis of MLH1 improves the selection of patients for genetic testing in Lynch syndrome. J Mol Diagn 2010;12:498504.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Nakagawa H, Nagasaka T, Cullings HM, et al. Efficient molecular screening of Lynch syndrome by specific 3′ promoter methylation of the MLH1 or BRAF mutation in colorectal cancer with high-frequency microsatellite instability. Oncol Rep 2009;21:15771583.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Gausachs M, Mur P, Corral J, et al. MLH1 promoter hypermethylation in the analytical algorithm of Lynch syndrome: a cost-effectiveness study. Eur J Hum Genet 2012;20:762768.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    McGivern A, Wynter CV, Whitehall VL, et al. Promoter hypermethylation frequency and BRAF mutations distinguish hereditary non-polyposis colon cancer from sporadic MSI-H colon cancer. Fam Cancer 2004;3:101107.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Loughrey MB, Waring PM, Tan A, et al. Incorporation of somatic BRAF mutation testing into an algorithm for the investigation of hereditary non-polyposis colorectal cancer. Fam Cancer 2007;6:301310.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    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;41:664668.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25.

    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;10:191195.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26.

    Hampel H, Pearlman R, Beightol M, et al. Assessment of tumor sequencing as a replacement for Lynch syndrome screening and current molecular tests for patients with colorectal cancer. JAMA Oncol 2018;4:806813.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27.

    Gazzoli I, Loda M, Garber J, et al. A hereditary nonpolyposis colorectal carcinoma case associated with hypermethylation of the MLH1 gene in normal tissue and loss of heterozygosity of the unmethylated allele in the resulting microsatellite instability-high tumor. Cancer Res 2002;62:39253928.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Miyakura Y, Sugano K, Akasu T, et al. Extensive but hemiallelic methylation of the hMLH1 promoter region in early-onset sporadic colon cancers with microsatellite instability. Clin Gastroenterol Hepatol 2004;2:147156.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Suter CM, Martin DI, Ward RL. Germline epimutation of MLH1 in individuals with multiple cancers. Nat Genet 2004;36:497501.

  • 30.

    Valle L, Carbonell P, Fernandez V, et al. MLH1 germline epimutations in selected patients with early-onset non-polyposis colorectal cancer. Clin Genet 2007;71:232237.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Goel A, Nguyen TP, Leung HC, et al. De novo constitutional MLH1 epimutations confer early-onset colorectal cancer in two new sporadic Lynch syndrome cases, with derivation of the epimutation on the paternal allele in one. Int J Cancer 2011;128:869878.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32.

    Auclair J, Vaissière T, Desseigne F, et al. Intensity-dependent constitutional MLH1 promoter methylation leads to early onset of colorectal cancer by affecting both alleles. Genes Chromosomes Cancer 2011;50:178185.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33.

    Pineda M, Mur P, Iniesta MD, et al. MLH1 methylation screening is effective in identifying epimutation carriers. Eur J Hum Genet 2012;20:12561264.

  • 34.

    Hitchins MP, Rapkins RW, Kwok CT, et al. Dominantly inherited constitutional epigenetic silencing of MLH1 in a cancer-affected family is linked to a single nucleotide variant within the 5'UTR. Cancer Cell 2011;20:200213.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35.

    Hitchins MP, Wong JJ, Suthers G, et al. Inheritance of a cancer-associated MLH1 germ-line epimutation. N Engl J Med 2007;356:697705.

  • 36.

    Hitchins MP, Owens SE, Kwok CT, et al. Identification of new cases of early-onset colorectal cancer with an MLH1 epimutation in an ethnically diverse South African cohort. Clin Genet 2011;80:428434.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37.

    Kwok CT, Vogelaar IP, van Zelst-Stams WA, et al. The MLH1 c.-27C>A and c.85G>T variants are linked to dominantly inherited MLH1 epimutation and are borne on a European ancestral haplotype. Eur J Hum Genet 2014;22:617624.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38.

    Leclerc J, Flament C, Lovecchio T, et al. Diversity of genetic events associated with MLH1 promoter methylation in Lynch syndrome families with heritable constitutional epimutation. Genet Med 2018;20:15891599.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39.

    Morak M, Schackert HK, Rahner N, et al. Further evidence for heritability of an epimutation in one of 12 cases with MLH1 promoter methylation in blood cells clinically displaying HNPCC. Eur J Hum Genet 2008;16:804811.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40.

    Morak M, Koehler U, Schackert HK, et al. Biallelic MLH1 SNP cDNA expression or constitutional promoter methylation can hide genomic rearrangements causing Lynch syndrome. J Med Genet 2011;48:513519.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 41.

    Gylling A, Ridanpää M, Vierimaa O, et al. Large genomic rearrangements and germline epimutations in Lynch syndrome. Int J Cancer 2009;124:23332340.

  • 42.

    Ward RL, Dobbins T, Lindor NM, et al. Identification of constitutional MLH1 epimutations and promoter variants in colorectal cancer patients from the Colon Cancer Family Registry. Genet Med 2013;15:2535.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43.

    Cini G, Carnevali I, Quaia M, et al. Concomitant mutation and epimutation of the MLH1 gene in a Lynch syndrome family. Carcinogenesis 2015;36:452458.

  • 44.

    Morak M, Ibisler A, Keller G, et al. Comprehensive analysis of the MLH1 promoter region in 480 patients with colorectal cancer and 1150 controls reveals new variants including one with a heritable constitutional MLH1 epimutation. J Med Genet 2018;55:240248.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 45.

    Crépin M, Dieu MC, Lejeune S, et al. Evidence of constitutional MLH1 epimutation associated to transgenerational inheritance of cancer susceptibility. Hum Mutat 2012;33:180188.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 46.

    Sloane MA, Nunez AC, Packham D, et al. Mosaic epigenetic inheritance as a cause of early-onset colorectal cancer. JAMA Oncol 2015;1:953957.

  • 47.

    Castillejo A, Hernández-Illán E, Rodriguez-Soler M, et al. Prevalence of MLH1 constitutional epimutations as a cause of Lynch syndrome in unselected versus selected consecutive series of patients with colorectal cancer. J Med Genet 2015;52:498502.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 48.

    Niessen RC, Hofstra RM, Westers H, et al. Germline hypermethylation of MLH1 and EPCAM deletions are a frequent cause of Lynch syndrome. Genes Chromosomes Cancer 2009;48:737744.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 49.

    Pineda M, Mur P, Iniesta MD, et al. MLH1 methylation screening is effective in identifying epimutation carriers. Eur J Hum Genet 2012;20:12561264.

  • 50.

    Crucianelli F, Tricarico R, Turchetti D, et al. MLH1 constitutional and somatic methylation in patients with MLH1 negative tumors fulfilling the revised Bethesda criteria. Epigenetics 2014;9:14311438.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 51.

    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;96:261268.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 52.

    van Roon EH, van Puijenbroek M, Middeldorp A, et al. Early onset MSI-H colon cancer with MLH1 promoter methylation, is there a genetic predisposition? BMC Cancer 2010;10:180.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 53.

    Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med 2005;352:18511860.

  • 54.

    Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for Lynch syndrome among patients with colorectal cancer. J Clin Oncol 2008;26:57835788.

  • 55.

    Pearlman R, Frankel WL, Swanson BJ, et al. Prospective statewide study of universal screening for hereditary colorectal cancer: the Ohio Colorectal Cancer Prevention Initiative. JCO Precis Oncol 2021;5:PO.20.00525.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 56.

    Deng G, Chen A, Hong J, et al. Methylation of CpG in a small region of the hMLH1 promoter invariably correlates with the absence of gene expression. Cancer Res 1999;59:20292033.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 57.

    Deng G, Peng E, Gum J, et al. Methylation of hMLH1 promoter correlates with the gene silencing with a region-specific manner in colorectal cancer. Br J Cancer 2002;86:574579.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 58.

    Newton K, Jorgensen NM, Wallace AJ, et al. Tumour MLH1 promoter region methylation testing is an effective prescreen for Lynch syndrome (HNPCC). J Med Genet 2014;51:789796.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 59.

    Dámaso E, Castillejo A, Arias MDM, et al. Primary constitutional MLH1 epimutations: a focal epigenetic event. Br J Cancer 2018;119:978987.

  • 60.

    Hitchins MP, Ward RL. Erasure of MLH1 methylation in spermatozoa- implications for epigenetic inheritance. Nat Genet 2007;39:1289.

  • 61.

    Wong JJ, Hawkins NJ, Ward RL, et al. Methylation of the 3p22 region encompassing MLH1 is representative of the CpG island methylator phenotype in colorectal cancer. Mod Pathol 2011;24:396411.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 62.

    Dámaso E, Canet-Hermida J, Vargas-Parra G, et al. Highly sensitive MLH1 methylation analysis in blood identifies a cancer patient with low-level mosaic MLH1 epimutation. Clin Epigenetics 2019;11:171.

    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 5795 2127 444
PDF Downloads 3846 1238 67
EPUB Downloads 0 0 0