POLD1 c.1275C>T, p.(Ala425=)

Reasoning

The allele is far more common than expected for a rare POLD1-associated dominant cancer-predisposition disorder: gnomAD v4.1 reports 1.9228% in Ashkenazi Jewish and 1.3787% in African/African American individuals, with concordant v2.1 frequencies. This meets generic BS1; it does not meet BA1 because the observed AF is below 5%.

Rule

Generic ACMG/AMP BS1: allele frequency greater than expected for the disorder; satisfied by reproducible 1.3-1.9% group frequencies for this rare dominant POLD1-associated condition.

Evidence considered

• No POLD1 ClinGen VCEP specification or local gene-specific framework was retrieved; generic ACMG/AMP frequency guidance therefore applies.
• gnomAD v4.1 has 1,949 variant alleles among 1,613,918 alleles (overall AF 0.1208%); the highest frequencies are Ashkenazi Jewish AF 1.9228% (569/29,592; 7 homozygotes) and African/African American AF 1.3787% (1,034/74,998; 5 homozygotes).
• gnomAD v2.1 corroborates enrichment: Ashkenazi Jewish AF 1.7685% (183/10,348; 2 homozygotes) and African/African American AF 1.3676% (341/24,934).
• gnomAD-Canada also detects the allele at overall AF 0.1739% (32/18,404) and African AF 1.5717% (16/1,018).
• POLD1 germline exonuclease-domain variants are associated with polyposis and predisposition to colorectal, endometrial, and possibly brain cancer; a persistent 1.3-1.9% frequency across ancestry groups is incompatible with a pathogenic contribution to this rare dominant syndrome. The population frequency, not the ClinVar classification, is the basis for BS1.

Sources

Reasoning

SpliceAI max delta score of 0.035 for NM_002691.4:c.1275C>T is well below the ~0.1 threshold typically used to indicate no predicted splicing consequence (per the SpliceAI delta-score framework of Jaganathan et al. 2019, which defines 0.2/0.5/0.8 as high-recall/recommended/high-precision thresholds for a splice-altering call; a score of 0.035 falls far below even the most permissive threshold). No VCEP-specific POLD1 lookup table was found to override this generic assessment. BP4 (supporting benign, in silico prediction of no deleterious effect on splicing) is met on the splice-impact path. No missense in silico predictor evidence contributes here since the variant is synonymous.

Evidence considered

• SpliceAI max delta score = 0.035 (DS_AG=0.035, DS_AL=0.001, DS_DG=0.001, DS_DL=0.005) for NM_002691.4:c.1275C>T, well below thresholds associated with splice-altering variants.
• SpliceAI delta-score interpretation thresholds (0.2 high recall / 0.5 recommended / 0.8 high precision) are from Jaganathan et al. 2019 (Cell), PMID 30661751; the observed score of 0.035 is well below even the lowest of these thresholds, supporting no predicted splicing impact.

Sources

Reasoning

NM_002691.4:c.1275C>T is a synonymous variant (p.(Ala425=), most_severe_consequence=synonymous_variant per gnomAD VEP annotation) with SpliceAI predicting no impact on splicing (max delta = 0.035). This satisfies the splicing-prediction component of BP7 (synonymous variant not predicted to create/disrupt a splice site). The bundle does not contain a nucleotide-level conservation score (e.g., PhyloP/GERP) for this exact position, so the second, conservation-based component of the classic BP7 definition could not be directly confirmed from the evidence provided; this is flagged as a gap requiring human review even though the splicing evidence alone strongly supports application of BP7.

Evidence considered

• gnomAD VEP annotation: most_severe_consequence = synonymous_variant for this variant, confirming p.(Ala425=) is a silent change.
• SpliceAI max delta score = 0.035, indicating no predicted disruption of splice donor/acceptor sites or creation of a new splice site.

Still missing

• No PhyloP/GERP nucleotide conservation score for c.1275 was available in the prefetched evidence to confirm the 'not highly conserved' component of BP7.

Sources

Reasoning

PM2 is not applicable because the variant is recurrent in multiple population datasets: gnomAD v4.1 overall AF is 0.1208%, with 1.9228% in Ashkenazi Jewish and 1.3787% in African/African American individuals. The corresponding group maximum filtering allele frequency is 1.3089%, so the apparent population signal is not a singleton or sampling artifact.

Rule

Generic ACMG/AMP PM2: absent or extremely rare in population databases; not met.

Evidence considered

• No POLD1 ClinGen VCEP specification or local gene-specific framework was retrieved; generic ACMG/AMP frequency guidance therefore applies.
• gnomAD v4.1: overall AF 0.1208% (1,949/1,613,918); Ashkenazi Jewish AF 1.9228% (569/29,592) with grpmax FAF 1.3089%; African/African American AF 1.3787% (1,034/74,998).
• gnomAD v2.1: overall AF 0.2098% (593/282,654), Ashkenazi Jewish AF 1.7685%, African/African American AF 1.3676%, and grpmax FAF 1.2136%.
• gnomAD-Canada: overall AF 0.1739% (32/18,404) and African AF 1.5717% (16/1,018).
• Generic ACMG/AMP PM2 requires absence or extreme rarity in population databases (Richards et al., 2015; PMID:25741868); this recurrent, high-frequency allele does not satisfy that requirement.

Sources

Reasoning

NM_002691.4:c.1275C>T is a synonymous variant (p.(Ala425=)) in POLD1 with no VCEP/gene-specific splicing lookup available (framework_mode=generic_acmg, no cspec found). SpliceAI predicts no significant splice impact (max delta score = 0.035; DS_AG=0.035, DS_AL=0.001, DS_DG=0.001, DS_DL=0.005), well below the commonly used 0.2 high-recall threshold for a splice-altering prediction. There is no applicable missense-path predictor (REVEL/BayesDel scores are both null/not found in prefetch, and this is not a missense change, so REVEL/aGVGD/BayesDel are not applicable). No in silico evidence supports a deleterious effect, so PP3 is not met.

Evidence considered

• SpliceAI scores for NM_002691.4:c.1275C>T: DS_AG=0.035, DS_AL=0.001, DS_DG=0.001, DS_DL=0.005; max delta = 0.035, indicating no predicted splicing impact.
• Variant is synonymous (p.(Ala425=)), so missense computational predictors (REVEL, BayesDel, aGVGD) are not applicable to PP3 for this substitution.
• REVEL and BayesDel scores were both queried in prefetch and returned null/not found for this variant, so no missense-path evidence exists regardless of applicability.

Sources

Reasoning

ClinVar reports no expert-panel submissions for this exact variant. The observed benign/likely benign laboratory submissions and aggregate classification are not eligible for PP5.

Evidence considered

• The exact ClinVar record has zero expert-panel submissions; only an exact-variant Pathogenic or Likely pathogenic ClinVar expert-panel classification could support PP5.

Sources

Reasoning

The highest population allele frequency is 1.9228% in the Ashkenazi Jewish gnomAD v4.1 group (569/29,592 alleles; 7 homozygotes); the African/African American group is also 1.3787% (1,034/74,998; 5 homozygotes). These frequencies are below the generic ACMG/AMP BA1 stand-alone threshold of 5%.

Rule

Generic ACMG/AMP BA1: population allele frequency >5%.

Evidence considered

• No POLD1 ClinGen VCEP specification or local gene-specific framework was retrieved; generic ACMG/AMP frequency guidance therefore applies.
• gnomAD v4.1: overall AF 0.1208% (1,949/1,613,918), Ashkenazi Jewish AF 1.9228% (569/29,592), and African/African American AF 1.3787% (1,034/74,998).
• gnomAD v2.1 independently showed Ashkenazi Jewish AF 1.7685% (183/10,348) and African/African American AF 1.3676% (341/24,934).
• Generic ACMG/AMP BA1 is stand-alone benign evidence only when allele frequency exceeds 5% (Richards et al., 2015; PMID:25741868).

Sources

Reasoning

ClinVar reports no expert-panel submissions for this exact variant. The aggregate benign/likely benign result and ordinary laboratory submissions are not eligible for BP6.

Evidence considered

• The exact ClinVar record has zero expert-panel submissions; only an exact-variant Benign or Likely benign ClinVar expert-panel classification could support BP6.

Sources

Reasoning

No case-level de novo observation is available. The bundle contains no proband result, parental genotypes, parentage confirmation, or phenotype information for an affected individual.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PS2 requires a confirmed de novo variant with both biological parents tested and appropriate clinical context.

Evidence considered

• The case evidence bundle supplies population, ClinVar, prediction, and literature-triage data but no family-based testing or phenotype data for this variant.
• Generic ACMG/AMP de novo guidance requires negative testing in both parents, confirmed biological parentage, and a phenotype consistent with the gene-disease association for PS2.

Still missing

• Variant result in an affected proband
• Genotype results for both biological parents
• Parentage confirmation and phenotype consistency assessment

Sources

Reasoning

No ClinGen VCEP specification exists for POLD1 (cspec.found=false; framework_mode=generic_acmg), so generic ACMG applies. The variant NM_002691.4:c.1275C>T is synonymous (NP_002682.2:p.(Ala425=)). The case bundle (prefetch, evidence, literature_pass, vcep_materials) contains no functional/biochemical assay data (no polymerase fidelity, proofreading, splicing minigene/RT-PCR, or cell-based complementation study) referencing this variant. The three validated ClinVar-linked PMIDs (25741868, 26467025, 25394175) are methodology/guideline papers (ACMG-AMP interpretation standards, Sherloc scoring system reference note, and a cancer predisposition referral guideline), not primary functional studies of this variant, and none mention it.

Still missing

• No published or unpublished functional assay data (polymerase activity, splicing assay, minigene, or complementation study) identified for this variant in the case evidence bundle.

Reasoning

No applicable POLD1 VCEP or gene-specific specification was available, so generic ACMG/AMP was used. The evidence bundle contains no exact-variant case series or case-control study with affected and control counts; PS4 cannot be evaluated.

Evidence considered

• The VCEP materials report no ClinGen specification or local gene-specific framework for POLD1, requiring generic ACMG/AMP fallback.
• The literature bundle identifies no exact-variant case-control or phenotype-enrichment publication.

Still missing

• Exact-variant case-control or statistically evaluated case-series data for NM_002691.4:c.1275C>T.

Reasoning

No affected individual carrying NM_002691.4(POLD1):c.1275C>T with a pathogenic or likely pathogenic POLD1 variant on the opposite allele is documented in the consolidated case-evidence bundle. PM3 cannot be evaluated without an affected-proband observation and phase/allelic evidence in a relevant recessive disease context.

Rule

Generic ACMG/AMP PM3: apply only when the variant is observed in trans with a pathogenic/likely pathogenic variant in an affected individual with a recessive disorder.

Evidence considered

• No POLD1 VCEP/ClinGen gene-specific specification was retrieved; generic ACMG/AMP assessment is therefore the applicable fallback.
• The bundle contains no allelic case observation, second POLD1 variant, phase result, or affected-proband evidence for NM_002691.4:c.1275C>T.

Still missing

• Affected-proband observations for this variant in a POLD1-associated recessive condition.
• Identity and classification evidence for any second POLD1 variant and confirmed trans phase (or sufficient phase evidence).
• Disease-specific evidence establishing that the relevant POLD1 condition is recessive for PM3 application.

Reasoning

No de novo observation with parental testing is available; therefore PM6 cannot be evaluated.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PM6 may apply to an apparent de novo variant when parentage is not confirmed; no such observation is supplied.

Evidence considered

• The case evidence bundle contains no reported trio, parental testing, or family-history data for this variant.
• Generic ACMG/AMP guidance reserves PM6 for an apparent de novo event when biological parentage is assumed rather than confirmed.

Still missing

• Variant result in an affected proband
• Genotype results for both parents showing absence of the variant
• Phenotype and family-history information

Sources

Reasoning

No informative affected relatives, genotypes, phenotypes, or meioses are reported, so segregation cannot be assessed.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PP1 requires co-segregation with disease in multiple affected family members; no segregation dataset is available.

Evidence considered

• The case evidence bundle contains no pedigree, relative-level genotype, phenotype, or meiosis data.

Still missing

• Pedigree with affected and unaffected relatives
• Variant genotypes and phenotype status for informative relatives
• Number of informative meioses and assessment of phenocopies or reduced penetrance

Sources

Reasoning

No affected individual phenotype, phenotype specificity analysis, or exact-variant phenotypic segregation data were provided. PP4 cannot be evaluated from condition labels attached to ClinVar laboratory submissions.

Evidence considered

• The available ClinVar condition labels are submission-associated labels and do not provide a patient-level phenotype or establish phenotype specificity for this exact variant.

Still missing

• Detailed phenotype for an individual carrying NM_002691.4:c.1275C>T and evidence that the phenotype is highly specific for a POLD1-related disorder.

Reasoning

Population databases report homozygotes, but no source provides phenotyped healthy adults or establishes that the relevant POLD1 disorders are fully penetrant and expected at an age at which unaffected status is informative. Therefore homozygote counts alone cannot establish BS2.

Rule

Generic ACMG/AMP BS2 requires well-phenotyped healthy adult observation for a fully penetrant early-onset disorder; population-database homozygote counts alone are insufficient.

Evidence considered

• gnomAD v4.1 reports 12 total homozygotes (7 Ashkenazi Jewish and 5 African/African American); gnomAD v2.1 reports 2 total homozygotes; gnomAD-Canada reports 1 homozygote.
• These datasets supply allele and genotype counts but no individual-level phenotype, disease-history, or age-at-evaluation evidence for the homozygotes.
• Generic ACMG/AMP BS2 requires observation in healthy adults for a disorder expected to be fully penetrant at an early age; that prerequisite is not demonstrated here (Richards et al., 2015; PMID:25741868).

Still missing

• Phenotyped adult carriers/homozygotes with age and relevant cancer/polyposis history, together with a condition-specific penetrance and age-of-onset assessment, are required for BS2.

Sources

Reasoning

No ClinGen VCEP specification exists for POLD1, so generic ACMG applies. No well-established functional assay data (with appropriate controls) demonstrating no damaging effect on POLD1 protein function or splicing was found for this synonymous variant in the case evidence bundle or the associated publications.

Still missing

• No functional assay evidence (with normal/wild-type controls) identified showing no damaging effect for this variant.

Reasoning

No informative non-segregation evidence is reported. The bundle does not document an affected non-carrier or an unaffected carrier evaluated in a setting where penetrance, age, and phenotype specificity permit interpretation.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): BS4 requires lack of segregation with disease; no informative discordant family observations are available.

Evidence considered

• The case evidence bundle contains no family segregation or non-segregation observations.

Still missing

• Pedigree with phenotype and genotype data for informative relatives
• Evidence of variant non-segregation in affected relatives or other robust discordance
• Penetrance, age-at-evaluation, and phenotype-specificity context

Sources

Reasoning

No individual-level co-occurrence of NM_002691.4(POLD1):c.1275C>T with a pathogenic or likely pathogenic POLD1 variant is documented in the consolidated case-evidence bundle, and no cis/trans phase result is available. BP2 cannot be assessed from aggregate variant classifications or population data alone.

Rule

Generic ACMG/AMP BP2: apply only to a documented observation with a pathogenic/likely pathogenic variant in cis, or in trans for a relevant dominant disorder.

Evidence considered

• No POLD1 VCEP/ClinGen gene-specific specification was retrieved; generic ACMG/AMP assessment is therefore the applicable fallback.
• The bundle contains no co-occurring POLD1 pathogenic/likely pathogenic variant and no phase information for NM_002691.4:c.1275C>T.

Still missing

• An individual-level report of co-occurrence with a pathogenic or likely pathogenic POLD1 variant.
• Confirmed cis phase, or confirmed trans phase in a disorder for which POLD1 is established to act dominantly in the relevant phenotype.

Reasoning

No patient-level phenotype or molecular diagnostic workup documenting an alternate cause was provided. BP5 cannot be evaluated.

Still missing

• A documented molecular diagnosis that fully explains the affected individual's phenotype, with adequate clinical context to assess BP5.