POLE c.4744C>T, p.(Pro1582Ser)

Reasoning

BS1 is met at Supporting strength because the variant exceeds the generic non-VCEP BS1 operating threshold of 0.3% in the largest relevant population estimate. gnomAD v4.1 reports a South Asian AF of 0.33819% and grpmax FAF of 0.30709%; gnomAD v2.1 reports a South Asian AF of 0.39527% and grpmax FAF of 0.33803%. The frequency is below the 1% BA1 threshold, so it supports BS1 rather than BA1.

Rule

Generic non-VCEP BS1 operating threshold: apply Supporting when the maximum credible population frequency is greater than 0.3%; do not upgrade to BA1 unless frequency exceeds 1%.

Evidence considered

• gnomAD v4.1 reports the highest population AF in South Asian individuals as 0.33819% (308/91,074 alleles) and grpmax FAF as 0.30709%.
• gnomAD v2.1 reports the highest population AF in South Asian individuals as 0.39527% (121/30,612 alleles) and grpmax FAF as 0.33803%.
• gnomAD-Canada independently reports AF 0.02172% overall (4/18,418 alleles), with South Asian AF 0.22026% (3/1,362 alleles), providing additional population evidence but not exceeding 0.3%.
• The observed elevation is ancestry-specific and should not be interpreted as a disease-prevalence estimate; nevertheless, it exceeds the defined BS1 frequency threshold.

Sources

Reasoning

NM_006231.4:c.4744C>T predicts the exonic missense substitution NP_006222.2:p.(Pro1582Ser), not a nonsense, frameshift, or canonical ±1/2 splice-site variant. It is therefore outside the generic PVS1 null-variant entry categories; NMD and exon criticality are not relevant to this consequence.

Rule

Generic ACMG/AMP PVS1 with the ClinGen SVI PVS1 decision framework: apply only to a predicted loss-of-function variant class in a gene with an established loss-of-function disease mechanism; this missense substitution is not a PVS1 variant class.

Evidence considered

• The variant-level PVS1 assessment identifies NM_006231.4:c.4744C>T as a missense consequence, with protein change NP_006222.2:p.(Pro1582Ser), no canonical splice-consensus involvement, and placement in the 'other' bucket outside the default PVS1 null-variant classes.
• No official POLE CSPEC/VCEP PVS1 specification was retrieved. The local León-Castillo POLE framework explicitly customizes only PM1, PS4, PP3, and BP4, so generic ClinGen SVI PVS1 guidance is the applicable fallback.

Sources

Reasoning

PS3 requires well-established functional studies demonstrating a deleterious effect (ACMG/AMP 2015, PMID:25741868). No functional assay evidence exists for POLE p.(Pro1582Ser) in this case bundle: OncoKB lists the variant as 'Unknown Oncogenic Effect' with no variant-specific reviewed functional evidence (not reviewed); COSMIC reports a single somatic occurrence (COSV108850878, n=1) which is an observational recurrence count, not a functional assay; SpliceAI (max delta 0.00-0.004) is an in silico splice prediction, not a functional study; the custom POLE framework (Leon-Castillo et al. 2020) does not customize PS3 and does not include p.Pro1582Ser among the functionally characterized exonuclease-domain hotspots (P286R, V411L, S297F, A456P, S459F); and neither fetched publication (PMID:25741868, PMID:25394175) mentions the variant. With no validated assay data for this exact variant, PS3 cannot be met.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PS3 - Well-established functional studies show a deleterious effect. Not customized by the Leon-Castillo 2020 custom POLE framework.

Evidence considered

• Reviewed generic ACMG/AMP 2015 PS3 definition (well-established functional studies show a deleterious effect) from PMID:25741868; the paper contains no variant-specific functional data for c.4744C>T/p.(Pro1582Ser) (paper_extracts confirms variant not mentioned).
• OncoKB (source registry key 'oncokb') reports no variant-specific reviewed functional evidence for POLE P1582S: classification 'Unknown Oncogenic Effect', is_variant_specific=false, not_reviewed=true, no citations. Not usable as PS3 evidence.
• COSMIC (source registry key 'cosmic') reports a single somatic occurrence (COSV108850878, n=1); a somatic recurrence count is not a well-established functional study and was not used for PS3.
• SpliceAI (source registry key 'spliceai') predicts no significant splice impact (max delta score 0.00-0.004); this is an in silico prediction, not a functional assay, and was not used for PS3.
• The local custom POLE framework (Leon-Castillo et al. 2020 package, vcep_materials) customizes only PM1/PS4/PP3/BP4 and does not define a PS3 rule; it contains no functional assay for p.Pro1582Ser (variant absent from the discussed hotspot set and supplementary tables).
• PMID:28492532 (Sherloc) was available only as metadata (full text not fetched); it is a classification-methods reference with no variant-specific functional content and was not used.

Still missing

• No functional assay evidence for p.(Pro1582Ser) is available (e.g., POLE polymerase-epsilon activity/proofreading or replication-fidelity assays, yeast or cell-based complementation, MMR-proficiency assays). A validated assay showing a deleterious effect would be required to meet PS3.

Sources

Reasoning

The applicable local POLE framework permits PS4_Supporting only for exact variants recurrent in both COSMIC and TCGA endometrial-carcinoma cohorts, with a combined count of at least 10, and belonging to its established/pathogenic set. p.(Pro1582Ser) is not among those enumerated variants, and COSMIC records only one somatic occurrence; no case-control enrichment is available.

Rule

León-Castillo et al. 2020 custom POLE framework: PS4_Supporting requires exact-variant recurrence in both COSMIC and TCGA endometrial-carcinoma cohorts, combined count >=10, and membership in the established/pathogenic set.

Evidence considered

• The local POLE framework specifies PS4_Supporting only for p.P286R, p.V411L, p.A456P, and p.S297F when recurrent in both COSMIC and TCGA endometrial-carcinoma cohorts with combined count at least 10.
• The queried p.(Pro1582Ser) is not in that exact-variant list; available COSMIC evidence records one somatic occurrence, not germline case-control enrichment.

Still missing

• No germline case-control dataset or appropriately ascertained POLE p.(Pro1582Ser) case series with controls was provided.

Sources

Reasoning

No CSPEC exists for POLE, so the governing framework is the local custom Leon-Castillo et al. 2020 POLE framework (framework_mode 'custom_gene_framework'; official_clingen false). Its PM1 rules are exact-substitution only: PM1_Strong for the five established exonuclease-domain hotspot EDMs (p.P286R, p.V411L, p.S297F, p.A456P, p.S459F), PM1_Moderate for p.F367S, p.L424I, p.M295R, p.P436R, p.M444K, p.D368Y, and PM1_Supporting for p.A465V, p.L424V, p.T278M, p.A428T - all located within the N-terminal exonuclease domain. p.(Pro1582Ser) is not any of these exact substitutions. vcep_materials.json domain_tables is empty ({}) for POLE, so there is no authoritative VCEP-approved domain-table entry whose residue range includes 1582. Direct inspection of PATH-250-323.pdf and Supplementary Tables S1-S3 found zero occurrences of residue 1582, c.4744C>T, or p.P1582S; the paper's hotspot/variant sets span the exonuclease domain (~aa 268-471) and it states that >95% of POLE mutations outside the exonuclease domain are classified non-pathogenic by POLE-score. CancerHotspots reports the variant is not in a statistically significant hotspot, and COSMIC shows a single somatic occurrence (n=1), i.e., no recurrence. PM1 is therefore not met under the governing framework or generic fallback.

Rule

Custom POLE framework PM1 (Leon-Castillo 2020): PM1_Strong for the five established exonuclease-domain hotspot EDMs (P286R, V411L, S297F, A456P, S459F); PM1_Moderate for exact recurrent non-hotspot EDM variants (F367S, L424I, M295R, P436R, M444K, D368Y); PM1_Supporting for POLE-score-3 tier variants (A465V, L424V, T278M, A428T). Exact substitutions only. Generic PM1 fallback: located in a mutational hot spot and/or critical and well-established functional domain without benign variation.

Evidence considered

• The governing custom POLE framework (final_classification_framework.json / case_summary cspec block) specifies PM1 only for exact substitutions: PM1_Strong {P286R, V411L, S297F, A456P, S459F}, PM1_Moderate {F367S, L424I, M295R, P436R, M444K, D368Y}, PM1_Supporting {A465V, L424V, T278M, A428T}, with the explicit note 'Use this only for these exact substitutions, not all substitutions at the same codon'. p.(Pro1582Ser) is not among them.
• vcep_materials.json domain_tables is empty ({}), i.e., no authoritative domain-table entry for POLE was extracted from the VCEP materials; per the PM1 instruction, the residue was checked against every available entry (none) and against the framework's exact-substitution lists.
• Direct text extraction of PATH-250-323.pdf (13 pages) found no occurrence of '1582', 'P1582', 'Pro1582', or '4744'. The paper's five established hotspot EDMs (P286R, V411L, S297F, A456P, S459F) and its tiered variant sets all fall within the N-terminal exonuclease domain, and the paper states 'nearly all (>95%) POLE mutations outside the exonuclease domain are classified as non-pathogenic by POLE-score'. Residue 1582 lies far outside this region (protein length 2286 aa).
• Direct inspection of PATH-250-323-s002.xlsx (Supplementary Table S1, 101 data rows of recurrent POLE variants with COSMIC/TCGA EC counts) found no row for p.P1582S/c.4744C>T.
• CancerHotspots reports 'This variant does not lie in a statistically significant hotspot' (no result for POLE P1582), and COSMIC lists a single somatic occurrence (COSV108850878, n=1), providing no recurrence-based hotspot support.

Still missing

• vcep_materials.json domain_tables is empty for POLE; the not-met conclusion rests on the custom framework's exact-substitution enumeration and direct review of the source PDF/supplementary tables rather than an official approved domain table.

Sources

Reasoning

PM2 is not met. Although the overall gnomAD v4.1 AF is low (0.01983%) and the gnomAD-Canada AF is 0.02172%, the maximum population frequency is above the generic PM2 rarity threshold of 0.1%: South Asian AF is 0.33819% in gnomAD v4.1, 0.39527% in gnomAD v2.1, and 0.22026% in gnomAD-Canada. The presence of four and two homozygotes in the respective gnomAD datasets further argues against treating the allele as sufficiently rare without a disease-specific frequency model.

Rule

Generic non-VCEP PM2 operating threshold: apply only when the maximum relevant population frequency is below 0.1%; ancestry-specific frequencies and homozygote observations must be considered rather than pooled AF alone.

Evidence considered

• gnomAD v4.1 reports overall AF 0.01983% but South Asian AF 0.33819% and grpmax FAF 0.30709%, with four homozygotes.
• gnomAD v2.1 reports overall AF 0.04945% but South Asian AF 0.39527% and grpmax FAF 0.33803%, with two homozygotes.
• gnomAD-Canada reports overall AF 0.02172% and South Asian AF 0.22026%, with no homozygotes.
• For PM2, the maximum credible ancestry-specific frequency is more informative than the pooled overall AF because the pooled value can conceal an ancestry-enriched allele and create a threshold artifact.

Still missing

• No POLE-specific maximum credible allele frequency based on disease prevalence, penetrance, and allelic contribution was retrieved for this variant or disorder.

Sources

Reasoning

p.(Pro1582Ser) replaces one amino acid without an insertion, deletion, stop-loss, or other protein-length alteration. PM4 is not met.

Rule

Generic ACMG/AMP PM4: protein-length-changing variants, including in-frame insertions/deletions in a non-repeat region or stop-loss variants, may qualify. A missense substitution does not meet this criterion.

Evidence considered

• The normalized variant consequence is the single-residue substitution NP_006222.2:p.(Pro1582Ser), rather than an in-frame insertion/deletion or stop-loss variant.
• No POLE-specific PM4 specification was retrieved; the local POLE framework does not customize PM4, so generic ACMG/AMP PM4 semantics apply.

Sources

Reasoning

PP3 is not met. The applicable local POLE rule requires the exact missense variant to be present in Supplementary Table S2 or S3 with REVEL classified as likely disease causing and no more than one benign in-silico result; p.(Pro1582Ser) is absent from both tables. Under generic calibrated computational evidence, the available REVEL score (0.564) is intermediate and does not reach the ClinGen SVI PP3 supporting threshold (REVEL >=0.773; PMID:36413997). SpliceAI shows no predicted splice impact (maximum delta 0.004), so it provides no splice-based PP3 support.

Rule

Local POLE PP3 rule: exact variant in Supplementary Table S2 or S3, REVEL class likely disease causing, and <=1 benign in-silico result; otherwise generic calibrated computational-evidence fallback.

Evidence considered

• The local POLE framework specifies PP3_Supporting only for an exact variant represented in Supplementary Table S2 or S3 with REVEL class 'likely disease causing' and <=1 benign in-silico result. Direct review of both tables found no p.(Pro1582Ser) or c.4744C>T entry.
• REVEL is 0.564. The ClinGen SVI calibrated thresholds reported by Pejaver et al. (PMID:36413997) are >=0.773 for PP3 supporting and <=0.290 for BP4 supporting; this score is intermediate.
• SpliceAI delta scores are DS_AG 0.004, DS_AL 0.004, DS_DG 0.000, and DS_DL 0.000 (maximum 0.004), predicting no significant splice effect.

Sources

Reasoning

ClinVar contains no exact-variant expert-panel pathogenic or likely pathogenic assertion. Its exact-variant submissions are ordinary single-submitter laboratory assertions and include benign, likely benign, and uncertain-significance classifications, which cannot trigger PP5.

Rule

Apply PP5 only for an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification.

Evidence considered

• ClinVar Variation ID 473690 reports zero expert-panel submissions; its listed assertions are single-submitter laboratory submissions with benign, likely benign, and uncertain-significance classifications.

Still missing

• An exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification would be required to apply PP5 under this review.

Sources

Reasoning

BA1 is not met. The highest observed population frequency is below the generic BA1 threshold of 1%: gnomAD v4.1 South Asian AF is 0.33819% (grpmax FAF 0.30709%), and gnomAD v2.1 South Asian AF is 0.39527% (grpmax FAF 0.33803%).

Rule

Generic BA1: apply only when the variant allele frequency exceeds 1% in a population dataset; no applicable POLE population-specific modification was retrieved.

Evidence considered

• gnomAD v4.1 reports 320/1,613,542 alleles overall (AF 0.01983%), with 308/91,074 South Asian alleles (AF 0.33819%), four homozygotes, and grpmax FAF 0.30709%.
• gnomAD v2.1 reports 124/250,770 alleles overall (AF 0.04945%), with 121/30,612 South Asian alleles (AF 0.39527%), two homozygotes, and grpmax FAF 0.33803%.
• The maximum observed frequency is well below 1%, so the BA1 threshold is not reached.

Sources

Reasoning

BS3 requires well-established functional studies demonstrating no damaging effect (ACMG/AMP 2015, PMID:25741868). No functional assay evidence of normal function for POLE p.(Pro1582Ser) exists in this case bundle. gnomAD population frequencies (v2.1 AF 0.049%, v4.1 AF 0.020%, grpmax FAF ~0.003; highest in South Asians ~0.34-0.40% with homozygotes) and the ClinVar 'Benign/Likely benign' single-submitter classifications are not functional studies and were not used as BS3 evidence; SpliceAI (max delta ~0.00) is an in silico splice prediction, not a functional assay; no paper in the bundle describes any assay on this variant. With no validated assay data for this exact variant, BS3 cannot be met.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): BS3 - Well-established functional studies show no damaging effect. Not customized by the Leon-Castillo 2020 custom POLE framework.

Evidence considered

• Reviewed generic ACMG/AMP 2015 BS3 definition (well-established functional studies show no damaging effect) from PMID:25741868; the paper contains no variant-specific functional data for c.4744C>T/p.(Pro1582Ser) (paper_extracts confirms variant not mentioned).
• gnomAD v2.1 and v4.1 (source registry keys 'gnomad_v2', 'gnomad_v4') show the variant is present at low overall frequency (0.049% and 0.020%) with highest South Asian frequency (~0.34-0.40%, homozygotes present); population frequency data address BA1/BS1, not BS3, and were not used as functional evidence.
• OncoKB (source registry key 'oncokb') provides no variant-specific functional characterization (Unknown Oncogenic Effect, not reviewed); no benign functional assay data available from this source.
• SpliceAI (source registry key 'spliceai') predicts no significant splice impact (max delta score 0.00-0.004); in silico prediction is not a well-established functional study and was not used for BS3.
• ClinVar (source registry key 'clinvar') classifications (Likely benign x2, Benign x1, VUS x1, single-submitter) were reviewed but not used for BS3, since final classifications are not functional studies and must not be used to infer criteria.
• The local custom POLE framework (Leon-Castillo et al. 2020 package) does not define a BS3 rule and contains no functional assay for p.Pro1582Ser.

Still missing

• No functional assay evidence demonstrating no damaging effect for p.(Pro1582Ser) is available (e.g., POLE polymerase-epsilon activity/proofreading-fidelity assays or other validated functional studies). A validated assay showing normal function would be required to meet BS3; population frequency and laboratory classifications do not substitute for BS3.

Sources

Reasoning

BP1 applies only to genes for which primarily truncating variants are known to cause disease. POLE does not meet this description: both loss-of-function and missense mechanisms cause POLE-associated disease. Germline missense variants in the exonuclease domain are an established, well-characterized pathogenic mechanism (polymerase proofreading-associated polyposis / colorectal cancer susceptibility 12, and somatic POLE-ultramutated endometrial carcinoma), which is the entire basis of the governing custom framework; ClinVar conditions for POLE include 'Colorectal cancer, susceptibility to, 12'. Because missense variants are a bona fide disease mechanism for POLE, BP1 (which would argue missense is NOT disease-causing because only truncations are) is not met.

Rule

BP1: Missense variant in a gene for which primarily truncating variants are known to cause disease (ACMG/AMP 2015, PMID 25741868).

Evidence considered

• Generic ACMG/AMP 2015 BP1 definition (PMID 25741868): 'Missense variant in a gene for which primarily truncating variants are known to cause disease.'
• The governing custom POLE framework (Leon-Castillo 2020) documents established pathogenic exonuclease-domain missense hotspots (P286R, V411L, S297F, A456P, S459F) and tiered missense variants - demonstrating that missense, not primarily truncating variation, is a central POLE disease mechanism.
• ClinVar conditions associated with POLE include 'Colorectal cancer, susceptibility to, 12', the phenotype caused by exonuclease-domain missense variants, confirming missense-mediated pathogenicity for this gene.
• POLE loss-of-function also causes germline disease (biallelic, e.g., FILM/FILS-related phenotypes per pvs1_gene_context supporting literature), so neither mechanism is 'primarily truncating'.

Sources

Reasoning

p.(Pro1582Ser) is a missense substitution, not an in-frame insertion or deletion. Therefore, repeat-region context cannot support BP3 for this variant.

Rule

Generic ACMG/AMP BP3: apply only to an in-frame insertion/deletion in a repetitive region without known functional significance. This missense substitution has no in-frame length change and does not meet BP3.

Evidence considered

• The variant-level consequence assessment identifies an exonic missense substitution with no insertion or deletion and no protein-length change.
• No POLE-specific BP3 specification was retrieved; the local POLE framework does not customize BP3, so generic ACMG/AMP BP3 semantics apply.

Sources

Reasoning

BP4 is not met. The applicable local POLE rule requires the exact missense variant to be present in Supplementary Table S2 or S3 with REVEL classified as likely benign and at least four benign in-silico results; p.(Pro1582Ser) is absent from both tables. The generic calibrated REVEL score (0.564) is intermediate rather than at or below the ClinGen SVI BP4 supporting threshold (REVEL <=0.290; PMID:36413997). Although SpliceAI predicts no splice impact, that single benign splice prediction does not establish benign computational evidence for this missense substitution.

Rule

Local POLE BP4 rule: exact variant in Supplementary Table S2 or S3, REVEL class likely benign, and >=4 benign in-silico results; otherwise generic calibrated computational-evidence fallback.

Evidence considered

• The local POLE framework specifies BP4_Supporting only for an exact variant represented in Supplementary Table S2 or S3 with REVEL class 'Likely benign' or 'Likely-benign' and >=4 benign in-silico results. Direct review of both tables found no p.(Pro1582Ser) or c.4744C>T entry.
• REVEL is 0.564. The ClinGen SVI calibrated thresholds reported by Pejaver et al. (PMID:36413997) are >=0.773 for PP3 supporting and <=0.290 for BP4 supporting; this score is intermediate.
• SpliceAI maximum delta is 0.004 and predicts no significant splice impact; this addresses splice disruption only and is insufficient to apply BP4 for the protein-level missense consequence without calibrated benign missense-predictor support.

Sources

Reasoning

ClinVar contains no exact-variant expert-panel benign or likely benign assertion. Ordinary single-submitter benign and likely benign laboratory submissions cannot trigger BP6.

Rule

Apply BP6 only for an exact-variant ClinVar expert-panel Benign or Likely benign classification.

Evidence considered

• ClinVar Variation ID 473690 reports zero expert-panel submissions; benign and likely benign assertions are from ordinary single-submitter laboratories.

Still missing

• An exact-variant ClinVar expert-panel Benign or Likely benign classification would be required to apply BP6 under this review.

Sources

Reasoning

PS1 requires a previously established pathogenic variant producing the same amino acid change (p.Pro1582Ser) via a different nucleotide change. No such same-amino-acid comparator was identified in the available sources: the pm5_candidates artifact reports pm5_mode 'unknown' with automatic same-residue candidate harvesting skipped (0 candidates collected), so no systematic search for other variants at codon 1582 was completed; the ClinVar record (variation 473690) contains only this exact c.4744C>T change, submitted as Likely benign (2 labs), Benign (1 lab), and VUS (1 lab), with no pathogenic same-amino-acid variant. Because the comparator evidence was never gathered, PS1 is not_assessed rather than affirmatively not met.

Rule

PS1: Same amino acid change as a previously established pathogenic variant regardless of nucleotide change (ACMG/AMP 2015, PMID 25741868).

Evidence considered

• pm5_candidates.json reports pm5_mode 'unknown' and 'Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting', with total_candidates_seen 0 and same_residue_candidates 0 - therefore no same-amino-acid PS1 comparator candidates were collected for evaluation.
• ClinVar variation 473690 (exact match to c.4744C>T) contains submissions of Likely benign (2 clinical laboratories), Benign (1), and Uncertain significance (1); no submission or criterion lead identifies a pathogenic variant at the same amino acid position.
• No publication or database in the bundle reports p.(Pro1582Ser) arising from a different nucleotide substitution.

Still missing

• No systematic same-residue (codon 1582) variant search was performed: pm5 candidate harvesting was skipped (pm5_mode 'unknown', eligible_for_classic_pm5_search false).
• No established pathogenic variant producing p.(Pro1582Ser) via a different nucleotide change was found in the case evidence bundle.

Sources

Reasoning

No de novo occurrence evidence is available for NM_006231.4:c.4744C>T (p.Pro1582Ser). The case bundle contains no proband phenotype, no parental testing results, and no paternity/maternity-confirmed de novo observation. The ClinVar record (VCV000473690) submissions carry no de novo criterion signals (evidence_signals criteria_found are empty except for an in-silico PP3 signal in one submission), and neither fetched full-text paper (PMID:25741868, PMID:25394175) mentions the variant. Without any parental-testing data, PS2 (de novo with confirmed parentage) cannot be applied. Applicable rule: generic ACMG/AMP 2015 (Richards et al., PMID:25741868), because no POLE-specific CSPEC/VCEP exists and the local custom POLE framework (León-Castillo et al. 2020) specifies rules only for PM1, PS4, PP3, and BP4.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PS2 - de novo (paternity and maternity confirmed), Strong. No POLE-specific specification exists for PS2 (local custom framework covers only PM1, PS4, PP3, BP4).

Evidence considered

• Generic ACMG/AMP 2015 rule definition consulted (Richards et al., PMID:25741868): PS2 applies as strong evidence for de novo occurrence with confirmed paternity and maternity; no gene-specific specification covers PS2, so generic rule applies.
• ClinVar VCV000473690 reviewed: 5 germline submissions (Labcorp Benign, KCCC Likely benign, Ambry Likely benign, GeneDx VUS, Sinai Health System Likely benign, research); none report a de novo observation or parental testing, and no submission lists a PS2/PM6 criterion signal.
• Full-text extraction of PMID:25394175 reviewed: variant c.4744C>T / p.(P1582S) is not mentioned; no de novo or family data present.
• Full-text extraction of PMID:25741868 reviewed: variant not mentioned; consulted only as the generic ACMG/AMP rule source.

Still missing

• No proband or parental testing data (paternity/maternity confirmation) in the case bundle; requires family trio data or a documented confirmed de novo observation to assess PS2.

Sources

Reasoning

No affected-proband or family-level evidence was provided showing NM_006231.4:c.4744C>T (p.Pro1582Ser) in trans with a pathogenic or likely pathogenic POLE variant. Phase and inheritance information are unavailable, so PM3 cannot be applied.

Rule

Generic PM3: apply only when the variant is observed in trans with a pathogenic or likely pathogenic variant in an affected individual with a disorder known to be caused by biallelic variants; phase should be established or adequately supported.

Evidence considered

• The consolidated case-evidence bundle contains no case report, pedigree, segregation result, parental testing, haplotype result, or affected-proband observation establishing this variant in trans with a pathogenic/likely pathogenic variant.
• The POLE custom framework is limited to PM1, PS4, PP3, and BP4; it provides no gene-specific PM3 specification, so generic PM3 requirements were considered.

Still missing

• Affected-proband observations involving this variant are not documented.
• No pathogenic or likely pathogenic variant in the same proband is documented.
• No parental testing, haplotyping, or other phase evidence establishes the variants in trans.
• The relevant disease inheritance context for applying a recessive allelic criterion is not documented for this case.

Reasoning

PM5 requires a different missense change at the same residue (p.Pro1582) that has been determined pathogenic. No such comparator was identified: the pm5_candidates artifact explicitly states classic same-residue PM5 semantics could not be confirmed safely and automatic candidate harvesting was skipped (total_candidates_seen 0, same_residue_candidates 0), and the ClinVar record for this variant contains only the single c.4744C>T change (LB/B/VUS submissions) with no other missense change at codon 1582. Without a collected same-residue comparator set, PM5 cannot be evaluated and is returned not_assessed; the artifact's own instruction is to review collected comparators before applying PM5, and none were collected.

Rule

PM5: Novel missense change at an amino acid residue where a different missense change determined to be pathogenic has been seen before (ACMG/AMP 2015, PMID 25741868).

Evidence considered

• pm5_candidates.json: pm5_mode 'unknown', eligible_for_classic_pm5_search false, recommendation level 'not_applicable' with summary 'Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting'; candidate_summary shows 0 candidates seen and 0 same-residue candidates.
• ClinVar variation 473690 lists only NM_006231.4:c.4744C>T (p.Pro1582Ser) across all five submissions (Benign, Likely benign x3, VUS); no alternate missense change at residue 1582 with a pathogenic classification appears in the case evidence bundle.

Still missing

• No same-residue comparator variants at p.Pro1582 were collected: pm5 candidate harvesting was skipped (pm5_mode 'unknown', 'Unable to confirm classic same-residue PM5 semantics safely').
• A dedicated search for other missense changes at codon 1582 (e.g., p.Pro1582Thr, p.Pro1582Leu) with pathogenic classification was not performed.

Sources

Reasoning

No assumed-de-novo evidence is available. There is no proband record, no parental genotyping, and no report in ClinVar submissions or the fetched literature of the variant arising de novo without parental confirmation. PM6 (assumed de novo without paternity/maternity confirmation) therefore cannot be applied. Rule basis: generic ACMG/AMP 2015 (PMID:25741868), as no POLE-specific framework rule exists for PM6.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PM6 - assumed de novo, but without confirmation of paternity and maternity, Moderate.

Evidence considered

• Generic ACMG/AMP 2015 rule definition consulted (Richards et al., PMID:25741868): PM6 applies as moderate evidence for assumed de novo without confirmation of paternity and maternity; no gene-specific specification covers PM6.
• ClinVar VCV000473690 submissions reviewed: none report an unconfirmed de novo occurrence; no PM6 criterion signal present in any submission.
• Full-text extraction of PMID:25394175 reviewed: variant not mentioned; no de novo or parental-testing data.
• Full-text extraction of PMID:25741868 reviewed: variant not mentioned; consulted only as the generic ACMG/AMP rule source.

Still missing

• No parental testing or de novo observation of any kind in the case bundle; family/trio data would be required to assess PM6.

Sources

Reasoning

No segregation data is available. The case bundle contains no affected-family-member genotype information, no pedigrees, no meioses, and no co-segregation observations for c.4744C>T. ClinVar submissions report no segregation evidence, and neither fetched paper mentions the variant. Although POLE is a gene with established germline disease associations (e.g., colorectal cancer susceptibility 12; gene-level context in gene_context.json and OncoKB), PP1 also requires observed co-segregation in multiple affected family members, which is entirely absent. Rule basis: generic ACMG/AMP 2015 (PMID:25741868); the local custom POLE framework does not specify PP1.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PP1 - co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease, Supporting.

Evidence considered

• Generic ACMG/AMP 2015 rule definition consulted (Richards et al., PMID:25741868): PP1 applies as supporting evidence for co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease; no gene-specific specification covers PP1.
• ClinVar VCV000473690 submissions reviewed: no submission reports segregation in affected relatives; no PP1 criterion signal present.
• Full-text extraction of PMID:25394175 reviewed: variant not mentioned; no segregation or family data.
• Full-text extraction of PMID:25741868 reviewed: variant not mentioned; consulted only as the generic ACMG/AMP rule source.

Still missing

• No segregation analysis in affected family members (no pedigrees, meioses, or affected-relative genotypes) in the case bundle; family segregation studies are required to assess PP1.

Sources

Reasoning

The custom POLE framework does not customize PP2, so the generic ACMG/AMP 2015 PP2 applies: a 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. The second component is supported for POLE (exonuclease-domain missense variants are an established disease mechanism - polymerase proofreading-associated polyposis/CRC susceptibility 12 and somatic EC ultramutation, per the governing framework and ClinVar conditions). However, the first component - a low rate of benign missense variation in POLE - cannot be evaluated from the case evidence bundle: no gene-level missense constraint metric (e.g., gnomAD missense Z-score or observed/expected ratio) is present in the source registry or any evidence file. Because one required element of the conjunctive criterion is unassessable, PP2 is returned not_assessed.

Rule

PP2: 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 (ACMG/AMP 2015, PMID 25741868).

Evidence considered

• Generic ACMG/AMP 2015 PP2 definition (PMID 25741868): '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.'
• Missense is an established POLE disease mechanism: the governing custom framework is built entirely around exonuclease-domain missense pathogenicity (hotspot EDMs), and ClinVar conditions for this gene include 'Colorectal cancer, susceptibility to, 12', consistent with a missense-mediated disease mechanism.
• The evidence bundle contains no gene-level constraint metric for POLE (no gnomAD Z-score/o-e, no equivalent source in source_registry.json), so the low-benign-missense-rate conjunct cannot be verified.

Still missing

• No gene-level missense constraint data for POLE (e.g., gnomAD missense Z-score or o/e ratio) is available in the evidence bundle to establish a low rate of benign missense variation.
• PP2 is not customized by the governing POLE framework, so the generic ACMG/AMP definition applies and both conjuncts must be satisfied.

Sources

Reasoning

No phenotype, clinical features, or phenotype-to-genotype specificity evidence for the individual carrying p.(Pro1582Ser) was provided. The available POLE framework has no PP4-specific rule, so generic ACMG/AMP requires phenotype specificity evidence that is absent here.

Rule

Generic ACMG/AMP PP4: the patient's phenotype or family history is highly specific for a disease with a single genetic etiology.

Evidence considered

• The local POLE framework customizes PS4 but does not supply a PP4 rule; no individual-level phenotype data are present in the consolidated evidence.

Still missing

• Provide the proband phenotype, indication for testing, relevant tumor features if applicable, and evidence that the phenotype is highly specific for POLE-related disease.

Sources

Reasoning

BS2 cannot be established from the available population data alone. gnomAD contains four homozygotes in v4.1 and two homozygotes in v2.1, but the case bundle does not provide phenotype, age, ascertainment, or penetrance information for those individuals. In addition, POLE-related disease mechanisms include both heterozygous cancer-predisposition contexts and recessive biallelic disease, so homozygote observation is not by itself sufficient to classify this allele as benign for the relevant disorder.

Rule

Generic BS2 requires the variant to be observed in healthy individuals in a manner inconsistent with the relevant disease model; homozygote counts without verified phenotype, age, ascertainment, and disease-model context are insufficient.

Evidence considered

• gnomAD v4.1 reports four homozygotes overall, all four within the South Asian population estimate.
• gnomAD v2.1 reports two homozygotes overall, both within the South Asian population estimate.
• The bundle supplies no clinical phenotype or unaffected-status verification for these homozygotes.

Still missing

• No individual-level phenotype or age-at-evaluation data are available for the gnomAD homozygotes.
• The relevant POLE disease model and expected penetrance for this variant are not sufficiently defined to interpret homozygote observations as evidence against pathogenicity.

Sources

Reasoning

No non-segregation (lack-of-segregation) evidence is available. BS4 requires documented lack of segregation in affected family members (e.g., an affected relative who does not carry the variant). No affected relatives, pedigree data, or genotype results appear anywhere in the case bundle, and none of the ClinVar submissions or fetched papers report segregation testing. Population frequency data (gnomAD v2.1/v4.1, gnomAD-Canada) is not BS4 evidence and is handled under population-frequency criteria by other groups. Rule basis: generic ACMG/AMP 2015 (PMID:25741868); the local custom POLE framework does not specify BS4.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): BS4 - lack of segregation in affected family members, Supporting (benign).

Evidence considered

• Generic ACMG/AMP 2015 rule definition consulted (Richards et al., PMID:25741868): BS4 applies as supporting benign evidence for lack of segregation in affected family members; no gene-specific specification covers BS4.
• ClinVar VCV000473690 submissions reviewed: none report segregation testing results in affected relatives; no BS4 criterion signal present.
• Full-text extraction of PMID:25394175 reviewed: variant not mentioned; no segregation or non-segregation data.
• Full-text extraction of PMID:25741868 reviewed: variant not mentioned; consulted only as the generic ACMG/AMP rule source.

Still missing

• No segregation testing in affected family members (no affected relatives genotyped, no non-segregation observations) in the case bundle; family studies are required to assess BS4.

Sources

Reasoning

No affected-proband, family, or phase evidence was provided showing NM_006231.4:c.4744C>T (p.Pro1582Ser) in trans with a pathogenic variant for a dominant disorder or in cis with a pathogenic variant for a recessive disorder. BP2 cannot be applied without a qualifying allelic observation and established phase.

Rule

Generic BP2: apply only when the variant is observed in trans with a pathogenic variant for a disorder inherited dominantly, or in cis with a pathogenic variant for a disorder inherited recessively; the relevant phase and inheritance context must be established.

Evidence considered

• The consolidated case-evidence bundle contains no case report, pedigree, segregation result, parental testing, haplotype result, or other phase evidence relevant to BP2.
• The POLE custom framework is limited to PM1, PS4, PP3, and BP4; it provides no gene-specific BP2 specification, so generic BP2 requirements were considered.

Still missing

• No qualifying affected-proband or family observation involving this variant and another pathogenic variant is documented.
• No pathogenic variant with which this variant is in cis or trans is identified.
• No parental testing, haplotyping, or other phase evidence is available.
• The inheritance model relevant to a qualifying BP2 observation is not documented for this case.

Reasoning

No individual-level case report, phenotype, or independent molecular diagnosis was provided to establish that p.(Pro1582Ser) was found in a case with an alternate molecular basis for disease.

Rule

Generic ACMG/AMP BP5: variant found in a case with an alternate molecular basis for disease.

Evidence considered

• The consolidated evidence contains no case-level alternate molecular diagnosis for a carrier of p.(Pro1582Ser).

Still missing

• Provide case-level evidence of the phenotype and a confirmed alternative molecular diagnosis that fully explains it.