MUTYH c.53C>T, p.(Pro18Leu)

Reasoning

No MUTYH-specific BS1 threshold is populated in the governing InSiGHT specification, so the supplied generic BS1 threshold of allele frequency >= 0.01 (1%) applies, evaluated against the highest credible ancestry-specific frequency rather than the all-comers total. gnomAD v2.1 East Asian AF is 0.01302997 (260/19,954 alleles, 4 homozygotes) with grpmax FAF 0.0132979; gnomAD v2.1 non-cancer East Asian exome frequency is 0.01324538 (255/19,252); gnomAD v4.1 East Asian AF is 0.01133883 (509/44,890 alleles, 7 homozygotes) with grpmax FAF 0.01052478. All three independently exceed 1% in a well-powered ancestry group (hundreds of carriers), whereas the all-comers totals are an order of magnitude lower (0.00105348 in v2.1; 0.00044112 in v4.1), so ancestry is decisive. This ancestry-specific frequency is corroborated by control cohorts in which the exact variant was screened: 3/343 healthy Chinese controls (0.87% haplotype frequency) in PMID:18811933 and 3/129 healthy controls (1.18%) in PMID:16929514. BS1 is met at strong.

Rule

{'source': 'generic_svi_default', 'threshold': 0.01, 'comparison': '>=', 'unit': 'allele frequency', 'population_selection': 'highest observed ancestry frequency'}

Evidence considered

• The InSiGHT MUTYH specification retrieved for this case has an empty rule payload and no BS1 population threshold, so the supplied generic ACMG/ClinGen-SVI default was applied.
• gnomAD v2.1 East Asian AF 0.01302997 (260/19,954 alleles) and grpmax FAF 0.0132979 exceed 0.01; gnomAD v4.1 East Asian AF 0.01133883 (509/44,890 alleles) and grpmax FAF 0.01052478 also exceed 0.01.
• gnomAD v2.1 non-cancer East Asian exomes independently give 0.01324538 (255/19,252 alleles); the non-cancer subsets are consistent with the all-comers data and do not change the verdict.
• The all-comers frequencies are far lower than 1%: 0.00105348 in gnomAD v2.1 (298/282,872 alleles) and 0.00044112 in gnomAD v4.1 (712/1,614,074 alleles), demonstrating that the criterion is met on the ancestry-specific frequency.
• PMID:18811933 genotyped the exact variant in 343 healthy Chinese controls (3 carriers, 0.87% haplotype frequency; 6/138 sporadic colorectal cancer cases) and PMID:16929514 found it in 3/129 same-region healthy controls (1.18% haplotype frequency), providing independent control-cohort support for an East Asian allele frequency near 1%; these were used as corroboration, not as the threshold calculation.
• The supplied generic BS1 threshold is allele frequency >= 0.01 (1%), with per-gene derivation preferred, as attributed to Whiffin et al. 2017 (PMID:28518168).
• PMID:24033266 supplies the recessive-disease caveat that the carrier frequency of a pathogenic allele must be under twice the square root of the disease prevalence; because no MUTYH-associated polyposis prevalence estimate was available, a gene-derived recessive cutoff could not be computed and the generic 1% default was retained.

Still missing

• Per-gene derivation is preferred for BS1 but no MUTYH-specific threshold is published; because MUTYH-associated polyposis is autosomal recessive, its pathogenic alleles can tolerate frequencies above the 1% default that was calibrated for dominant disease, so this strong benign call should be re-derived if an InSiGHT MUTYH threshold is released.

Sources

Reasoning

Scope check performed before any predictor score was read: NM_001128425.2:c.53C>T is a missense substitution (NP_001121897.1:p.(Pro18Leu)), so BP4 is in scope and its evidence comes from the missense protein-impact path (REVEL) only; a missense variant's splice prediction is never BP4 support. The MUTYH InSiGHT Expert Panel specification (doc 1742141534) was retrieved with an empty criteria payload and pre-assigns no BP4 code, so the generic ClinGen SVI REVEL calibration applies (Pejaver et al. 2022, Am J Hum Genet, PMID:36413997): BP4 supporting <= 0.29, moderate <= 0.183, strong <= 0.016. The local REVEL v1.3 lookup returns 0.2, which is <= 0.29 and therefore meets BP4 at supporting strength; it is above the moderate (0.183) and strong (0.016) cutoffs, so no higher strength is claimed. This is not a gray-zone result: the score falls inside the calibrated benign-supporting band rather than between the BP4 and PP3 supporting cutoffs. A ClinVar submission-level criterion lead for variation 41759 independently reports that '2/3 in silico tools predict a benign outcome for this variant', which is directionally concordant, but it is an uncalibrated submitter statement and was not used to set strength. No double-counting: the same REVEL score is the sole computational basis for both PP3 (not met) and BP4 (met), and no SpliceAI or BayesDel value was used.

Rule

Generic fallback, missense path: BP4 (supporting) requires REVEL <= 0.29; met at REVEL 0.2 (Pejaver et al. 2022, PMID:36413997).

Evidence considered

• Scope: missense variant (NP_001121897.1:p.(Pro18Leu)), so BP4 is evaluated on the REVEL missense path only; the SpliceAI splice path is not applicable to this variant class.
• Governing framework: MUTYH ClinGen InSiGHT Expert Panel Specification v1.0 (doc 1742141534) retrieved with an empty criteria payload (framework_complete=false), pre-assigning no BP4 code; generic ACMG/AMP fallback used.
• REVEL = 0.2 (local REVEL v1.3 lookup) satisfies the BP4 supporting threshold REVEL <= 0.29 and does not reach BP4 moderate (<= 0.183) or strong (<= 0.016) (Pejaver et al. 2022, Am J Hum Genet, PMID:36413997). BP4 is met at supporting strength only.
• Corroborating but non-decisive: a ClinVar submission for variation 41759 carries the criterion-level statement '2/3 in silico tools predict a benign outcome for this variant (snpsandgo not captured due to low reliability index)'; this is an uncalibrated submitter statement and was not used to assign strength.
• No double-counting: the same REVEL 0.2 result is the sole basis for both PP3 (not met) and BP4 (met supporting); no splice prediction and no BayesDel score was used for either criterion.

Sources

Reasoning

The case variant is NM_001128425.2:c.53C>T, encoding NP_001121897.1:p.(Pro18Leu). PS1 requires the variant's amino-acid change to be identical to that of a variant already established as pathogenic. p.Pro18Leu can arise only from c.53C>T (codon 18 spans c.52-54; a CCN-to-CTN proline-to-leucine change requires substitution of the second base, c.53C>T), so there is no independent nucleotide change that could supply a same-amino-acid pathogenic comparator. ClinVar carries this variant in two records (VCV000041759 and the companion record) with conflicting classifications, 9 Benign and 3 Likely benign laboratory submissions, 2 Uncertain significance submissions and no expert-panel submission and no pathogenic submission, so the amino-acid change is not established as pathogenic. The three reports of p.Pro18Leu available in this case describe it as a heterozygous haplotype change: PMID:18422726 found it once among 66 Japanese adenomatous polyposis patients, always in cis and compound-heterozygous with p.Gly25Asp, p.Gln324His and c.1389G>C, and explicitly did not perform functional analysis of p.Pro18Leu because proline 18 is not conserved across species; PMID:16929514 found it only in cis with p.Gly25Asp in gastric-cancer cases and controls and interpreted the T/A haplotype as a polymorphism; PMID:18811933 found the isolated p.Pro18Leu protein mislocalised no differently from wild type. PS1 is not met.

Rule

PS1 (ACMG/AMP 2015): same amino-acid change as a previously established pathogenic variant, regardless of nucleotide change. No MUTYH-specific PS1 rule is published in the governing specification, so the generic definition was applied.

Evidence considered

• The case variant is NP_001121897.1:p.(Pro18Leu), i.e. a proline-to-leucine change at residue 18, from NM_001128425.2:c.53C>T.
• ClinVar (variation IDs 41759 / 41763; VCV000041759, last updated 2026-06-20) contains 20 submissions for this variant: 9 laboratories Benign, 3 Likely benign, 2 Uncertain significance, 1 likely benign, no expert-panel submission and no pathogenic submission; the aggregate API call returns 'Conflicting classifications of pathogenicity'.
• PMID:18422726 reports c.53C>T (p.Pro18Leu) once as a heterozygous germ-line change in 66 Japanese adenomatous polyposis patients (Table 2, allele frequency 0.76%, no homozygotes); the affected patient carried it as a compound heterozygote with p.Gly25Asp, p.Gln324His and c.1389G>C and was counted as MAP only because of that biallelic set. The authors state they did not functionally analyse p.Pro18Leu because proline 18 is not conserved across species.
• PMID:18811933 reports c.53C>T in cis with c.74G>A on the same allele (T/A haplotype; 6/138 sporadic CRC cases vs 3/343 controls) and shows that the single p.Pro18Leu mutant protein remained mitochondrial like wild type, concluding that either single substitution alone is insufficient to impair the mitochondrial targeting sequence.
• PMID:16929514 reports c.53C>T only in cis with c.74G>A (p.Gly25Asp) in suspected hereditary gastric cancer patients and controls, interprets the haplotype as a polymorphism rather than a pathogenic mutation, and found no biallelic MYH mutation in any patient.
• No source in this case classifies any variant encoding p.Pro18Leu as pathogenic, so the PS1 precondition (a previously established pathogenic variant with the same amino-acid change) is absent.

Sources

Reasoning

PS3 requires a well-established functional study showing a damaging effect. The governing framework is the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH v1.0 (cspec doc 1742141534), but it is held for this case with an empty rule_payload and an empty criteria list, so no VCEP-specific PS3 rule or strength calibration was available and the generic ACMG/AMP 2015 PS3 definition was applied. The most comprehensive variant-specific functional dataset is the E. coli MutY-complementation (rifampicin-resistance mutation-rate) assay of 47 MUTYH variants (Komine et al. 2015, Hum Mutat, PMID:25820570; open-access full text PMC4682456, retrieved directly because this paper was never fetched into the case). p.(Pro18Leu) gave a 2.7-fold increase in spontaneous mutation rate over wild type (26.0/10^8 vs 9.67/10^8) and normal protein level (0.96 of wild type), which the authors classified as 'partially defective' under their own cut-offs (<1.7-fold functionally retained; >=1.7 and <4.8-fold partially defective; >=4.8-fold defective). An intermediate, partial-defect result of this kind is not the well-established damaging effect PS3 requires, and the assay's calibration is demonstrably weak at exactly this point in its range: the same assay scored the definitively pathogenic founder variant p.Gly396D at only 1.7-fold (the retained/partial boundary) while benign population variants p.Val22M, p.Gln338H and p.Ser515F scored 1.2-, 0.7- and 1.0-fold, i.e. p.Pro18Leu at 2.7-fold sits above the pathogenic p.Gly396D value and overlaps the benign-to-partial range, so this assay cannot separate damaging from tolerated substitutions here. The authors themselves state that 'the clinical application of our results might be limited by the difficulty in establishing an appropriate cut-off value, as we could not estimate how intermediate or subtle functional defects would contribute to the pathogenesis of MAP' and that the cut-offs were 'tentatively set for convenience'. The only other variant-specific functional dataset is subcellular-localization work in cos-7 cells (PMID:18811933) which found the single p.Pro18Leu protein indistinguishable from wild type (mitochondrial retention); normal mitochondrial targeting is not a damaging effect, and that assay is a one-component proxy readout. No glycosylase-activity (8-oxoG:A adenine excision) measurement exists for this variant: the nicking assay in PMID:18422726 tested only p.Gly272Glu/p.Ala359Val, and its authors explicitly declined to assay P18L/G25D because residues 18 and 25 'were not homologous throughout the species'. PS3 is therefore not met at any strength.

Evidence considered

• Governing framework: the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH v1.0 (cspec doc 1742141534, svi URL cspec.genome.network/cspec/ui/svi/doc/1742141534) is present in this case but carries no machine-readable PS3/BS3 rules (framework_mode 'unstructured_ruleset', framework_complete false, rule_payload empty, criteria list empty), so the generic ACMG/AMP 2015 PS3/BS3 definitions were applied (PMID:25741868).
• PMID:25820570 (Komine et al., Hum Mutat 2015, 'Functional Complementation Assay for 47 MUTYH Variants in a MutY-Disrupted Escherichia coli Strain'; open access, PMC4682456; full text retrieved directly from Europe PMC for this assessment). Table 1: 'p.P18L c.53C>T 26.0 (2.7) 0.96 T (0.50) B (0.028) NE NE' — i.e. mutation rate 26.0/10^8 giving 2.7-fold wild type, protein level 0.96, SIFT tolerated 0.50, PolyPhen-2 benign 0.028, no glycosylase assay (NE), no previous rifampicin assay (NE). Categorisation text: 'Relative to the wild type, we categorized variant that had less than 1.7-fold higher mutation rate as functionally retained, more than or equal to 1.7-fold, and less than 4.8-fold as partially defective, and more than or equal to 4.8-fold as defective. According to this categorization, 17 variants (p.P18L, p.Y128H, p.P157L, p.G189E, p.A227V, p.V234M, p.R241W, p.R245C, p.R309C, p.P380T, p.G396D, p.P405S, p.R437P, p.A489T, p.V493F, p.G503E, and p.E480del) were partially defective'. Direct quote on limitation: 'the clinical application of our results might be limited by the difficulty in establishing an appropriate cut-off value, as we could not estimate how intermediate or subtle functional defects would contribute to the pathogenesis of MAP.'
• PMID:25820570 assay controls and calibration anchors (checked before scoring): wild-type human MUTYH suppressed the MutY-deficient E. coli CC104mutY mutation rate to 9.67/10^8 versus 137/10^8 for empty vector, demonstrating that complementation is measurable and providing the internal wild-type reference; benign population variants gave p.V22M 1.2-fold, p.Q338H 0.7-fold, p.S515F 1.0-fold; pathogenic founders gave p.Y179C 4.8-fold ('defective') and p.G396D 1.7-fold ('partially defective'). p.Pro18Leu at 2.7-fold is above the pathogenic p.G396D anchor and only ~2x the benign p.V22M anchor, so the assay does not demonstrate a damaging effect for this substitution.
• PMID:18811933 (Chen et al. 2008) variant-specific functional data: cos-7 cells transfected with FLAG-tagged hMYH type 1 constructs (wild type, single c.53C>T, single c.74G>A, double c.53C>T+c.74G>A haplotype), anti-FLAG immunofluorescence with MitoTracker Red and confocal microscopy. Direct quote: 'Interestingly, immunofluorescence showed that the hMYH proteins with single missense mutation, p.Pro18Leu or p.Gly25Asp, remained in the mitochondria, similar to the wild-type protein. It seems that the effect of the mutations are additive and either one of the missense mutations in the haplotype is not enough to impair the function of MTS domain.' No damaging effect of p.Pro18Leu alone was demonstrated, so this source cannot support PS3.
• PMID:18422726 (Yanaru-Fujisawa et al. 2008) records the variant in one heterozygous carrier among 66 Japanese adenomatous polyposis patients (Table 2, allele frequency 0.76%) and in a compound-heterozygous MAP patient (Table 5, patient 2), but reports no assay for it. Direct quote: 'we did not perform functional analysis of the MUTYH mutations (P18L and G25D) identified in patient 2 because the proline residue at position 18 and the glycine residue at position 25 were not homologous throughout the species.' The nicking assay in this paper used only recombinant mouse MUTYH G257E (impaired) and A341V (normal) constructs, corresponding to patient 1's p.Gly272Glu/p.Ala359Val.
• PMID:16929514 (case-control germline screening in suspected hereditary gastric cancer, Jiangsu, China): c.53C>T was found only in cis with c.74G>A (the 'T/A' haplotype), 10/101 patients vs 3/129 controls (p=0.019; haplotype allele frequency 4.95% vs 1.18%); no biallelic MYH mutation was found, and no functional assay of the variant was performed — the authors only speculate that N-terminal modification could affect MYH function. Checked and contributes no damaging-effect functional evidence.
• ClinVar submission text captured for this record (two criterion leads: Ambry Genetics SCV000184108 and Sinai Health Systems SCV000592673) references 'functional studies' for the c.53C>T/c.74G>A haplotype, but the captured sentence is truncated ('...were shown to retain co') and concerns the cis haplotype, so it was not relied on as PS3 evidence; no expert-panel submission exists for this record (0 of 20 submissions).

Still missing

• No glycosylase-activity or 8-oxoG:A excision measurement exists for p.(Pro18Leu); the only variant-specific datasets are a bacterial complementation assay with author-declared tentative cut-offs and a subcellular-localization proxy.
• The InSiGHT MUTYH VCEP specification is stored for this case as an unstructured ruleset with an empty rule payload and no criteria list, so any VCEP-specific PS3 rules/thresholds could not be applied; if the specification defines a validated MUTYH functional assay or its own cut-offs, this criterion should be re-run against it.
• PMID:25820570 (the 47-variant complementation assay, the decisive dataset here) is not a source_registry key in this case and its full text was never fetched; its data are therefore recorded only in evidence_used prose and not in the machine-readable citation fields.
• Candidate papers triaged for functional content but never fetched (PMID:26684191, PMID:11295288) could not be checked for variant-specific assay data; their abstracts describe mutation/SNP screening only, with no activity assay for p.Pro18Leu.

Sources

Reasoning

Governing framework: the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel Specification for MUTYH v1.0 (cspec doc 1742141534, svi_url https://cspec.genome.network/cspec/ui/svi/doc/1742141534) is the applicable gene-specific framework for this case (autosomal recessive, familial adenomatous polyposis 2 / MONDO:0012041, transcript NM_001128425.2). Its criterion-level rules were not captured in this case (rule_payload empty, framework_complete=false, criteria list empty), so no gene-specific PS4 numeric threshold or operator is available; the generic ACMG/AMP 2015 PS4 definition was applied instead (case-control prevalence significantly increased in affected vs unaffected individuals; that definition carries no numeric cutoff). Two studies report statistically significant over-representation, but neither isolates c.53C>T. (1) PMID:16929514: c.53C>T was reported 10/101 suspected hereditary gastric cancer patients vs 3/129 healthy controls (p=0.019, Table II); the c.53C>T/c.74G>A (T/A) haplotype allele frequency was 4.95% vs 1.18%, OR 4.43, 95% CI 1.33-14.72, p=0.021. Explicit comparison: lower 95% CI bound 1.33 >= 1.0? yes (significant association). However the association is reported for the T/A haplotype, the two substitutions were shown by cloning of the PCR product to lie on the same allele, the authors state 'Our study indicates that the variant haplotype T/A is a polymorphism in MYH gene', and no biallelic MYH mutation was found in any gastric cancer patient. (2) PMID:18811933: heterozygous T/A haplotype 6/138 (4.35%) sporadic CRC cases vs 3/343 (0.87%) controls, P=0.019; variant allele 2.17% vs 0.44%, P=0.020, OR 5.06, 95% CI 1.26-20.4; explicit comparison: lower 95% CI bound 1.26 >= 1.0? yes. This is likewise haplotype-level, and the same paper showed that each single substitution tested alone (including p.Pro18Leu) remained mitochondrially localized like wild type, concluding that 'either one of the missense mutations in the haplotype is not enough to impair the function of MTS domain'. Against this, the only cohort with the MUTYH-associated phenotype sought for PS4 (PMID:18422726; 66 Japanese adenomatous polyposis patients) found c.53C>T once, as a heterozygous change, allele frequency 0.76%, with no homozygotes and no control comparison, which is not case-control enrichment. The enriched phenotypes (gastric cancer; sporadic CRC risk in heterozygous carriers) are not the gene-disease relationship under assessment (familial adenomatous polyposis 2 / MUTYH-associated polyposis), and the enrichment signal is inseparable from the cis c.74G>A (p.Gly25Asp) haplotype. PS4 is therefore not met; the two significant case-control signals are flagged for human review because they are statistically real but haplotype-attributable and in a non-MAP phenotype.

Rule

Generic ACMG/AMP 2015 PS4 (PMID:25741868): case-control prevalence of the variant significantly increased in affected versus unaffected individuals (no numeric cutoff in the generic definition). MUTYH InSiGHT v1.0 gene-specific PS4 rule and threshold not available in this case (cspec doc 1742141534, criterion-level rules not captured; framework_complete=false).

Evidence considered

• VCEP/rule check: the governing framework for MUTYH v1.0 (ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel Specification, cspec doc 1742141534) is present but unstructured in this case - rule payload empty and framework_complete=false - so no VCEP-specific PS4 threshold or comparison operator exists to apply; generic ACMG/AMP 2015 PS4 (PMID:25741868) was used, and that definition is qualitative (statistically significant increase in affected vs controls) with no numeric cutoff.
• PMID:16929514: case-control germline screen of 101 suspected hereditary gastric cancer patients vs 129 healthy controls of the same geographic origin (Jiangsu, China; PCR-DHPLC plus TA cloning to phase the two substitutions). Table II row for c.53C>T (p.Pro18Leu): 10/101 patients vs 3/129 controls, p=0.019. The authors' reported association is for the in-cis c.53C>T/c.74G>A (T/A) haplotype: allele frequency 4.95% vs 1.18%, OR 4.43, 95% CI 1.33-14.72, p=0.021. Explicit inequality for the cited statistic: 1.33 >= 1.0? yes. The authors explicitly conclude the haplotype is a polymorphism of MYH, report no biallelic MYH mutation in any gastric cancer patient, and note the effect was driven by family-history and young-onset subgroups.
• PMID:18811933: case-control study of 138 Chinese sporadic CRC patients vs 343 healthy controls. Heterozygous T/A haplotype in 6/138 (4.35%) cases vs 3/343 (0.87%) controls, P=0.019; variant allele frequency 2.17% vs 0.44%, P=0.020, OR 5.06, 95% CI 1.26-20.4; explicit inequality for the cited statistic: 1.26 >= 1.0? yes. The paper states it could not exclude selection bias given the low variant frequency, and its functional arm showed that the single p.Pro18Leu protein remained mitochondrially localized like wild type ('either one of the missense mutations in the haplotype is not enough to impair the function of MTS domain').
• PMID:18422726: the only retrieved cohort with the MUTYH-associated phenotype (66 Japanese patients with adenomatous polyposis) detected c.53C>T once as a heterozygous germ-line change (allele frequency 0.76%, no homozygotes) and reports no control group or enrichment statistics; a single heterozygous carrier in a case series does not demonstrate increased prevalence in affected individuals.
• Phenotype/attribution mismatch: the two statistically significant case-control signals concern heterozygous carriers of the c.53C>T/c.74G>A haplotype in gastric cancer and in sporadic CRC risk, not the recessive gene-disease relationship under assessment (familial adenomatous polyposis 2 / MUTYH-associated polyposis), so they cannot be attributed to c.53C>T through the MUTYH-associated phenotype.
• Numeric-threshold protocol: no VCEP numeric threshold exists for PS4 in this case, so direction of comparison was taken only from the statistics actually reported (odds-ratio 95% CI lower bounds vs 1.0, written above as explicit inequalities); no operator was borrowed from another criterion in this group.

Still missing

• MUTYH InSiGHT v1.0 criterion-level PS4 rule (threshold/operator) not captured in this case; no gene-specific numeric cutoff could be applied.
• PMID:11295288 and PMID:22641385 (nominated by ClinVar submissions SCV000184108 / SCV001470575) were not retrieved as full text; the available abstract of PMID:11295288 describes MYH polymorphism screening in lung cancer rather than a MAP case-control design, so it is unlikely to change the PS4 verdict.
• No case-control study of c.53C>T in MUTYH-associated polyposis (biallelic or MAP-suspected cases vs controls) was identified in the retrieved evidence.

Sources

Reasoning

PM1 requires the residue to lie in a mutational hotspot and/or a critical, well-established functional domain that is not a site of benign variation. The governing ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis specification for MUTYH (v1.0, CSPEC document 1742141534) is listed as 'Pilot Rules In Prep' and publishes no approved criteria codes and no domain table, and vcep_materials.json contains an empty domain_tables object, so there is no authoritative, approved MUTYH critical-domain entry against which residue 18 could be placed inside a PM1 region. Independently of the absent specification, Pro18 lies in the N-terminal mitochondrial-targeting / RPA-binding region rather than in the adenine DNA glycosylase catalytic core or the C-terminal 8-oxoG-recognition/MSH6-interaction region; PMID:16929514 places p.Pro18Leu in that N-terminal mitochondrial targeting signal / RPA-binding region and only speculates that an N-terminal modification could weaken MYH function. That region is populated by benign missense variation: p.Pro18Leu itself reaches 1.3% (260/19,954 alleles, 4 homozygotes) in East Asian gnomAD v2.1 and is classified Benign by the majority of ClinVar submitters, and its cis partner p.Gly25Asp is carried as a common benign haplotype in cases and controls (PMID:16929514, PMID:18811933). CancerHotspots contains no statistically significant hotspot at MUTYH codon 18. PM1 is therefore not met.

Rule

PM1 (ACMG/AMP 2015): located in a mutational hot spot and/or critical and well-established functional domain (e.g. active site) without benign variation. Authoritative VCEP domain-table entries take precedence when present; none are present for MUTYH.

Evidence considered

• The governing framework is the ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH v1.0 (CSPEC doc 1742141534, InSiGHT VCEP). Its CSPEC page carries status badges 'Classification Rules In Prep' (1/24/2025), 'Classification Rules Submitted' (4/11/2025), 'Pilot Rules In Prep' (4/24/2025, active) and 'Pilot Rules Submitted' (7/1/2024), and its 'Rules for MUTYH' panel lists only gene, transcripts NM_001128425.2 / NM_001048174.2, disease familial adenomatous polyposis 2 (MONDO:0012041) and autosomal recessive inheritance, with no criteria table rendered.
• vcep_materials.json for this case contains an empty domain_tables object, so there is no authoritative list of approved MUTYH critical functional domains to test residue 18 against.
• PMID:16929514 speculates that p.Pro18Leu lies in the N-terminal RPA-binding domain / mitochondrial targeting signal and that an N-terminal modification could weaken MYH association with its product; it performed no functional assay and reported the change only as part of a haplotype with p.Gly25Asp.
• gnomAD v2.1 reports the variant (1-45800167-G-A) at total AF 0.00105348 (298/282,872 alleles, 4 homozygotes) with a maximum population frequency of AF 0.01303 (260/19,954 alleles, 4 homozygotes) in East Asian samples and grpmax FAF 0.0132979, i.e. the N-terminal region tolerates this missense change at appreciable population frequency.
• ClinVar submissions for the variant are dominated by Benign (9 laboratories) and Likely benign (3 laboratories) classifications without any expert-panel submission, providing no evidence that codon 18 is a constrained, hotspot-like residue.
• The case's CancerHotspots query for MUTYH P18 returned no hotspot row and the assessment sentence states the variant does not lie in a statistically significant hotspot (cancerhotspots.org was searched for this residue; that resource is not a keyed source in this case's source_registry).

Still missing

• No approved MUTYH domain table exists: the governing CSPEC specification publishes no criteria or domain entries and vcep_materials.json domain_tables is empty, so PM1 was assessed against generic definitions plus the case's residue-level evidence.

Sources

Reasoning

No MUTYH-specific PM2 rule is populated in the governing InSiGHT specification, so the supplied generic ClinGen-SVI PM2 supporting threshold of allele frequency <= 0.0001 (0.01%) applies to the all-comers source. The variant is far too common in every source and every ancestry: gnomAD v2.1 all-comers AF 0.00105348 (0.105%; 298/282,872 alleles) and gnomAD v4.1 all-comers AF 0.00044112 (0.044%; 712/1,614,074 alleles), with non-cancer exome AF 0.00111422 and non-cancer genome AF 0.000581584, and East Asian frequencies of 1.13-1.32%. The ACMG/AMP PM2 formulation - absence, or a frequency below the expected carrier frequency for a recessive disorder, in a race-matched large general population or control cohort - is therefore not satisfied, and the observed homozygotes further argue against rarity. PM2 is not met.

Rule

{'source': 'generic_clingen_svi_default', 'threshold': 0.0001, 'comparison': '<=', 'unit': 'allele frequency', 'population_selection': 'all-comers default unless the governing specification requires a different source'}

Evidence considered

• The retrieved InSiGHT MUTYH specification has no populated PM2 threshold, so the supplied generic PM2 supporting threshold was applied.
• gnomAD v2.1 all-comers AF 0.00105348 (298/282,872 alleles, 4 homozygotes) and gnomAD v4.1 all-comers AF 0.00044112 (712/1,614,074 alleles, 7 homozygotes) are both above 0.0001.
• gnomAD v2.1 non-cancer exomes AF 0.00111422 (264/236,936 alleles) and gnomAD v3.1 non-cancer genomes AF 0.000581584 (86/147,872 alleles) are likewise above 0.0001, so no available source supports rarity.
• Ancestry-specific frequencies are much higher still: 0.01302997 (v2.1 East Asian) and 0.01133883 (v4.1 East Asian).
• The supplied generic ClinGen-SVI PM2 recommendation used is allele frequency <= 0.0001 (0.01%), attributed to ACMG/AMP guidance (PMID:25741868), whose PM2 wording requires absence or a frequency below the expected carrier frequency for a recessive disorder in a race-matched population.

Sources

Reasoning

Framework and scope first: MUTYH-associated polyposis (familial adenomatous polyposis 2, MONDO:0012041) is inherited autosomal recessively, which is the only setting in which PM3 applies. The gene-level ruleset of the governing ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH v1.0 (doc 1742141534) records 'Autosomal recessive inheritance', and PMID:21325953 states that 'MYH-associated polyposis syndrome is a recently characterized, autosomal recessive, polyposis syndrome caused by biallelic mutations in the MYH gene.' The MUTYH specification was retrieved with an empty criteria payload (framework_complete=false, framework_mode=unstructured_ruleset), so it supplies no criterion-level PM3 rule and the generic ACMG/AMP PM3 definition was applied (PMID:25741868): the variant must be shown, with phase resolved, to be in trans with a pathogenic (or likely pathogenic) variant in an affected individual. The published phase and genotype data do not satisfy that requirement. PMID:18422726 describes one affected proband, patient 2 (male, polyposis diagnosed at 47 years, 147 colorectal adenomas, colorectal carcinoma, lipoma), who was heterozygous for four MUTYH changes - P18L (c.53C>T), G25D, Q324H and c.1389G>C - and whom the authors regarded as a compound heterozygote with MAP; however, the paper never states which of the other three changes lies on the opposite allele, and it expressly did not functionally analyse P18L or G25D. Independent phasing of the same two exon-2 changes in PMID:16929514 and PMID:18811933 places c.53C>T and c.74G>A (p.Gly25Asp) on the same allele (cis), which excludes G25D as the trans partner. That leaves as candidate trans partners Q324H, which PMID:18422726 itself designates a polymorphism (49% allele frequency in its cohort, no functional alteration in DNA repair activity), and c.1389G>C, for which no pathogenicity evidence is presented in any source of this case. Neither PMID:16929514 ('No biallelic MYH mutation was found in any GC patients') nor PMID:18811933 ('no homozygote was found in the present study') documents any biallelic carrier. No source therefore reports c.53C>T in trans with an established pathogenic MUTYH variant in an affected proband, and PM3 is not met.

Rule

Generic ACMG/AMP PM3 (PMID:25741868): for a recessive disorder, a pathogenic (or likely pathogenic) variant observed in trans with the variant under assessment in an affected individual, with phase and partner pathogenicity both established. No criterion-level PM3 rule is supplied by the governing MUTYH InSiGHT specification (empty criteria payload), so the generic rule governed.

Evidence considered

• Inheritance prerequisite for PM3: the gene-level ruleset of the governing ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH v1.0 (doc 1742141534, source_registry key 'cspec') records 'modeOfInheritance: Autosomal recessive inheritance' for familial adenomatous polyposis 2 (MONDO:0012041), and PMID:21325953 states that MAP is 'an autosomal recessive, polyposis syndrome caused by biallelic mutations in the MYH gene'. PM3 is only applicable to recessive disorders.
• Framework limitation: the MUTYH specification was retrieved with an empty criteria payload (framework_complete=false, framework_mode=unstructured_ruleset, rule_payload {}), so it pre-assigns no PM3 code and provides no gene-specific PM3 rule; the generic ACMG/AMP PM3 definition was used (PMID:25741868: for recessive disorders, detected in trans with a pathogenic variant).
• Affected-proband evidence: PMID:18422726 reports one affected proband relevant here - Table 5 patient 2, male, polyposis diagnosed at 47 years, 147 colorectal adenomas, colorectal carcinoma, lipoma - with the MUTYH genotype P18L, G25D, Q324H and c.1389G>C. The authors wrote 'Because these two patients had biallelic mutations of MUTYH, we regarded them as having MAP', but also 'As patient 2 is actually a compound heterozygote of those mutations together with Q324H and nucleotide change of 1389G.C, the possible functional alteration of MUTYH in such a combination of the variants needs to be examined further' - the phase of P18L relative to any particular partner is not established.
• Phase of c.53C>T itself: PMID:16929514 - 'Sequencing of several cloned PCR products revealed that the 2 variations affected the same allele with retention of a wild-type allele (Fig. 1c). No biallelic MYH mutation was found in any GC patients.' PMID:18811933 - 'Further cloning-sequencing showed that the mutations occurred at the same allele (haplotype T/A variation)' and 'no homozygote was found in the present study.' Both place c.53C>T in cis with c.74G>A (p.Gly25Asp), so G25D cannot be the allele in trans with c.53C>T.
• Partner-variant pathogenicity: the surviving candidate trans partners are Q324H, which PMID:18422726 concludes 'should be regarded as a polymorphism of MUTYH even in Japanese population, which is unrelated to phenotypic features of FAP' (allele frequency 49% in that cohort, no functional alteration in DNA repair activity) and c.1389G>C, for which no pathogenicity evidence appears anywhere in this case. No source documents c.53C>T in trans with a variant of established pathogenicity.
• No functional support exists for the partner allele in the one alleged compound heterozygote: PMID:18422726 states 'we did not perform functional analysis of the MUTYH mutations (P18L and G25D) identified in patient 2 because the proline residue at position 18 and the glycine residue at position 25 were not homologous throughout the species.'

Still missing

• The phase of the four heterozygous MUTYH changes in the PMID:18422726 patient 2 genotype was never resolved by the authors, so c.53C>T cannot be placed on a defined allele opposite a characterised variant.
• No source in this case reports c.53C>T in trans with a MUTYH variant that is itself classified pathogenic or likely pathogenic (for example Y165C or G382D); such a report would be needed to meet PM3.

Sources

Reasoning

PM5 requires a missense change at the same residue (Pro18) different from the one under assessment, with that comparator established as pathogenic. A residue-level search of ClinVar covering MUTYH (4,247 variant records retrieved and screened) found no missense variant at codon 18 other than the case variant itself: the only codon-18 entries are NM_001048174.2(MUTYH):c.11C>T (p.Pro4Leu) — the same c.53C>T change reported on the alternative transcript — in its two records, and all remaining 'Pro18x' MUTYH entries are p.Pro185 substitutions. The case's own PM5 candidate harvest independently returned zero candidates (eligible_for_classic_pm5_search false, total_candidates_seen 0). The literature reporting this region (PMID:18422726, PMID:18811933, PMID:16929514) reports p.Pro18Leu itself and the adjacent p.Gly25Asp at a different residue, never a second, pathogenic substitution at Pro18. With no same-residue pathogenic comparator, PM5 is not met.

Rule

PM5 (ACMG/AMP 2015): missense change at an amino-acid residue where a different missense change determined to be pathogenic has been seen before. No MUTYH-specific PM5 rule is published in the governing specification.

Evidence considered

• Target residue is MUTYH Pro18 with observed substitution Pro18Leu (NP_001121897.1:p.(Pro18Leu)).
• ClinVar was screened across all 4,247 MUTYH records (NCBI ClinVar E-utilities esearch/esummary): codon-18 entries are limited to NM_001048174.2(MUTYH):c.11C>T (p.Pro4Leu), which is the same c.53C>T variant reported on the alternative transcript (variation 41759 and its companion record, conflicting classifications, no pathogenic and no expert-panel submission). No different missense substitution at Pro18 exists in ClinVar, so no pathogenic same-residue comparator can be established.
• pm5_candidates.json for this case reports found=false, eligible_for_classic_pm5_search=false, total_candidates_seen=0 and same_residue_candidates=0, with the note that classic same-residue PM5 semantics could not be safely confirmed and that PM5 must not be applied from the artifact alone.
• PMID:18422726 reports p.Pro18Leu compound-heterozygous with p.Gly25Asp (residue 25), p.Gln324His and c.1389G>C, and cites prior detection of p.Pro18Leu with p.Ala359Val and p.Gly25Asp in Korean patients — no second Pro18 substitution and no pathogenic comparator at residue 18.
• PMID:18811933 and PMID:16929514 report p.Pro18Leu only in cis with p.Gly25Asp (a different residue) and interpret the haplotype as a polymorphism.

Sources

Reasoning

Generic PP2 has two arms and both must hold: missense variants must be a common mechanism of disease in the gene, and the gene must have a low rate of benign missense variation. The first arm is satisfied for MUTYH — biallelic MUTYH-associated polyposis has been attributed to biallelic missense genotypes (PMID:18422726 reports a patient with biallelic p.Gly272Glu and p.Ala359Val; PMID:21325953 describes the recessive biallelic mechanism). The second arm fails: gnomAD gene constraint for MUTYH (ENSG00000132781) gives a missense z-score of 0.63 with pLI approximately 0 and observed/expected LoF 0.88, i.e. no evidence of constraint against missense variation, and the gene carries well-documented benign/common missense alleles — p.Gln324His is described as a polymorphism (PMID:18422726) and the p.Pro18Leu/p.Gly25Asp haplotype is present at 1.3% in East Asian gnomAD v2.1 and in healthy controls (PMID:16929514, PMID:18811933). No MUTYH-specific PP2 rule is available (the InSiGHT CSPEC v1.0 rules are still in preparation), so the generic two-part test governs and PP2 is not met.

Rule

PP2 (ACMG/AMP 2015, supporting): missense variant in a gene that has a low rate of benign missense variation and in which missense variants are a common mechanism of disease.

Evidence considered

• The governing InSiGHT MUTYH specification (CSPEC doc 1742141534, v1.0) publishes no approved criteria codes, so no gene-specific PP2 rule was available and the generic ACMG/AMP 2015 definition was applied.
• gnomAD v4.1 gene constraint for MUTYH (gene_id ENSG00000132781, queried through the gnomAD API) returns missense z = 0.63, pLI = 1.6e-19, observed/expected LoF = 0.88 (upper bound 1.08) — no significant constraint on missense or LoF variation.
• gnomAD v2.1 shows the p.Pro18Leu haplotype allele at 1.3% in East Asian samples (260/19,954; 4 homozygotes), demonstrating that benign missense variation in MUTYH can reach appreciable population frequency.
• PMID:18422726 reports a MAP patient with biallelic missense changes (p.Gly272Glu, p.Ala359Val) while also stating that p.Gln324His should be regarded as a polymorphism — so missense can be a disease mechanism but the gene also carries common benign missense alleles.
• PMID:21325953 describes MYH-associated polyposis as an autosomal recessive polyposis syndrome caused by biallelic MYH mutations, establishing the recessive biallelic mechanism without supporting a low benign missense rate.

Still missing

• No approved VCEP PP2 rule or curated per-gene benign-missense-rate metric exists for MUTYH; the generic ACMG/AMP two-part definition plus gnomAD gene constraint were used instead.

Sources

Reasoning

Scope check performed before any predictor score was read: NM_001128425.2:c.53C>T in MUTYH is a missense substitution (NP_001121897.1:p.(Pro18Leu)) per the normalized prefetch consequence, so it is in scope for PP3, but its PP3 evidence must come from the missense protein-impact path (REVEL) and never from the splice path. The governing gene framework - ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel Specification to the ACMG/AMP guidelines for MUTYH, v1.0, doc 1742141534 - was retrieved, but its criteria payload is empty (framework_complete=false, framework_mode=unstructured_ruleset), so it pre-assigns no PP3 code and no gene-specific PP3 rule blocks or modifies the generic fallback. Under the generic ClinGen SVI REVEL calibration (Pejaver et al. 2022, Am J Hum Genet, PMID:36413997), PP3 requires REVEL >= 0.644 (supporting), >= 0.773 (moderate) or >= 0.932 (strong). The local REVEL v1.3 lookup returns 0.2 for this variant, below every PP3 threshold, so PP3 is not met at any strength. The splice path neither adds nor rescues PP3 for this variant: SpliceAI returned no score (query timeout; max_delta null, so it cannot be read as 0.00), and the Pangolin values that were returned (SG 0.007, SL -0.001) are not calibrated for PP3/BP4. BayesDel (-0.0488581) has no established ACMG-strength calibration and was treated as not available. No double-counting: the single REVEL score is the only computational result used for this group.

Rule

Generic fallback, missense path: PP3 (supporting) requires REVEL >= 0.644; PP3 not met at REVEL 0.2 (Pejaver et al. 2022, PMID:36413997).

Evidence considered

• Scope: normalized variant consequence is missense (NP_001121897.1:p.(Pro18Leu)), so only the REVEL missense path is evaluable for PP3; the SpliceAI splice path cannot supply PP3 for a missense variant.
• Governing framework: ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel Specification to ACMG/AMP for MUTYH v1.0 (doc 1742141534) was retrieved but carries an empty criteria payload (framework_complete=false, unstructured_ruleset), so it contains no gene-specific PP3 rule and no pre-assigned PP3 code; the generic ACMG/AMP fallback applies.
• REVEL score for NM_001128425.2:c.53C>T = 0.2 (local REVEL v1.3 lookup, source_registry key 'revel'). PP3 supporting threshold is REVEL >= 0.644, moderate >= 0.773, strong >= 0.932 (Pejaver et al. 2022, Am J Hum Genet, PMID:36413997); 0.2 is below all three, so PP3 is not met at any strength.
• SpliceAI returned no score for this variant (query error: timeout; max_delta null rather than 0.00) and the returned Pangolin scores (SG 0.007, SL -0.001) are not calibrated for PP3/BP4; neither was substituted for the REVEL result.
• BayesDel = -0.0488581 is present in the case data but no ACMG-strength calibration is established for it, so it was treated as not available for PP3/BP4.

Sources

Reasoning

Generic ACMG/AMP 2015 PP4 (PMID:25741868) requires the patient's phenotype or family history to be highly specific for a disease with a single genetic etiology; the InSiGHT MUTYH v1.0 specification could not be retrieved at criterion level for this case, so no gene-specific PP4 rule was available. This case is variant-level: no proband, indication, family history or HPO terms accompany c.53C>T, so the patient's own phenotype cannot be assessed. The only published individual carrying c.53C>T with a documented phenotype is patient 2 of PMID:18422726 (male, colorectal carcinoma at 47 years, 147 colorectal adenomas, lipoma, no extracolonic polyposis, APC-negative), who also carried p.Gly25Asp in cis plus Q324H and 1389G>C and was regarded by the authors as having MAP. An adenomatous polyposis/CRC presentation is consistent with, but not highly specific for, MUTYH-associated polyposis: the same phenotype is produced by APC-associated familial adenomatous polyposis and overlaps Lynch syndrome, and PMID:21325953 states directly that genetic testing for MYH is complicated by phenotypic overlap with other colorectal cancer syndromes. The enriched phenotype reported in the case-control literature was gastric cancer, which is not the MUTYH disease phenotype at all. PP4 demands specificity beyond mere consistency with the disease, which the available phenotype evidence does not provide, and there is no proband-level phenotype to test.

Rule

Generic ACMG/AMP 2015 PP4 (PMID:25741868): patient's phenotype or family history is highly specific for a disease with a single genetic etiology. MUTYH InSiGHT v1.0 gene-specific PP4 rule not available in this case (cspec doc 1742141534, criterion-level rules not captured).

Evidence considered

• PMID:18422726 documents the only phenotype-bearing individual known to carry c.53C>T (patient 2: male, age 47 at diagnosis, no family history, 147 colorectal adenomas, colorectal carcinoma positive, no duodenal adenomas, lipoma, mutations P18L, G25D, Q324H, 1389G>C, APC-negative), which the authors classed as MAP because they considered the MUTYH changes biallelic; the presentation is colon polyposis/CRC, a phenotype not specific to MUTYH.
• PMID:21325953 frames the specificity problem explicitly for this gene: MYH-associated polyposis is autosomal recessive and caused by biallelic MYH mutations, with a highly variable clinical picture that overlaps classical FAP-through-Lynch-spectrum phenotypes, and it notes that genetic testing for MYH is complicated by phenotypic overlap with other colorectal cancer syndromes.
• PMID:25741868 (generic ACMG/AMP 2015) defines PP4 as a patient's phenotype or family history being highly specific for a disease with a single genetic etiology; consistency alone is insufficient.
• Case-level gap: this case contains no proband phenotype, indication, family history or other clinical data for the individual in whom the variant was found, so a phenotype-specificity assessment rests solely on published case reports.

Still missing

• No proband-level phenotype, indication for testing, family history or HPO terms are present in the case evidence, so PP4 could not be tested against the patient's own presentation.
• MUTYH InSiGHT v1.0 criterion-level PP4 rule not captured in this case.

Sources

Reasoning

PP5 requires an exact-variant assertion from a reputable (expert-panel) source that the variant is pathogenic, and per the rules for this group only an exact-variant ClinVar expert-panel classification may trigger it; ordinary laboratory submissions and aggregate ClinVar labels must never do so. The exact-variant ClinVar record for MUTYH c.53C>T (ClinVarID 41759, VCV000041759, also 41763) contains 20 submissions from 19 clinical laboratories; zero submissions come from an expert panel (expert_panel_submissions = 0; is_expert_panel = false for every submission), and the record's review status is 'criteria provided, conflicting classifications' (1 star, api_classification = 'Conflicting classifications of pathogenicity'). The submitter classifications are Benign (9), Likely benign (3), likely benign (1) and Uncertain significance (2) - conflicting ordinary laboratory assertions, none of which is a Pathogenic or Likely pathogenic expert-panel classification. No ClinGen InSiGHT/VCEP expert-panel classification of c.53C>T exists in the retrieved evidence, so PP5 cannot be triggered.

Rule

PP5 as specified for this group: trigger only on an exact-variant ClinVar expert-panel classification of Pathogenic or Likely pathogenic; ordinary laboratory submissions, aggregate ClinVar labels and non-expert assertions never trigger it. Generic PP5 origin: ACMG/AMP 2015 (PMID:25741868).

Evidence considered

• ClinVar (ClinVarID 41759, VCV000041759, last updated 2026-06-20): 20 submissions, 19 submitters, review status 'criteria provided, conflicting classifications' = 1 star; api_classification 'Conflicting classifications of pathogenicity'; submissions by clinical laboratories: Benign 9, Likely benign 3, likely benign 1, Uncertain significance 2. Submitting laboratories include Ambry Genetics, LabCorp, GeneDx, ARUP, Mayo Clinic Laboratories, Color Health, PreventionGenetics, Invitae/Labcorp, CeGaT, Mendelics, Sema4, Quest Diagnostics and others - all ordinary clinical laboratories.
• Submission audit for this case: expert_panel_submissions = 0, and every one of the 20 submissions is flagged is_expert_panel = false; the ClinVar expert_panels field is empty. No Pathogenic or Likely pathogenic classification from an expert panel exists for the exact variant, so the PP5 trigger condition is absent.
• The InSiGHT Hereditary Colorectal Cancer/Polyposis MUTYH expert-panel specification (cspec doc 1742141534) is present as a framework in this case but contains no variant-level classification of c.53C>T, and its criterion-level rules were not captured.
• No inference was drawn from any other source's final classification; PP5 is decided solely on the presence or absence of an exact-variant expert-panel assertion.

Still missing

• No ClinGen expert-panel (VCEP) classification of MUTYH c.53C>T is available in ClinVar or in the retrieved evidence; if one is published, PP5 must be re-evaluated.

Sources

Reasoning

The ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis VCEP specification for MUTYH (CSPEC doc 1742141534, version 1.0) is identified for this case but carries an empty rule payload and no criteria-specific population thresholds, and the CSPEC record lists the MUTYH ruleset as still in preparation, so no gene-specific BA1 rule exists and the supplied generic stand-alone threshold of allele frequency >= 0.05 (5%) applies. The highest credible frequency for NM_001128425.2:c.53C>T (p.Pro18Leu) is 0.01302997 (260/19,954 alleles, 4 homozygotes) in the gnomAD v2.1 East Asian subgroup, with grpmax FAF 0.0132979; the gnomAD v2.1 non-cancer East Asian exome frequency is 0.01324538 (255/19,252), gnomAD v4.1 East Asian frequency is 0.01133883 (509/44,890) and gnomAD v4.1 all-comers frequency is 0.00044112 (712/1,614,074). No population approaches 5%, so the variant is far too common to be ignored but nowhere near stand-alone benign frequency. The single value above 5% seen anywhere (HGDP:YIZU 0.0555556, 1/18 alleles, in the v3.1 non-cancer genomes) is a small-sample artifact and was not treated as a population frequency. BA1 is not met.

Rule

{'source': 'generic_acmg_amp', 'threshold': 0.05, 'comparison': '>=', 'unit': 'allele frequency', 'population_selection': 'highest observed ancestry frequency'}

Evidence considered

• The retrieved InSiGHT MUTYH specification (CSPEC doc 1742141534, v1.0) has an empty rule payload and lists no criteria-specific population thresholds; the CSPEC record shows the MUTYH classification rules still in preparation, so the supplied generic ACMG/AMP BA1 threshold was applied.
• gnomAD v2.1: all-comers AF 0.00105348 (298/282,872 alleles, 4 homozygotes); highest subgroup East Asian AF 0.01302997 (260/19,954 alleles, 4 homozygotes); grpmax FAF 0.0132979.
• gnomAD v4.1: all-comers AF 0.00044112 (712/1,614,074 alleles, 7 homozygotes); East Asian AF 0.01133883 (509/44,890 alleles, 7 homozygotes); grpmax FAF 0.01052478.
• gnomAD non-cancer subsets were checked under the default all-comers source: v2.1 non-cancer exomes AF 0.00111422 (264/236,936 alleles, 4 homozygotes; East Asian 0.01324538, 255/19,252) and v3.1 non-cancer genomes AF 0.000581584 (86/147,872 alleles, no homozygotes). These are concordant with the all-comers totals and change no verdict; the governing specification designates no non-cancer or exome-only source, so the all-comers totals remain the default.
• The generic ACMG/AMP BA1 definition used is a control-population allele frequency greater than 5% for stand-alone benign support (Richards et al. 2015, PMID:25741868), evaluated against the highest credible ancestry-specific frequency.
• The gnomAD v3.1 non-cancer HGDP:YIZU value of 0.0555556 (1/18 alleles) was not used as a population frequency because of the very small sample size.
• PMID:24033266 states that for autosomal recessive disorders the heterozygous carrier frequency of a pathogenic allele must be less than twice the square root of the disease prevalence, i.e. a recessive gene tolerates a higher pathogenic allele frequency than a dominant one; no MUTYH-associated polyposis prevalence estimate was available in this case's evidence to convert that rule into a numeric BA1 cutoff.

Still missing

• No MUTYH-specific BA1 threshold is published (the InSiGHT MUTYH ruleset is unpopulated), so the recessive disease architecture could not be incorporated into a gene-derived stand-alone frequency cutoff.

Sources

Reasoning

BS3 requires well-established functional studies showing no damaging effect on protein function or splicing. No such dataset exists for p.(Pro18Leu). The best-controlled MUTYH functional dataset, the E. coli MutY-complementation assay of 47 variants (Komine et al. 2015, PMID:25820570), detects a partial loss of function for this variant (2.7-fold increased spontaneous mutation rate relative to wild type, authors' category 'partially defective') — not a wild-type-like result; in that assay functionally retained benign variants scored <=1.2-fold (p.Val22M 1.2, p.Ser515F 1.0, p.Gln338H 0.7), so a genuine 'no damaging effect' result would have required a value in that range. The only result pointing toward normal function is the subcellular-localization study PMID:18811933, where the single p.Pro18Leu protein remained mitochondrial like wild type in cos-7 cells. That readout is one component of MUTYH biology (N-terminal mitochondrial targeting) measured by heterologous overexpression, not the protein's full biological function (8-oxoG:A adenine glycosylase activity); the authors of that study state that glycosylase activity and mitochondrial genome stability 'have not been directly investigated' and call the data preliminary, and PMID:18422726 records that the residue is not conserved across species, further weakening the biological relevance of a single, non-validated localization readout as proof of normal function. The two datasets also do not agree: one shows a partial defect, the other apparent normality. Applying the ACMG/AMP 2015 requirement that assay evidence be judged by how closely it reflects the biological environment, and that evidence from one component of function is weaker than a full-function readout (PMID:25741868), neither dataset qualifies as a well-established no-damaging-effect study. BS3 is not awarded at any strength; the mitochondrial-localization result remains limited, benign-leaning functional information that is not criterion-grade.

Evidence considered

• Governing framework: ClinGen InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH v1.0 (cspec doc 1742141534) is present but has no machine-readable BS3 rule (empty rule_payload, empty criteria list), so the generic ACMG/AMP 2015 BS3 definition was applied (PMID:25741868).
• PMID:25820570 (Komine et al. 2015, Hum Mutat; open-access PMC4682456, retrieved directly) is the assay with the strongest controls available for this gene: wild-type human MUTYH restored E. coli CC104mutY rifampicin-resistance mutation rate from 137/10^8 (empty vector) to 9.67/10^8, and 47 variants were scored against two cut-offs. p.P18L (c.53C>T) = 26.0/10^8, i.e. 2.7-fold wild type, with normal protein level (0.96), and was listed by the authors among the 'partially defective' variants — a detectable partial functional defect, therefore not 'no damaging effect'. Direct quote: 'Relative to the wild type, we categorized variant that had less than 1.7-fold higher mutation rate as functionally retained, more than or equal to 1.7-fold, and less than 4.8-fold as partially defective, and more than or equal to 4.8-fold as defective.'
• PMID:18811933 variant-specific result relevant to BS3 (the only normal-function-leaning data): cos-7 cells expressing FLAG-tagged hMYH type 1; immunofluorescence with MitoTracker Red showed the single p.Pro18Leu construct 'remained in the mitochondria, similar to the wild-type protein', while only the double c.53C>T/c.74G>A haplotype construct showed dual nuclear plus mitochondrial localization. Limitations recorded by the authors themselves: 'the adenine glycosylase activity of the mutants in mitochondria and mitochondrial genome stability under the influence of the mutants have not been directly investigated. Thus, validation of these results on larger cohorts and further functional studies are imperatively necessary.'
• PMID:25741868 (ACMG/AMP 2015 standards, 'PS3 BS3 Functional studies' section) was used to calibrate assay acceptability rather than a numeric threshold: 'To assess the validity of a functional assay, one must consider how closely the functional assay reflects the biological environment.' and 'evidence is stronger if the assay reflects the full biological function of the protein (e.g. substrate breakdown by an enzyme) as compared to one component of function (e.g. ATP hydrolysis for a protein with additional binding properties).' A mitochondrial-targeting immunofluorescence readout is a one-component assay by this standard, and the bacterial complementation assay measures repair function but returns an intermediate value for this variant.
• PMID:18422726 documents that p.Pro18Leu was deliberately not functionally assayed ('the proline residue at position 18 and the glycine residue at position 25 were not homologous throughout the species'), which further limits the biological relevance of any single readout for this position.
• SpliceAI lookup for NM_001128425.2:c.53C>T returned no score (spliceai_available false; max_delta_score null; error 'timeout'; Pangolin only, SG 0.007) and the case record explicitly warns this is missing data, not evidence of absent splice impact; no splice-assay evidence therefore exists for BS3.
• Papers checked for BS3-relevant functional or splicing data on this variant and found to contain none: PMID:21325953 (MYH-associated polyposis review), PMID:22855150 (SEAP/SEOM CRC biomarker consensus), PMID:23852704 (EGTM tumour-marker guidelines), PMID:24033266 (variant assessment tool methodology) — all reviews/methodology documents with no MUTYH assay of c.53C>T.

Still missing

• BS3 could only be reconsidered with a validated, calibrated full-function MUTYH assay (e.g. in vitro adenine glycosylase activity on 8-oxoG:A substrates with established benign and pathogenic reference controls) showing p.(Pro18Leu) indistinguishable from wild type; no such data exist for this variant.
• No splicing assay or splice prediction is available for this variant (SpliceAI returned no score: upstream timeout, max_delta null and explicitly not to be read as 0.00), so splicing-dependent BS3 evidence is absent; this is not material for a single-nucleotide missense substitution.
• The InSiGHT MUTYH VCEP specification (doc 1742141534) is held for this case with an empty rule payload, so any VCEP-specific BS3 rules or assay-acceptability criteria could not be applied.
• PMID:25820570 is not a source_registry key in this case (full text not fetched by the case pipeline); its content is recorded in evidence_used prose only.

Sources

Reasoning

BP1 applies only to a missense variant in a gene for which disease-causing variation is primarily truncating, so that missense change is generally discounted. That premise does not hold for MUTYH: the recessive syndrome is caused by biallelic MUTYH inactivation through missense alleles, and PMID:18422726 documents an affected MUTYH-associated polyposis patient with a biallelic missense genotype (p.Gly272Glu plus p.Ala359Val) alongside a second patient compound-heterozygous for missense changes, PMID:21325953 describes the same biallelic recessive mechanism, and the case's own gene-context assessment records MUTYH loss of function as a supported germline disease mechanism for a polyposis syndrome rather than a truncating-only gene. Because missense variation is a recognised disease mechanism in this gene and the case variant is itself a missense change, BP1 cannot be used to argue benignity. BP1 is not met.

Rule

BP1 (ACMG/AMP 2015, supporting): missense variant in a gene for which primarily truncating variants are known to cause disease.

Evidence considered

• The governing InSiGHT MUTYH specification (CSPEC doc 1742141534, v1.0) publishes no approved criteria codes, so no gene-specific BP1 rule was available and the generic ACMG/AMP 2015 definition was applied.
• PMID:18422726 reports MUTYH-associated polyposis patients carrying biallelic missense changes (patient 1: p.Gly272Glu and p.Ala359Val; patient 2: p.Pro18Leu with p.Gly25Asp, p.Gln324His and c.1389G>C), so missense alleles are part of the disease-causing spectrum of this gene.
• PMID:21325953 states that MYH-associated polyposis is an autosomal recessive syndrome caused by biallelic mutations in the MYH gene with highly variable phenotype, establishing the recessive biallelic (not truncating-only) mechanism.
• pvs1_gene_context.json records MUTYH loss of function as a supported germline disease mechanism for the polyposis phenotype (pvs1_gene_gate 'eligible') with the caution that no official CSPEC/VCEP or local explicit framework was available; loss of function here is achieved by biallelic inactivation and is not restricted to truncating alleles.
• Because missense variation is a documented MUTYH disease mechanism, BP1's premise (primarily truncating disease, missense generally benign) is not satisfied for this gene.

Sources

Reasoning

BP2 has two limbs, and the limb structure has to be settled before any phase observation is read. The first limb - observed in trans with a pathogenic variant for a fully penetrant dominant gene or disorder - cannot apply to MUTYH, because MUTYH-associated polyposis is autosomal recessive: the gene-level ruleset of the governing InSiGHT MUTYH specification v1.0 (doc 1742141534) records 'Autosomal recessive inheritance' for familial adenomatous polyposis 2, and PMID:21325953 describes MAP as 'an autosomal recessive, polyposis syndrome caused by biallelic mutations in the MYH gene'. The specification was retrieved with an empty criteria payload, so it pre-assigns no BP2 code and supplies no criterion-level BP2 rule; the generic ACMG/AMP BP2 definition was therefore applied (PMID:25741868) and only the second limb - observed in cis with a pathogenic variant in any inheritance pattern - is evaluable for this variant. On that limb the cis phase is itself well documented: PMID:16929514 reported that 'the 2 missense mutations occurred on the same allele' and PMID:18811933 that cloned PCR products showed c.53C>T and c.74G>A 'on the same allele'. The decisive question is whether that cis partner is a pathogenic variant, and it is not. c.74G>A (p.Gly25Asp) is the only partner ever reported in cis with c.53C>T, and the two papers that established the phase both frame the combined c.53C>T/c.74G>A ('T/A') change as a polymorphism rather than a pathogenic allele - PMID:16929514 states 'Our study indicates that the variant haplotype T/A is a polymorphism in MYH gene', and PMID:18811933 found no homozygote and only a modest patient-vs-control excess (2.17% vs 0.44% allele frequency, P=0.020). PMID:18422726 never establishes G25D as pathogenic either, reporting it only as one of four heterozygous changes in a single MAP patient and expressly omitting functional testing of it. No source in this case places c.53C>T in cis with a variant classified pathogenic or likely pathogenic, so BP2 is not met.

Rule

Generic ACMG/AMP BP2 (PMID:25741868): observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder, or observed in cis with a pathogenic variant in any inheritance pattern. MUTYH is autosomal recessive, so only the cis limb is evaluable; no criterion-level BP2 rule is supplied by the governing MUTYH InSiGHT specification (empty criteria payload).

Evidence considered

• Limb applicability: MUTYH-associated polyposis is autosomal recessive (governing InSiGHT MUTYH specification v1.0, doc 1742141534, gene-level ruleset: 'modeOfInheritance: Autosomal recessive inheritance'; PMID:21325953: MAP is 'an autosomal recessive, polyposis syndrome caused by biallelic mutations in the MYH gene'), so BP2's 'in trans with a pathogenic variant for a fully penetrant dominant gene/disorder' limb is inapplicable and only the 'in cis with a pathogenic variant in any inheritance pattern' limb was evaluated.
• Framework limitation: the MUTYH specification was retrieved with an empty criteria payload (framework_complete=false, framework_mode=unstructured_ruleset), pre-assigning no BP2 code and no criterion-level BP2 rule; the generic ACMG/AMP BP2 definition was used (PMID:25741868).
• Cis-phase evidence: PMID:16929514 - 'These 2 missense mutations presented strong linkage disequilibrium. Sequencing of several cloned PCR products revealed that the 2 variations affected the same allele with retention of a wild-type allele (Fig. 1c).' PMID:18811933 - 'Further cloning-sequencing showed that the mutations occurred at the same allele (haplotype T/A variation)' and 'Sequencing of cloned PCR product revealed the 2 nucleotide substitutions C > T and G > A on the same allele.' c.53C>T is therefore in cis with c.74G>A (p.Gly25Asp).
• Pathogenicity of the cis partner is not established: PMID:16929514 states 'Our study indicates that the variant haplotype T/A is a polymorphism in MYH gene', with the haplotype enriched but not fixed in cases (allele frequency 4.95% vs 1.18%, P=0.021) and no biallelic carrier found; PMID:18811933 reports the heterozygous haplotype at 4.35% in sporadic CRC patients vs 0.87% in controls with 'no homozygote ... found in the present study' (allele frequency 2.17% vs 0.44%, P=0.020) and showed that the single p.Pro18Leu mutant protein behaved like wild type in its mitochondrial localisation assay, with only the double mutant showing an effect.
• PMID:18422726 lists c.74G>A (p.Gly25Asp) only as one of four heterozygous changes (P18L, G25D, Q324H, c.1389G>C) in a single polyposis patient, and explicitly did not functionally analyse G25D or P18L; the paper designates the co-inherited Q324H a polymorphism. No source in this case classifies c.74G>A, or any other reported cis neighbour of c.53C>T, as pathogenic or likely pathogenic, so the cis limb of BP2 is not satisfied.

Still missing

• No source establishes that c.74G>A (p.Gly25Asp) - the only variant reported in cis with c.53C>T - is pathogenic or likely pathogenic; a curated classification of that partner variant would be required to revisit the cis limb of BP2.
• No report places c.53C>T in cis with any other established pathogenic MUTYH allele (e.g. Y165C or G382D).

Sources

Reasoning

BP5 requires the variant to be observed in a case in which an alternate molecular basis for the disease is established. No such observation exists for MUTYH c.53C>T. The only adenomatous-polyposis patient documented to carry the variant (patient 2, PMID:18422726) was APC-negative and was classified by the authors as having MAP on the basis of biallelic MUTYH variants (p.Pro18Leu and p.Gly25Asp in cis, plus Q324H and 1389G>C) - i.e. the MUTYH genotype was accepted as the explanation, not displaced by another cause. Thirty-four of the 66 patients in that cohort carried both APC and MUTYH changes, but the single c.53C>T carrier was not among them. In the case-control cohorts (PMID:16929514, PMID:18811933) carriers are heterozygous for the c.53C>T/c.74G>A haplotype and no alternate genetic cause of their cancer is documented; notably no biallelic MYH mutation was found in any gastric cancer patient in PMID:16929514. The ClinVar submissions are conflicting ordinary laboratory assertions and provide no case-level statement of an alternate molecular basis. No numeric or likelihood-ratio-style BP5 threshold is defined in the captured MUTYH InSiGHT framework (criterion-level rules not retrieved), so the qualitative generic ACMG/AMP 2015 BP5 definition was applied.

Rule

Generic ACMG/AMP 2015 BP5 (PMID:25741868): variant observed in a case with an alternate molecular basis for the disease. MUTYH InSiGHT v1.0 gene-specific BP5 rule (including any likelihood-ratio-style value and its comparison direction) is not available in this case (cspec doc 1742141534, criterion-level rules not captured), so BP5 was applied qualitatively with no numeric threshold.

Evidence considered

• PMID:18422726 Table 5 / Results: patient 2 (APC-negative) carried P18L, G25D, Q324H and 1389G>C; 'Because these two patients had biallelic mutations of MUTYH, we regarded them as having MAP' - the authors assigned the polyposis to the MUTYH genotype, and the single c.53C>T carrier in this cohort was not one of the 34 patients carrying both APC and MUTYH changes.
• PMID:16929514: the T/A haplotype carriers are heterozygous and no biallelic MYH mutation was found in any of the 101 gastric cancer patients, so the authors did not assign the cancers to an alternate established germline cause either - there is no documented case in which c.53C>T was found alongside an established alternate molecular basis for the phenotype.
• PMID:18811933: c.53C>T carriers are heterozygous sporadic CRC cases; no alternate germline cause of CRC (e.g. MMR gene mutation) is documented for those individuals, and the study could not formally exclude selection bias.
• PMID:25741868 (generic ACMG/AMP 2015) defines BP5 as observation of the variant in a case with an alternate molecular basis for the disease; the retrieved evidence contains no such case.
• Numeric-threshold protocol: the captured MUTYH InSiGHT framework defines no numeric or likelihood-ratio-style BP5 value for this case, so no threshold direction (>= or <=) was applicable; BP5 was applied qualitatively. No operator was imported from another criterion in this group.

Still missing

• No proband-level data (co-occurring pathogenic variant in another gene, or a known alternate diagnosis) accompany this variant-only case.
• MUTYH InSiGHT v1.0 criterion-level BP5 rule not captured in this case; the four reviewed sources also do not document an alternate molecular basis in a c.53C>T carrier.

Sources

Reasoning

BP6 requires an exact-variant assertion from a reputable (expert-panel) source that the variant is benign, and per the rules for this group only an exact-variant ClinVar expert-panel classification may trigger it; aggregate ClinVar labels and ordinary laboratory submissions must never do so. The exact-variant ClinVar record (ClinVarID 41759, VCV000041759) does report Benign (9 laboratories) and Likely benign (3) plus likely benign (1), but all 20 submissions are from ordinary clinical laboratories, none is from an expert panel (expert_panel_submissions = 0; is_expert_panel = false for all 20), and the record-level review status is 'criteria provided, conflicting classifications' (1 star), with 2 laboratories submitting Uncertain significance. Because the benign signal rests entirely on non-expert laboratory submissions and an aggregate label, and no InSiGHT/VCEP expert-panel classification of c.53C>T exists in the retrieved evidence, BP6 is not triggered.

Rule

BP6 as specified for this group: trigger only on an exact-variant ClinVar expert-panel classification of Benign or Likely benign; ordinary laboratory submissions, aggregate ClinVar labels and non-expert assertions never trigger it. Generic BP6 origin: ACMG/AMP 2015 (PMID:25741868).

Evidence considered

• ClinVar (ClinVarID 41759 / VCV000041759, also 41763): aggregate submissions Benign 9, Likely benign 3, likely benign 1, Uncertain significance 2; api_classification 'Conflicting classifications of pathogenicity'; review status 'criteria provided, conflicting classifications' (1 star).
• Submission audit for this case: expert_panel_submissions = 0 and is_expert_panel = false for each of the 20 submissions; the ClinVar expert_panels field is empty, so there is no expert-panel Benign or Likely benign assertion for the exact variant.
• The InSiGHT Hereditary Colorectal Cancer/Polyposis MUTYH expert-panel specification (cspec doc 1742141534) is present as a framework in this case but provides no variant-level benign classification of c.53C>T and its criterion-level rules were not captured.
• Per the invocation rules for this group, the aggregate ClinVar label and the individual submitting laboratories' Benign/Likely benign classifications are explicitly excluded as BP6 triggers; no inference was drawn from any other source's final classification.

Still missing

• No ClinGen expert-panel (VCEP) classification of MUTYH c.53C>T is available in ClinVar or in the retrieved evidence; if one is published, BP6 must be re-evaluated.

Sources

Reasoning

The governing InSiGHT Hereditary Colorectal Cancer/Polyposis Expert Panel specification for MUTYH (doc 1742141534, v1.0) is present and applicable by gene and disease context, but its captured ruleset is an incomplete unstructured payload with no criterion-specific PS2 rule, so generic ACMG/AMP PS2 governs. Generic PS2 requires a variant observed to have arisen de novo in a patient, with maternity and paternity confirmed by identity testing, a family history consistent with de novo inheritance, and a phenotype matching the gene's disease association. For MUTYH c.53C>T (p.Pro18Leu) no proband-parent genotypes, parental identity testing, or de novo assertion exists in any consulted source. PMID:18422726 genotyped 66 Japanese adenomatous polyposis patients and found c.53C>T once as a heterozygous germline change, in a patient who carried it as part of a compound heterozygous genotype (P18L, G25D, 1389G>C, Q324H) and who explicitly had no family history of colorectal adenomatous polyposis; the paper reports no testing of that patient's parents. PMID:18811933 and PMID:16929514 report c.53C>T as a case-control haplotype (always in cis with c.74G>A) in unrelated Chinese colorectal and gastric cancer patients and healthy controls, with no parent-child transmission data. PMID:25741868 and PMID:24033266, which define the criterion and the requirement for biological-parent testing with exclusion of non-paternity and sample swap, were consulted and confirm the missing evidence type. PS2 therefore cannot be assessed rather than being met or refuted.

Rule

Generic ACMG/AMP PS2; no applicable criterion-specific MUTYH VCEP rule was available.

Evidence considered

• The MUTYH InSiGHT specification was retrieved (gene MUTYH, NM_001128425.2, autosomal recessive familial adenomatous polyposis 2, MONDO:0012041) but its rule payload is empty, so no criterion-specific PS2 rule exists and generic ACMG/AMP PS2 applies.
• PMID:25741868 defines PS2 as a de novo observation in a patient where both parental samples are shown by identity testing to be the biological parents, and notes PM6 applies when parentage is assumed but not confirmed.
• PMID:24033266 states that testing of biological parents and exclusion of non-paternity and sample swap are necessary to confirm a de novo variant.
• PMID:18422726 identified MUTYH c.53C>T (p.Pro18Leu) once as a heterozygous germ-line change among 66 Japanese patients with adenomatous polyposis; the carrier (patient 2) had a compound heterozygous MUTYH genotype and no family history of colorectal adenomatous polyposis, and no parental genotypes were reported.
• PMID:18811933 and PMID:16929514 genotyped unrelated patients and controls only, reporting c.53C>T in cis with c.74G>A as a haplotype; neither reports parental testing or a de novo determination.
• The ClinVar record for this variant (20 submissions, 19 submitters) provides classifications and cited PMIDs but no submission-level parental testing, trio, or de novo assertion for c.53C>T.

Still missing

• Documented proband genotype and phenotype for this variant
• Confirmed maternity and paternity by identity testing
• Testing results for both biological parents
• Explicit de novo interpretation with assay and laboratory details

Sources

Reasoning

The applicable MUTYH InSiGHT Expert Panel specification contains no criterion-specific PM6 rule, so generic ACMG/AMP PM6 applies. PM6 covers an assumed (presumed) de novo occurrence in a patient where maternity and paternity are not confirmed by identity testing; it still requires clinical documentation that the variant arose de novo, including parental genotype results in the absence of formal parentage confirmation. No source consulted for MUTYH c.53C>T (p.Pro18Leu) reports a presumed de novo interpretation, parental genotypes, or family genotyping. The variant is documented exclusively through population-scale and case-control germline screening: PMID:18811933 found the c.53C>T/c.74G>A haplotype in 6/138 (4.35%) sporadic Chinese colorectal cancer patients versus 3/343 (0.87%) healthy controls, and PMID:16929514 found it in 10/101 suspected hereditary gastric cancer patients versus 3/129 controls, both as heterozygous carriers without triplet or family transmission data. PMID:18422726 reports a single heterozygous carrier among 66 adenomatous polyposis patients, in a compound heterozygous MAP genotype with no family history and no relative testing. Because no presumed de novo claim exists to evaluate, PM6 is unassessable here.

Rule

Generic ACMG/AMP PM6; no applicable criterion-specific MUTYH VCEP rule was available.

Evidence considered

• The MUTYH InSiGHT specification has no criterion-specific PM6 rule payload; generic ACMG/AMP PM6 was applied.
• PMID:25741868 defines PM6 as the de novo criterion applied when identity of both parents is assumed but not confirmed by identity testing.
• PMID:24033266 requires testing of biological parents, with exclusion of non-paternity and sample swap, before de novo status can be concluded.
• No consulted source reports parental genotypes, a presumed de novo interpretation, or family testing for MUTYH c.53C>T: PMID:18422726 (1 heterozygous carrier of 66 polyposis patients, compound heterozygous MAP genotype, no family history), PMID:18811933 and PMID:16929514 (unrelated case-control carriers of the c.53C>T/c.74G>A haplotype).
• The ClinVar record contains no submission-level presumed de novo assertion for this variant.

Still missing

• Parental genotype results or other evidence establishing presumed de novo origin
• Proband and parental phenotype documentation
• Laboratory details (assay, confirmation method) supporting any de novo inference

Sources

Reasoning

The applicable MUTYH InSiGHT Expert Panel specification provides no criterion-specific PP1 rule, so generic ACMG/AMP PP1 applies. PP1 requires co-segregation of the variant with disease in multiple affected family members of a gene definitively known to cause the disease, assessed across informative meioses; PubMed-indexed guidance consulted here notes that roughly 10 informative segregations would approach a LOD score above 3.0, and that unaffected relatives should not contribute segregation information where penetrance, expressivity, and age of onset vary. No consulted source reports any pedigree, relative genotypes, or informative meioses for MUTYH c.53C>T (p.Pro18Leu). PMID:18422726 reports the only patient-level clinical description of a carrier: patient 2 carried c.53C>T within a compound heterozygous MUTYH genotype, had 147 colorectal adenomas and carcinoma diagnosed at age 47, and had no family history of colorectal adenomatous polyposis; the paper's only pedigree (Fig. 1a, consanguineous parents, affected brother) belongs to patient 1, who carried G272E/A359V and not c.53C>T. The remaining sources are case-control surveys of unrelated individuals (PMID:18811933, PMID:16929514) or functional localization work (PMID:18811933), none of which genotypes relatives. With zero informative meioses available, PP1 cannot be assessed.

Rule

Generic ACMG/AMP PP1; no applicable criterion-specific MUTYH VCEP rule was available.

Evidence considered

• The MUTYH InSiGHT specification has no criterion-specific PP1 rule payload; generic ACMG/AMP PP1 was applied.
• PMID:24033266 states that significant co-segregation of a variant with disease provides strong linkage evidence, that about 10 informative segregations would achieve a LOD score above 3.0, and that unaffected family members should not contribute where penetrance is incomplete and age of onset late.
• PMID:18422726 reports the only carrier of c.53C>T with clinical detail (patient 2, compound heterozygous P18L/G25D/1389G>C/Q324H, 147 adenomas, colorectal carcinoma at 47 years) and states that patient had no family history of colorectal adenomatous polyposis; the single pedigree shown (Fig. 1a) belongs to patient 1, who carried G272E/A359V instead.
• PMID:18811933 genotyped unrelated sporadic colorectal cancer patients and healthy controls (c.53C>T/c.74G>A haplotype in 6/138 patients vs 3/343 controls) and performed a cos-7 cell localization assay; no relatives were tested.
• PMID:16929514 recruited patients by gastric cancer family history but genotyped only unrelated patients and controls (10/101 vs 3/129); no relative genotypes, pedigrees, or meioses are reported.
• PMID:21325953 describes MAP as an autosomal recessive syndrome requiring biallelic MYH mutation, which is the inheritance frame in which any co-segregation would need to be evaluated; it reports no variant-level family data.
• The ClinVar record for this variant contains no submission reporting family-member genotyping or a segregation analysis.

Still missing

• Pedigree and phenotype data for relatives of a confirmed carrier
• Genotypes of affected and unaffected relatives
• Number of informative meioses
• Segregation analysis appropriate to autosomal recessive inheritance

Sources

Reasoning

The variant is a homozygous-observed allele in population data: gnomAD v2.1 reports 4 homozygotes (all within the East Asian subgroup, 4 of 260 East Asian alternate alleles) and gnomAD v4.1 reports 7 homozygotes (all East Asian, 7 of 509 alternate alleles), while the gnomAD v3.1 non-cancer genomes and all non-East-Asian subgroups show none. Those counts are the only BS2-relevant information available. The ACMG/AMP wording requires that the disease be fully penetrant at an early age and that the observation be in a well-documented healthy adult for a recessive (homozygous) finding (PMID:25741868), and the same guideline states that population databases cannot be assumed to contain only healthy individuals. MUTYH-associated polyposis is an autosomal recessive, adult-onset, variably penetrant cancer-predisposition disorder (PMID:21325953), so homozygous genotype counts in a reference cohort of unknown age and phenotype cannot establish the healthy-adult precondition, and no MUTYH-specific BS2 rule is published in the governing specification. BS2 is therefore not assessed rather than met or not met. If applied, the homozygote counts alone would be insufficient for a strong benign claim without individual-level phenotype.

Rule

{'source': 'generic_acmg_amp', 'requirement': 'observation in a healthy adult individual, with the disease fully penetrant at an early age, for a recessive (homozygous) finding', 'population_selection': 'phenotype-confirmed unaffected adults'}

Evidence considered

• gnomAD v2.1 reports 4 homozygotes (4 among 260 East Asian alternate alleles) and 0 homozygotes in every other subgroup; gnomAD v4.1 reports 7 homozygotes (7 among 509 East Asian alternate alleles).
• The gnomAD v3.1 non-cancer genome subset reports no homozygotes among 86 alternate alleles, so homozygote observations are confined to the exome-derived East Asian data.
• The population records provide cohort labels and genotype counts only; they carry no individual age, clinical examination, polyposis status or cancer history needed to document an unaffected healthy adult.
• The ACMG/AMP specification requires a fully penetrant, early-onset disease and a well-documented healthy adult for BS2, and cautions that population databases cannot be assumed to include only healthy individuals (PMID:25741868).
• PMID:21325953 establishes that MYH/MUTYH-associated polyposis is autosomal recessive with biallelic mutations required and a highly variable, adult-onset clinical picture, so homozygotes in a reference cohort may be undiagnosed rather than unaffected.
• The retrieved InSiGHT MUTYH specification contains no populated BS2 rule that would authorize treating an unphenotyped reference-population homozygote as a healthy adult.

Still missing

• Individual-level age, health status and MUTYH-associated phenotype (adenomatous polyposis/colorectal cancer) of the reported gnomAD homozygotes are unavailable, so the healthy-adult precondition of BS2 cannot be confirmed.
• No MUTYH-specific BS2 rule addressing reduced, adult-onset penetrance is published in the governing InSiGHT specification.

Sources

Reasoning

The applicable MUTYH InSiGHT Expert Panel specification contains no criterion-specific BS4 rule, so generic ACMG/AMP BS4 applies. BS4 requires lack of segregation of the variant with disease in affected family members, which presupposes an adequately ascertained family in which the variant has been genotyped in affected and unaffected relatives. No consulted source provides relative genotypes, affected-family cohorts, or a discordant genotype-phenotype observation for MUTYH c.53C>T (p.Pro18Leu). PMID:18422726, the only source with patient-level clinical detail for a c.53C>T carrier, describes an isolated patient with no family history of colorectal adenomatous polyposis and tested no relatives; PMID:18811933 and PMID:16929514 are case-control surveys of unrelated individuals in which the variant (as part of the c.53C>T/c.74G>A haplotype) was found in both patients and healthy controls, but co-occurrence in unrelated controls is not within-family non-segregation and cannot substitute for it. No pedigree-based non-segregation analysis, and therefore no evidence of the type BS4 requires, is available; a benign classification reported elsewhere was not used as a substitute for this criterion's own evidence.

Rule

Generic ACMG/AMP BS4; no applicable criterion-specific MUTYH VCEP rule was available.

Evidence considered

• The MUTYH InSiGHT specification has no criterion-specific BS4 rule payload; generic ACMG/AMP BS4 was applied.
• No consulted source genotypes affected relatives of a c.53C>T carrier or reports within-family non-segregation for this variant.
• PMID:18422726 reports a single compound heterozygous carrier of c.53C>T with no family history of colorectal adenomatous polyposis and no relative testing.
• PMID:18811933 and PMID:16929514 report the variant as a heterozygous haplotype in unrelated patients and healthy controls; observation in unrelated controls is not family-based non-segregation evidence.
• PMID:21325953 frames MAP as an autosomal recessive biallelic disorder, which is the context any segregation or non-segregation analysis would need to account for; it reports no variant-level family data.
• The ClinVar record for this variant provides classifications and PMID citations but no submission reporting non-segregation within a family.

Still missing

• Testing of affected family members for this variant
• Clinical ascertainment and phenotype data for relatives
• Documented discordant genotype-phenotype observations within a family
• Assessment of phenocopies, reduced penetrance, and autosomal recessive inheritance in the pedigree

Sources