MLH1 c.676C>T, p.(Arg226Ter)

Reasoning

Variant NM_000249.3:c.676C>T is a nonsense change creating a premature termination codon (PTC) at codon 226 (NP_000240.1:p.(Arg226Ter)) in MLH1. The governing framework is the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MLH1 v2.0 (CSPEC, precedence rank 1), whose PVS1 rule states: 'Nonsense/frameshift variant introducing Premature Termination Codon (PTC) at or before codon 753 in MLH1' = PVS1 Very Strong. Codon 226 is well before the codon 753 boundary, and germline loss of function is an established MLH1 disease mechanism (pvs1_gene_context: gene gate eligible). No downgrade considerations apply: the PTC is in exon 8 of 19 (not the last exon, NMD expected), it is not in a distal non-critical region, and the variant is ultra-rare in population data (gnomAD v4.1 AF = 6.2e-07, 1/1,605,126 alleles), so no population LoF-enrichment concern. Huang et al. (PMID:11343035) directly document the exact variant in a virulent HNPCC kindred: germline C>T at base pair 676, codon 226, in exon 8 of hMLH1, with in vitro transcription-translation (IVTT) of proband mRNA showing a truncated hMLH1 protein in the exons 1-12 segment, 'suggesting that the stop codon, and not altered splicing, is the predominant mode of hMLH1 inactivation in this family'. Note: the variant is 2 bp from the 3' end of exon 8 (donor splice-site proximity), but SpliceAI predicts no significant splicing impact (max delta 0.09) and the IVTT data support the PTC as the mechanism; PVS1 is therefore applied via the nonsense/PTC rule. Per the VCEP rule, PVS1 is not to be combined with PP3 (combination guidance for other groups).

Rule

Nonsense/frameshift variant introducing Premature Termination Codon (PTC) at or before codon 753 in MLH1 (InSiGHT MLH1 VCEP v2.0, PVS1 Very Strong).

Evidence considered

• cspec
• pvs1_gene_context
• pvs1_variant_assessment
• PMID:11343035
• PMID:10874307
• spliceai
• gnomad_v4

Sources

Reasoning

InSiGHT MLH1 v2.0 PM2 rule: absent/extremely rare allele frequency < 0.00002 (< 1 in 50,000 alleles) in the gnomAD v4 dataset, strength Supporting. gnomAD v4.1 total AF is 6.23e-07 (1/1,605,126 alleles; exome AF 6.88e-07, 1/1,452,898) — more than 30-fold below the 0.00002 threshold — with 0 homozygotes and no population exceeding 8.53e-07 (NFE). PM2_Supporting is met. No threshold artifact: the single observed allele is in the European (non-Finnish) population in both v2.1 and v4.1, and gnomAD-Canada v1.0 is concordantly absent. Note for the classifier: this is a truncating variant (p.Arg226Ter) and PM2 is scored as written; whether the framework elects to combine PM2 with PVS1 is a combination-policy decision outside this group's scope.

Rule

Absent/extremely rare allele frequency <0.00002 (<1 in 50,000 alleles) in gnomAD v4 dataset

Evidence considered

• gnomAD v4.1 total AF 6.23e-07 < 0.00002 threshold; 0 homozygotes; max subpopulation AF 8.53e-07 (NFE)
• gnomAD v2.1 total AF 3.98e-06 corroborates extreme rarity
• gnomAD-Canada v1.0: absent
• InSiGHT MLH1 v2.0 CSPEC PM2 rule text (supporting strength)

Sources

Reasoning

Governing framework: ClinGen InSiGHT HCRC/Polyposis VCEP MLH1 v2.0 (cspec), which makes PM3 applicable with a point system: P/LP variant in trans = 1.0 point per case, P/LP phase unknown = 0.5 points; >=4 pts = Very Strong, 2-3.5 pts = Strong, 1-1.5 pts = Moderate, 0.5 pts = Supporting; requires co-occurrence with a known P/LP sequence variant in the same gene in a patient with clinical features consistent with CMMRD per Aronson et al 2022, and for MLH1 variants PM2_supporting must be met. PM2_supporting gate is met: gnomAD v4.1 AF = 6.23e-07 (<0.00002). ClinVar variation 17087 submission descriptions (SCV004840892 All of Us; SCV002580789 MGZ) state the variant 'has also been reported in homozygous carriers affected with constitutional mismatch repair deficiency syndrome (pmid: 17889038)'. Homozygous CMMRD-affected carriers are biallelic observations: the variant is present on both alleles, i.e. in trans with an identical allele that is itself a known Pathogenic variant (InSiGHT), and affected status satisfies the CMMRD-consistent-features requirement; phase is definitively known from homozygosity. This supports >=1.0 point -> PM3 Moderate (floor). Strength could be Strong if >=2 independent homozygous cases exist (the ClinVar text says 'carriers', plural, but the count is not quantified in any available source). Confidence is low and human review is requested because the primary source (PMID 17889038) is not present in the source registry or the downloaded publications, the extracted ClinVar sentences are truncated auto-extracted fragments, and no phase-confirmed trans observation with a distinct second MLH1 P/LP variant is documented in the bundle. The downloaded papers that mention c.676C>T (PMID:10874307, PMID:16216036, PMID:11343035) were scanned and contain only heterozygous Lynch-syndrome observations with no biallelic/CMMRD/phase data, so they were not used for PM3.

Rule

InSiGHT MLH1 VCEP v2.0 PM3: co-occurrence with a known P/LP sequence variant in the same gene in a patient with clinical features consistent with CMMRD (Aronson et al 2022); for MLH1 the variant must meet PM2_supporting. Points: P/LP in trans = 1.0, phase unknown = 0.5 per case; >=4 pts = Very Strong, 2-3.5 pts = Strong, 1-1.5 pts = Moderate, 0.5 pts = Supporting.

Evidence considered

• ClinVar variation 17087 submission descriptions (SCV004840892 All of Us; SCV002580789 MGZ) documenting homozygous carriers affected with CMMRD
• gnomAD v4.1 AF 6.23004e-07 (1/1605126 alleles, 0 homozygotes) confirming the VCEP PM2_supporting gate (<0.00002) required before applying PM3 to MLH1 variants
• InSiGHT MLH1 VCEP v2.0 PM3 instructionsToUse point rules (P/LP in trans = 1.0 point; phase unknown = 0.5 points; strength thresholds)

Still missing

• Number of independent homozygous CMMRD cases is not quantified in the available sources; if >=2 independent cases exist, PM3 strength rises to Strong (2-3.5 points) or Very Strong (>=4 points)
• Primary source PMID 17889038 (homozygous CMMRD carriers) is not in source_registry or the publications directory and could not be verified directly
• No phase-confirmed trans observation with a distinct second MLH1 P/LP variant is documented; only homozygous (same-variant in trans) observations are traceable
• ClinVar evidence signals are truncated auto-extracted fragments of submission descriptions; the full submission text is not stored in the case bundle

Sources

Reasoning

InSiGHT MLH1 VCEP v2.0 PP4 (phenotype specificity via tumor MMR deficiency) is met at Moderate strength: 2 independent MSI-H colorectal tumors with loss of MLH1 protein expression consistent with the variant location in germline carriers of the exact variant. Tumor 1: PMID:11343035 - a sigmoid adenocarcinoma from the 14-year-old proband of an Amsterdam-criteria HNPCC kindred showed high-frequency MSI using the standard NCI 5-locus panel (BAT-25, BAT-26, D2S123, D5S346, D17S250; >=40% loci unstable = MSI-H) and loss of hMLH1 expression, explicitly interpreted as bi-allelic inactivation of hMLH1. Tumor 2: PMID:16216036 - Table 1, family 490 (patient 1), MLH1 c.676C>T (R226X, truncating) in an MSI-H colorectal cancer with IHC showing loss of nuclear MLH1 expression (IHC for MLH1 '-') and normal MSH2 expression (IHC for MSH2 '+'), from an independent carrier in the Bonn cohort. Both tumors are MSI-H with MMR-protein loss consistent with the variant gene, satisfying the rule's 'MSI-H AND/OR loss of MMR protein expression consistent with the variant location' clause; independent tumors may be from the same patient/family, and these are from two different families. Does not reach PP4_Strong (requires >=3 tumors).

Rule

InSiGHT MLH1 VCEP v2.0 PP4: Moderate = 2 independent CRC/Endometrial MSI-H tumors using a standard panel of 5-10 markers and/or loss of MMR protein expression consistent with the variant location; for MLH1 variants, MLH1 promoter methylation is to be excluded in the tumors

Evidence considered

• PMID:11343035
• PMID:16216036

Still missing

• MLH1 promoter methylation exclusion required by the InSiGHT PP4 rule is not explicitly documented in either tumor source (PMID:11343035 and PMID:16216036 both predate routine MLH1 promoter methylation testing); however, PMID:11343035 directly demonstrates bi-allelic inactivation (germline mutation + loss of MLH1 expression in the tumor). A reviewer may wish to confirm methylation was excluded before finalizing Moderate vs Supporting strength.

Sources

Reasoning

Expert panel International Society for Gastrointestinal Hereditary Tumours (InSiGHT) classified as Pathogenic.

Rule

Global PP5/BP6 policy: exact-variant ClinVar expert-panel classification Pathogenic/Likely pathogenic (3-star) -> PP5 at Supporting strength, applied regardless of VCEP 'Not Applicable' flag

Evidence considered

• clinvar
• ClinVar expert panel classification

Sources

Reasoning

Governing framework is the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis EP specification for MLH1 v2.0 (cspec, framework_mode=vcep), which scores PS2 as de novo points (2 points/proband: confirmed de novo with both maternity and paternity confirmed AND MMR-deficient LS-spectrum tumor with no MLH1 methylation; 1 point/proband: confirmed de novo with LS-spectrum tumor but no tumor MSI/IHC/methylation data; 0.5 points/proband: assumed de novo with maternity/paternity unconfirmed). No de novo occurrence is documented for c.676C>T in any available source. On the contrary, the only family study with parental testing (PMID:11343035) shows the variant was inherited: it is present in the proband (14-yr-old with sigmoid adenocarcinoma, MSI-H, loss of MLH1 expression) AND in her affected mother, definitively excluding a de novo origin. ClinVar submissions (VCV000017087, expert-panel Pathogenic) contain no de novo/parental-testing annotations. PS2 therefore cannot be met.

Rule

InSiGHT MLH1 VCEP v2.0 PS2: >=4 de novo points = Very Strong; 2-3.5 points = Strong; 1-1.5 points = Moderate; 0.5 points = Supporting. Points: 2/proband if de novo confirmed (both parents) with MMR-deficient LS-spectrum tumor; 1/proband if de novo confirmed with LS-spectrum tumor without tumor data; 0.5/proband if assumed de novo with parents unconfirmed.

Evidence considered

• Sequencing of proband, mother, and 8 family members shows the c.676C>T variant in the proband AND her mother (maternal transmission), excluding de novo origin (PMID:11343035).
• ClinVar VCV000017087 submissions provide no de novo or parental-testing annotations for this variant.

Still missing

• No de novo occurrence of c.676C>T with parental confirmation exists in the case evidence; a documented de novo proband (with both parents tested) would be required to revisit PS2.

Sources

Reasoning

Variant is nonsense p.(Arg226Ter) (PTC at codon 226, exon 8). Governing framework is the ClinGen InSiGHT MLH1 v2.0 specification (cspec, doc 1564688410), which requires calibrated functional assays with functional odds for pathogenicity (Strong >18.7; Moderate >4.3 and <=18.7 or MMR function defect per functional assay flowchart or monoallelic complete loss of expression <10% of wild-type in cDNA without puromycin, full-length transcript with and without NMD block; Supporting >2.08 and <=4.3). No VCEP-approved calibrated assay (CIMRA/Drost, chemical-selection/Jia-Scott, hESC-CRISPR/Rath, Thompson CIM-type repair assays, Morak biallelic allelic-expression, Bouvet survival assay per Functional-assay-SVI-documentation-MMR.xlsx) was performed for this variant, and no functional odds are available. The only variant-specific functional data are qualitative: IVTT from proband mRNA demonstrating a truncated hMLH1 polypeptide in segment 1 (exons 1-12) confirming the stop codon is used rather than altered splicing (PMID:11343035), and tumor IHC showing complete loss of MLH1 expression (PMID:11343035; PMID:16216036 Table 1 family 501). IVTT and IHC are not on the VCEP approved-assay list and cannot be converted to functional odds; the PS3_Moderate monoallelic-expression arm was not quantitatively assessed (no cDNA with/without NMD block measurement of variant-allele expression). PS3 is therefore not met at any strength. The truncating mechanism is captured by PVS1 (nonsense at or before codon 753 per InSiGHT) and must not be double-counted via PS3.

Rule

Strong: Calibrated functional assays with functional odds for pathogenicity > 18.7. Moderate: Calibrated functional assays with functional odds for pathogenicity >4.3 and <=18.7 OR MMR function defect following functional assay flowchart OR variants with monoallelic expression: complete loss of expression (<10% of wild-type in cDNA without puromycin) of the variant allele (full-length transcript with and without NMD block). Supporting: Calibrated functional odds for pathogenicity >2.08 and <=4.3.

Evidence considered

• PMID:11343035: IVTT shows truncated hMLH1 polypeptide (PTC used, not altered splicing); tumor IHC shows complete loss of MLH1 expression — qualitative, not an approved calibrated assay
• PMID:16216036: Table 1, family 501 — c.676C>T (R226X) truncating, IHC MSH2 retained / MLH1 lost in MSI-H tumor — expression-level context only
• PMID:10874307: c.676C>T listed only as a known-mutation control for the EMD DNA-detection assay; no functional data
• vcep_functional_assay_svi_documentation_mmr: approved/calibrated assay list (CIMRA, chemical selection, hESC-CRISPR, Thompson CIM, Morak allelic expression, Bouvet survival) — IVTT and IHC not included
• cspec: InSiGHT MLH1 v2.0 PS3 thresholds require calibrated functional odds or flowchart-defined MMR function/monoallelic-expression evidence

Still missing

• No VCEP-approved calibrated functional assay (with functional odds) performed on c.676C>T
• No quantitative mRNA allelic-expression analysis with and without NMD block (PS3_Moderate monoallelic-expression arm)

Sources

Reasoning

Variant NM_000249.3:c.676C>T is a nonsense (stop-gain) change p.(Arg226Ter) in exon 8 (c.589-677; c.676 is the penultimate base of exon 8, -2 relative to the intron 8 donor). Under the governing ClinGen InSiGHT MLH1 v2.0 specification, PP3 has two paths: (1) missense variant with HCI prior probability >0.68 (Supporting) or >0.81 (Moderate) — not applicable because this is not a missense variant, and the deterministic HCI prior lookup for c.676C>T returned found:false (the LOVD-derived MLH1 HCI prior table covers missense substitutions only); (2) predicted splice defect for non-canonical splicing nucleotides using SpliceAI with delta score >= 0.2 (Walker et al 2023) — evaluated: SpliceAI max delta score = 0.09 (DS_AG 0.0, DS_AL 0.01, DS_DG 0.09, DS_DL 0.01), well below the 0.2 threshold, so no splice defect is predicted. PP3 is therefore not met. BayesDel (0.66) and REVEL (score null/not found) are missense pathogenicity predictors not used by the InSiGHT MLH1 framework and are not applicable to a nonsense variant; they were not used to support PP3.

Rule

InSiGHT MLH1 v2.0 PP3: missense variant with HCI prior probability for pathogenicity >0.68 (PP3_Supporting) or >0.81 (PP3_Moderate) as per hci-priors.hci.utah.edu; OR predicted splice defect for non-canonical splicing nucleotides using SpliceAI with delta score >= 0.2 as per Walker et al 2023

Evidence considered

• SpliceAI max delta score 0.09 (all four sub-scores < 0.2) — fails the VCEP PP3 splice-defect threshold (>= 0.2, Walker et al 2023)
• HCI prior lookup (LOVD_MLH1_priors 2026-05-05): no entry for c.676C>T (found:false) — HCI missense path inapplicable to a nonsense variant
• Variant consequence: nonsense p.(Arg226Ter), exon 8 (c.589-677), penultimate base (-2) — not a missense substitution, outside HCI prior scope

Sources

Reasoning

Governing framework is the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel MLH1 v2.0 specification (CSPEC doc 1564688410), which defines BA1 as gnomAD v4 Grpmax filtering allele frequency >= 0.001 (0.1%) with the variant excluded as a founder pathogenic variant. gnomAD v4.1 total AF is 6.23e-07 (1/1,605,126 alleles; exome AF 6.88e-07, 1/1,452,898) with 0 homozygotes; the highest population frequency is European (non-Finnish) at 8.53e-07 (1/1,171,694). grpmax_faf is null in the fetched record, but the observed AF is >1000-fold below the 0.001 threshold, so the stand-alone benign criterion is not approached under any population or FAF interpretation. gnomAD v2.1 (AF 3.98e-06) and gnomAD-Canada (absent) corroborate extreme rarity.

Rule

GnomAD v4 Grpmax filtering allele frequency ≥ 0.001 (0.1%) and variant is excluded as founder pathogenic variant.

Evidence considered

• gnomAD v4.1 total AF 6.23e-07 (1/1,605,126 alleles), 0 homozygotes, max subpopulation AF 8.53e-07 (NFE)
• gnomAD v2.1 total AF 3.98e-06 (1/250,982 alleles), 0 homozygotes
• gnomAD-Canada v1.0: absent
• InSiGHT MLH1 v2.0 CSPEC BA1 rule text

Still missing

• grpmax_faf (filtering allele frequency) was null in the gnomAD v4.1 fetch; total/population AF used instead — margin to threshold is >1000x, so this does not affect the call.

Sources

Reasoning

InSiGHT MLH1 v2.0 defines BS1 as gnomAD v4 Grpmax FAF >= 0.0001 and < 0.001 (0.01-0.1%) with the variant excluded as a founder pathogenic variant. gnomAD v4.1 total AF is 6.23e-07 (0.00006%), well below the 0.0001 (0.01%) floor; even the highest ancestry-specific observation across datasets (gnomAD v2.1 NFE_SEU subpopulation, AF 8.72e-05, 1/11,466 alleles) remains below 0.0001. 0 homozygotes in both gnomAD v2.1 and v4.1. The variant is far too rare for BS1 under any ancestry stratification.

Rule

GnomAD v4 Grpmax filtering allele frequency ≥ 0.0001 and < 0.001 (0.01-0.1%) and variant is excluded as founder pathogenic variant.

Evidence considered

• gnomAD v4.1 total AF 6.23e-07 (0.00006%), max subpopulation AF 8.53e-07 (NFE)
• gnomAD v2.1 highest sub-subpopulation AF 8.72e-05 (NFE_SEU, 1/11,466) — still below 0.0001
• gnomAD-Canada v1.0: absent
• InSiGHT MLH1 v2.0 CSPEC BS1 rule text

Still missing

• grpmax_faf null in fetched gnomAD v4.1 record; adjudicated on total and per-population AF, which remain >100-fold below the BS1 floor.

Sources

Reasoning

Under InSiGHT MLH1 v2.0, BS2 is a clinical co-occurrence criterion, not a population-frequency criterion: it requires co-occurrence in trans with a known pathogenic sequence variant in the same gene in a patient with colorectal cancer after age 45 (or other LS cancer above the median LS age of onset) with no CMMRD manifestations, with phase confirmed by parental/offspring testing. No such phase-confirmed trans observation for c.676C>T exists anywhere in this case bundle. The generic ACMG BS2 (homozygous observation in healthy adults) is superseded by the VCEP specification; gnomAD v2.1 and v4.1 both report 0 homozygotes, which is consistent with rarity but does not satisfy the VCEP BS2 rule.

Rule

Co-occurrence in trans with a known pathogenic sequence variant in the same gene in a patient with colorectal cancer after age 45 (or other LS cancer above the median age of onset for that cancer in LS), and who has no previous or current evidence of clinical manifestations of CMMRD as per Aronson et al 2022 (Refer to 'Table for CMMRD diagnosis.pdf'). Confirmation of phase requires testing of parents or offspring.

Evidence considered

• InSiGHT MLH1 v2.0 CSPEC BS2 rule text (trans co-occurrence definition)
• gnomAD v4.1 homozygote count 0
• gnomAD v2.1 homozygote count 0

Still missing

• No phase-confirmed trans co-occurrence of c.676C>T with a known pathogenic MLH1 variant in a CRC patient >45y (or other LS cancer above median onset) without CMMRD was identified; such case-level data would be required to meet BS2.

Sources

Reasoning

Governing framework is the ClinGen InSiGHT MLH1 v2.0 specification. BS3_Strong requires calibrated functional assays with functional odds for pathogenicity <=0.05 or synonymous/intronic variants with no mRNA aberration (splicing or allelic imbalance) determined with NMD inhibition; BS3_Supporting requires calibrated functional odds >0.05 and <=0.48 or variant-specific proficient function in protein and mRNA-based lab assays per the MMR functional assay flowchart. None of these conditions are met: no calibrated assay with functional odds was performed; the variant is a nonsense change (not eligible for the synonymous/intronic mRNA-aberration arm); and the available evidence points in the opposite direction — IVTT-confirmed protein truncation (PMID:11343035) and complete loss of MLH1 protein expression in a carrier tumor (PMID:11343035; PMID:16216036 Table 1 family 501) contradict any proficient function. BS3 is therefore not met at any strength.

Rule

Strong: Calibrated functional assays with functional odds for Pathogenicity <= 0.05 OR synonymous substitutions and intronic variants with no associated mRNA aberration (splicing or allelic imbalance) determined with NMD inhibition. Supporting: Calibrated functional assays with functional odds for Pathogenicity >0.05 and <=0.48 OR variant-specific proficient function in protein and mRNA-based lab assays as per MMR functional assay flowchart.

Evidence considered

• PMID:11343035: IVTT truncation of hMLH1 and tumor IHC loss contradict proficient function
• PMID:16216036: complete loss of MLH1 IHC in MSI-H tumor of c.676C>T carrier contradicts proficient function
• vcep_functional_assay_svi_documentation_mmr: BS3 requires calibrated odds <=0.05 (Strong) or >0.05-<=0.48 / flowchart-defined proficient function (Supporting) — none present
• cspec: InSiGHT MLH1 v2.0 BS3 rule text (no applicable arm satisfied)

Still missing

• No calibrated functional assay (functional odds <=0.48) performed on c.676C>T
• No mRNA assay demonstrating normal splicing/allelic balance (relevant only to synonymous/intronic variants, which this variant is not)

Sources

Reasoning

VCEP BS4 (InSiGHT MLH1 v2.0) requires lack of co-segregation with disease: combined Bayes LR <0.05 (Strong) or >0.05 and <=0.48 (Supporting). No lack-of-segregation evidence exists in the available data. In the only pedigree with segregation data (PMID:11343035), both tested affected individuals carry the variant and no affected non-carrier was identified among the tested family members; the only unaffected carrier (younger brother) is a child/adolescent of unreported age, uninformative for non-segregation. No pedigree shows an affected relative without the variant or an unaffected carrier at an age beyond typical LS onset. The available pattern is consistent with co-segregation, not against it; BS4 is not met.

Rule

InSiGHT MLH1 VCEP v2.0 BS4: combined Bayes LR <0.05 = Strong; >0.05 and <=0.48 = Supporting (lack of co-segregation; LR from multiple pedigrees multiplied).

Evidence considered

• No affected non-carrier among 9 tested individuals; no advanced-age unaffected carrier documented (PMID:11343035).
• Available segregation pattern (2 affected carriers, no affected non-carrier) argues against lack of co-segregation (PMID:11343035).

Still missing

• No pedigrees with documented affected non-carriers or unaffected carriers at informative (post-onset) ages are available; these would be required to evaluate BS4.

Sources

Reasoning

No evidence for the benign tumor-phenotype direction of InSiGHT MLH1 VCEP v2.0 BP5 (>=2 CRC/Endometrial tumors with MSS and/or no loss of MMR protein expression and/or LS-spectrum tumors with MMR-protein loss inconsistent with the variant gene; or BRAF V600E/MLH1 promoter methylation with MSI-H). The only tumor-phenotype data available for carriers of this exact variant point in the opposite direction: both documented tumors (PMID:11343035 proband; PMID:16216036 family 490) are MSI-H with loss of MLH1 expression consistent with the variant, and no MSS or MMR-protein-intact tumor in a carrier is reported in any registered source. gnomAD frequencies are negligible (v4 total AF 6.23e-07) and are not BP5 evidence. Criterion therefore not met.

Rule

InSiGHT MLH1 VCEP v2.0 BP5: Supporting = 2 or 3 tumors with MSS and/or no loss of MMR protein expression and/or LS-spectrum tumors with inconsistent MMR-protein loss; Strong = >=4 tumors

Still missing

• No registered source reports MSS tumors, MMR-protein-intact tumors, BRAF V600E, or MLH1 promoter methylation in carriers of c.676C>T, so a benign phenotype signal cannot be positively excluded beyond the reviewed sources.

Sources

Reasoning

BP6 applies only when an exact-variant ClinVar expert-panel classification is Benign/Likely benign. No such assertion exists for c.676C>T: the only expert-panel submission (InSiGHT, SCV000106813) classifies the variant as Pathogenic, and the aggregate ClinVar label is Pathogenic. The benign direction is therefore unsupported. The InSiGHT MLH1 VCEP v2.0 additionally marks BP6 as 'Not Applicable', but the decisive point is the absence of any benign expert-panel assertion for this variant.

Rule

Global PP5/BP6 policy: BP6 only for exact-variant ClinVar expert-panel classification Benign/Likely benign (none exists for this variant); VCEP marks BP6 'Not Applicable'

Reasoning

VCEP PP1 (InSiGHT MLH1 v2.0) requires a computed combined Bayes Likelihood Ratio from co-segregation in pedigree(s) (COOL v3): Supporting >2.08 and <=4.3; Moderate >4.3 and <=18.7; Strong >18.7 in >=2 families. Directionally positive segregation evidence exists in PMID:11343035 (HNPCC kindred meeting Amsterdam criteria): c.676C>T was found in the affected proband (CRC at age 14), her affected mother (CRC in third decade, recurrence at 48), and an asymptomatic younger brother; the mutation was NOT found in the remaining family members tested, and no affected non-carrier was identified among the 9 individuals sequenced (proband + 8 family members). A crude 2-affected-carrier estimate approximates LR ~4 (supporting range), but the Bayes factor cannot be reliably computed from the published data because: (1) the age of the asymptomatic carrier brother (which determines his informativeness) is not reported; (2) the affected maternal aunt (died age 29) and maternal great-aunt (CRC third decade) were deceased and not tested; (3) no second family with usable segregation data is available locally (the German kindred reported by Moslein et al. 1996, referenced via PMID:10874307 and discussed in PMID:11343035, has no segregation details retrievable here). Because the VCEP makes strength assignment conditional on the COOL-computed Bayes LR, and the available data do not support a defensible point estimate, PP1 is not assessed rather than assigned supporting; human segregation analysis with the full pedigree is required.

Rule

InSiGHT MLH1 VCEP v2.0 PP1: combined Bayes LR >18.7 in >=2 families = Strong; >4.3 and <=18.7 = Moderate; >2.08 and <=4.3 = Supporting (LRs from multiple pedigrees multiplied; COOL v3 recommended).

Evidence considered

• 2 affected carriers (14-yr-old proband with stage II sigmoid adenocarcinoma, MSI-H, MLH1-loss; mother with CRC diagnosed in third decade) among 9 sequenced individuals (PMID:11343035).
• Asymptomatic younger brother is a carrier; age at last follow-up not reported (PMID:11343035).
• Mutation 'not found in the remaining family members who were tested' - no affected non-carrier documented among tested individuals (PMID:11343035).
• Variant previously reported in a German HNPCC kindred (Moslein et al. 1996, earliest CRC age 39) - confirms familial recurrence but no segregation data retrievable (PMID:10874307, PMID:11343035 discussion).

Still missing

• COOL v3 Bayes likelihood ratio not computable from bundle data (age of asymptomatic carrier brother unknown; affected maternal aunt/great-aunt untested).
• Segregation data from the second (German) kindred with this variant (Moslein et al. 1996) not available locally for a combined >=2-family LR.
• Full pedigree (Fig 1 of PMID:11343035) needed for penetrance-informed analysis.

Sources