POLE c.3332G>A, p.(Arg1111Gln)

Reasoning

Extremely low population frequency: gnomAD v4.1 total AF 3.53e-05 (0.0035%, 57/1,614,138 alleles) and gnomAD v2.1 total AF 2.39e-05 (0.0024%, 6/251,478 alleles); highest subpopulation frequency is European (Finnish) 1.09e-04 (0.0109%) in v4.1. All observed frequencies are well below the project non-VCEP PM2 threshold of AF <0.1%. Zero homozygotes in both datasets, with grpmax FAF 7.01e-06 (v2.1) and 3.15e-05 (v4.1), providing no evidence of a common benign allele or population artifact. Applied at Supporting strength per the ClinGen SVI PM2_Supporting recommendation (Tavtigian et al. 2018, PMID 30311386).

Rule

PM2: absent from controls or at extremely low frequency in population databases (Richards et al. 2015, PMID 25741868); project non-VCEP gnomAD threshold AF <0.1%; applied at Supporting strength per ClinGen SVI PM2_Supporting (Tavtigian et al. 2018, PMID 30311386).

Evidence considered

• gnomAD v2.1 (exome): 6/251,478 alleles, AF 2.39e-05 (0.0024%), 0 homozygotes, grpmax FAF 7.01e-06 (source_registry key gnomad_v2)
• gnomAD v4.1 (joint): 57/1,614,138 alleles, AF 3.53e-05 (0.0035%), 0 homozygotes, joint grpmax FAF 3.15e-05; highest subpopulation Finnish AF 1.09e-04 (0.0109%) (source_registry key gnomad_v4)
• PM2 criterion definition from Richards et al. 2015 ACMG/AMP (PMID 25741868), applied via generic_acmg_combination_rules; project non-VCEP gnomAD PM2 threshold AF <0.1% (max observed 0.0109%, met)
• PM2_Supporting strength per ClinGen SVI Bayesian calibration, Tavtigian et al. 2018, Genet Med 20(9):1054-1060 (PMID 30311386)

Still missing

• gnomAD-Canada (HostSeq) data not retrieved; not limiting for PM2 because gnomAD v2.1/v4.1 frequencies are already well below threshold and mutually consistent.

Sources

Reasoning

PS1 requires the same amino-acid change (p.Arg1111Gln) as a previously established pathogenic variant, arising from a different nucleotide change. No source in the case bundle establishes p.Arg1111Gln as pathogenic: ClinVar VCV000405761 classifies this exact variant as Uncertain significance (3 clinical laboratories; no expert-panel submission; review_stars 2), the pm5_candidates artifact collected zero candidate comparators, and the governing POLE framework package (León-Castillo et al. 2020) plus its three supplementary tables contain no mention of residue 1111 or c.3332. The only full-text publication in the bundle (PMID:25394175, ACMG/NSGC referral-indications guideline) contains no mention of POLE or this variant. Somatic COSMIC recurrence (COSV57681369, n=4) was reviewed as context but does not establish an 'established pathogenic' germline amino-acid change. PS1 is therefore not met.

Rule

PS1: Same amino acid change as a previously established pathogenic variant regardless of nucleotide change (ACMG/AMP 2015, PMID 25741868 via generic_acmg_combination_rules); no POLE-specific PS1 customization exists in the governing custom framework.

Evidence considered

• ClinVar VCV000405761: NM_006231.4(POLE):c.3332G>A (p.Arg1111Gln) classified Uncertain significance by 3 clinical laboratories (GeneDx, Ambry, Labcorp); no expert-panel submission and no pathogenic classification of this amino-acid change.
• pm5_candidates.json: total_candidates_seen = 0; no same-change or same-residue pathogenic comparator candidates collected.
• PATH-250-323.pdf (León-Castillo et al. 2020): full-text search shows no occurrence of '1111', 'R1111', or '3332'; no established pathogenic variant at this residue.
• PMID:25394175 (ACMG/NSGC referral indications guideline): full text contains 0 occurrences of POLE/1111/3332/Arg1111; no variant-specific evidence.

Still missing

• No dedicated search for alternate nucleotide changes at codon 1111 that would produce p.Arg1111Gln (same amino-acid change) was available; no such variant appeared in any reviewed source.
• No ClinVar expert-panel or multi-laboratory pathogenic classification exists for p.Arg1111Gln that could anchor PS1.

Sources

Reasoning

Governing framework is the local custom POLE framework (León-Castillo et al. 2020, J Pathol 250:323-335; official_clingen=false, precedence_rank=2). Its PS4 rule is an exact-variant somatic-recurrence adaptation capped at Supporting: it triggers only for the precise POLE missense variants recurrent in BOTH the COSMIC and TCGA endometrial carcinoma (EC) cohorts in Supplementary Table S1 with a combined EC count >=10 and belonging to the established/pathogenic set (p.P286R, p.V411L, p.A456P, p.S297F). Direct parsing of PATH-250-323-s002.xlsx (Table S1, 102 rows) shows no row for p.R1111Q / c.3332G>A, and this variant is not in the established pathogenic set; the framework's PS4 rule therefore does not trigger. The case bundle's COSMIC entry (COSV57681369, somatic count n=4) is a general somatic-cancer recurrence count that does not meet the dual-cohort EC >=10 requirement, and the framework explicitly caps PS4 as a somatic-recurrence adaptation rather than germline case-control evidence. No germline case-control enrichment study for this exact variant exists in the bundle, so generic ACMG/AMP PS4 cannot be met either.

Rule

Custom POLE framework (León-Castillo et al. 2020, local custom framework v1) PS4_Supporting: exact missense variant recurrent in both COSMIC and TCGA EC cohorts in Table S1, combined EC count >=10, and in the established pathogenic set. Generic ACMG/AMP PS4 (statistically significant case-control enrichment) applies outside the framework.

Evidence considered

• vcep_path_250_323: León-Castillo et al. 2020 custom POLE framework PS4 rule (exact-variant COSMIC+TCGA EC recurrence, combined EC count >=10, established pathogenic set P286R/V411L/A456P/S297F; capped at Supporting)
• vcep_path_250_323_s002: Supplementary Table S1 parsed directly; no row for p.R1111Q or c.3332G>A
• Case-bundle COSMIC lookup (COSV57681369, somatic count 4) noted as context but does not satisfy the framework rule or generic case-control PS4

Still missing

• No germline case-control enrichment study for c.3332G>A (p.Arg1111Gln) is present in the evidence bundle; only somatic COSMIC recurrence (n=4) is available, which is insufficient for generic PS4.
• Variant is absent from León-Castillo Table S1 (COSMIC+TCGA EC recurrence), so the custom POLE PS4 exact-variant rule cannot trigger.

Sources

Reasoning

The governing framework is the local custom POLE framework derived from León-Castillo et al. 2020 ('leon-castillo-2020-custom-framework-v1'; no official ClinGen CSPEC was available). Its PM1 rules are exact-substitution rules: PM1_Strong applies only to p.P286R, p.V411L, p.S297F, p.A456P, p.S459F; PM1_Moderate only to p.F367S, p.L424I, p.M295R, p.P436R, p.M444K, p.D368Y; PM1_Supporting only to p.A465V, p.L424V, p.T278M, p.A428T. The framework explicitly states these apply only to the exact substitutions, not all substitutions at the same codon. p.Arg1111Gln is not among any PM1 tier. Consistently, residue 1111 is outside the exonuclease-domain hotspot region occupied by the five established hotspots, is absent from the recurrent-variant list in Supplementary Table S1 and from all three supplementary in-silico tables (searched for 1111/c.3332 with no hits), is not flagged as a statistically significant hotspot by CancerHotspots (context only, not a registry key), and OncoKB reports Unknown Oncogenic Effect with no variant-specific functional annotation. PM1 is therefore not met.

Rule

PM1 (custom POLE framework): Apply PM1_Strong to the five established exonuclease-domain hotspots p.P286R, p.V411L, p.S297F, p.A456P, p.S459F; PM1_Moderate to p.F367S, p.L424I, p.M295R, p.P436R, p.M444K, p.D368Y; PM1_Supporting to p.A465V, p.L424V, p.T278M, p.A428T - exact substitutions only, not all substitutions at the same codon.

Evidence considered

• Custom framework PM1 rules (case_summary cspec payload): PM1 tiers enumerate exact substitutions only (Strong: P286R, V411L, S297F, A456P, S459F; Moderate: F367S, L424I, M295R, P436R, M444K, D368Y; Supporting: A465V, L424V, T278M, A428T); p.R1111Q appears in none.
• PATH-250-323-s002.xlsx (Supplementary Table S1): recurrent POLE variants with COSMIC/TCGA EC counts; contains no codon-1111 or c.3332 entry.
• ClinVar VCV000405761: variant is Uncertain significance, providing no hotspot/domain pathogenic support.
• OncoKB POLE/R1111Q: 'Unknown Oncogenic Effect', not reviewed, no variant-specific functional evidence.

Still missing

• No official ClinGen POLE VCEP/CSPEC PM1 rule set was available; PM1 was adjudicated under the exact-substitution custom framework, which does not award PM1 to any residue-level (codon-level) domain argument.

Sources

Reasoning

PM4 (ACMG/AMP 2015, Richards et al., PMID 25741868) is defined as protein length change resulting from an in-frame deletion/insertion in a non-repeat region or from a stop-loss variant. The variant is a single-nucleotide substitution c.3332G>A producing a single amino-acid substitution p.Arg1111Gln: Mutalyzer translation of NM_006231.4(NP_006222.2) shows the reference and predicted protein sequences identical in length with only residue 1111 changed (Arg->Gln), i.e. no in-frame insertion/deletion and no stop-loss. There is therefore no protein length change of any kind, and PM4 is not met. SpliceAI max delta 0.02 additionally excludes any splicing-mediated in-frame exon skipping that could theoretically create a length change. Residue context (exon 27, mid-protein, outside the exonuclease domain) does not alter this determination because the criterion is strictly about length alteration.

Evidence considered

• ACMG/AMP 2015 PM4 definition (Richards et al., PMID 25741868): PM4 = protein length changes as a result of in-frame deletions/insertions in a nonrepeat region or stop-loss variants; c.3332G>A is a missense SNV (p.Arg1111Gln) causing no length change, so PM4 is not met.
• Mutalyzer transcript/protein normalization (prefetch.json): predicted translation of NM_006231.4(NP_006222.2):p.(Arg1111Gln) is identical in length to the reference with only residue 1111 differing (Arg->Gln); no insertion, deletion, or stop-loss.
• SpliceAI (Jaganathan et al. 2019, Cell; PMID 30661751): max delta 0.02, excluding splice-mediated in-frame exon skipping as a route to protein length change.

Sources

Reasoning

PM5 requires a novel missense change at an amino-acid residue where a DIFFERENT missense change has previously been determined to be pathogenic. No such comparator exists in any reviewed source: the pm5_candidates artifact (source_registry key pm5_candidates) reports found=false with 0 same-residue candidates and mode 'unknown' (automatic harvesting was skipped because classic same-residue PM5 semantics could not be confirmed), ClinVar submissions for this locus show only c.3332G>A (p.Arg1111Gln) as VUS, and the governing framework package - main PDF plus all three supplementary tables - was searched for codon-1111/c.3332 entries with no hits. Because no established pathogenic (or likely pathogenic) missense at residue 1111 with a different amino-acid change was identified, PM5 is not met on the available evidence. A human review is flagged because the automated same-residue comparator harvest was skipped, so the absence of codon-1111 comparators rests on the sources reviewed here rather than a systematic ClinVar codon-level query.

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 via generic_acmg_combination_rules); no POLE-specific PM5 customization exists in the governing custom framework.

Evidence considered

• pm5_candidates.json: found=false, candidate_summary.total_candidates_seen = 0, same_residue_candidates = 0; recommendation level 'not_applicable' with instruction not to apply PM5 from this artifact alone.
• ClinVar VCV000405761: all 3 submissions concern the identical variant p.Arg1111Gln (VUS); no other missense change at residue 1111 is reported in the record.
• PATH-250-323.pdf and PATH-250-323-s002/s003/s004.xlsx: string searches for '1111', 'R1111', and '3332' returned no occurrences; no same-residue variants are listed in the framework package.

Still missing

• Automatic PM5 comparator harvesting was skipped (pm5_candidates mode 'unknown'; reason: 'Unable to confirm classic same-residue PM5 semantics safely'); no systematic ClinVar query for other variants at codon 1111 (e.g., p.R1111C/H/L/P/W) was performed.
• If a ClinVar codon-level search later identifies a pathogenic/likely-pathogenic missense at residue 1111, PM5 would need re-adjudication at the appropriate strength.

Sources

Reasoning

Governing framework is the local custom POLE framework (León-Castillo et al. 2020, J Pathol 250:323-335), which pre-assigns PP3_Supporting only when the exact missense variant appears in Supplementary Table S2 or S3 with a REVEL class of 'likely disease causing' (minor spelling variants included) and <=1 benign in silico result. The queried variant p.R1111Q (c.3332G>A) was searched in both lookup spreadsheets (PATH-250-323-s003.xlsx = Table S2, PATH-250-323-s004.xlsx = Table S3) and is absent from both, so the custom PP3 rule does not trigger and the framework's own fallback to the generic in-silico workflow applies. Under the generic fallback, missense variants are evaluated with REVEL only: REVEL score for this exact variant is 0.36, which is below the ClinGen SVI REVEL PP3 supporting threshold of >=0.932 (Pejaver et al. 2022, PMID 36413997), so PP3 is not met (gray zone). BayesDel score (-0.0290682) is present for the variant but has no verified published threshold and is treated as not usable for PP3. SpliceAI max delta 0.02 predicts no splice impact; per the generic rule a missense variant's SpliceAI score does not clear its REVEL-assessed protein-level effect and the splice path is not applied to a missense variant. Single-submitter (1-star) ClinVar entries for this variant include a PP3 signal, but these are not expert-panel evaluations and are not adopted as independent evidence.

Rule

Custom POLE framework (León-Castillo et al. 2020) PP3 rule: PP3_Supporting if exact missense variant is in Supplementary Table S2/S3 with REVEL class 'likely disease causing' and <=1 benign in-silico result. Not triggered (variant absent from both tables); generic fallback for missense variants uses REVEL only, PP3_Supporting if REVEL >= 0.932 (ClinGen SVI REVEL calibration, Pejaver et al. 2022, PMID 36413997).

Evidence considered

• REVEL score 0.36 for the exact variant (source_registry: revel); below the ClinGen SVI PP3 supporting threshold REVEL >= 0.932 (Pejaver et al. 2022, PMID 36413997) -> PP3 not met.
• Custom POLE framework PP3 rule requires the exact missense variant in Supplementary Table S2 or S3 with REVEL class 'likely disease causing' and <=1 benign in-silico result; p.R1111Q is absent from both tables (vcep_path_250_323_s003, vcep_path_250_323_s004) -> rule not triggered, generic fallback applies.
• BayesDel score -0.0290682 (source_registry: bayesdel) not used for PP3: no verified published threshold can be named.
• SpliceAI max delta 0.02 (source_registry: spliceai) not applied to PP3 for this missense variant: generic rule restricts missense variants to the REVEL path.
• Generic fallback rule set (source_registry: generic_acmg_combination_rules, based on Richards et al. 2015, PMID 25741868) applied because the custom rule did not trigger.

Still missing

• Custom POLE PP3 lookup rule could not be applied because the exact variant p.R1111Q is absent from Supplementary Tables S2 and S3.
• Only REVEL (0.36) was usable for the missense in-silico evaluation; BayesDel could not contribute because no verified published calibration threshold is available to the pipeline.
• No gene-specific REVEL calibration beyond the generic fallback was available for POLE.

Sources

Reasoning

PP5 is applied only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic. The ClinVar submission audit for VCV000405761 shows expert_panel_submissions = 0: all three submissions (GeneDx SCV002756592, Ambry SCV005927433, Labcorp/Invitae SCV000544082) are single-submitter clinical laboratories, each classifying the variant as Uncertain significance with 'criteria provided, single submitter'. No expert-panel Pathogenic/LP classification exists for this exact variant, so PP5 is not triggered. Aggregate ClinVar labels and ordinary laboratory submissions are explicitly excluded from PP5 per the group's gate.

Rule

PP5 applies only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic (supporting strength); ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions never trigger PP5.

Evidence considered

• clinvar: VCV000405761 submission audit - 3 single-submitter laboratory submissions, all Uncertain significance, 0 expert-panel submissions; no exact-variant expert-panel Pathogenic/LP classification

Sources

Reasoning

Highest observed allele frequency for NM_006231.4:c.3332G>A (p.Arg1111Gln) is in gnomAD v4.1 European (Finnish): AF 1.09e-04 (0.0109%, 7/64,034 alleles); gnomAD v4.1 total AF 3.53e-05 (0.0035%, 57/1,614,138 alleles) with joint grpmax FAF 3.15e-05, and gnomAD v2.1 total AF 2.39e-05 (0.0024%, 6/251,478 alleles) with grpmax FAF 7.01e-06. Every observed frequency is roughly two orders of magnitude below the project non-VCEP BA1 threshold of AF >1% (also far below the >5% ExAC illustration in Richards et al. 2015). Zero homozygotes in both gnomAD datasets. The allele is far too rare for the stand-alone benign BA1 criterion.

Rule

BA1: allele frequency is too high for the disorder, stand-alone benign (Richards et al. 2015, PMID 25741868); project non-VCEP gnomAD threshold AF >1%.

Evidence considered

• gnomAD v2.1 (exome, GRCh37 12-133234500-C-T): 6/251,478 alleles, AF 2.39e-05 (0.0024%), 0 homozygotes, grpmax FAF 7.01e-06 (source_registry key gnomad_v2)
• gnomAD v4.1 (joint, GRCh38 12-132657914-C-T): 57/1,614,138 alleles, AF 3.53e-05 (0.0035%), 0 homozygotes, joint grpmax FAF 3.15e-05; highest subpopulation Finnish AF 1.09e-04 (0.0109%) (source_registry key gnomad_v4)
• BA1 criterion definition from Richards et al. 2015 ACMG/AMP (PMID 25741868), applied via generic_acmg_combination_rules; project non-VCEP gnomAD BA1 threshold AF >1% (max observed 0.0109%, not met)

Still missing

• gnomAD-Canada v1.0 (HostSeq) lookup was registered in source_registry but returned no variant data in this case; not limiting for BA1 because gnomAD v2.1/v4.1 frequencies are already far below threshold.

Sources

Reasoning

BS1 requires an allele frequency greater than expected for the disorder. Highest observed frequency is gnomAD v4.1 Finnish AF 1.09e-04 (0.0109%, 7/64,034 alleles); total AFs are 3.53e-05 (v4.1) and 2.39e-05 (v2.1). These are approximately 30-fold below the project non-VCEP BS1 threshold of AF >0.3% and far below any frequency expected to be incompatible with an autosomal dominant, adult-onset cancer predisposition phenotype. Zero homozygotes in both datasets. BS1 not met.

Rule

BS1: allele frequency is greater than expected for the disorder (Richards et al. 2015, PMID 25741868); project non-VCEP gnomAD threshold AF >0.3%.

Evidence considered

• gnomAD v2.1 (exome): 6/251,478 alleles, AF 2.39e-05 (0.0024%), 0 homozygotes (source_registry key gnomad_v2)
• gnomAD v4.1 (joint): 57/1,614,138 alleles, AF 3.53e-05 (0.0035%), 0 homozygotes; highest subpopulation Finnish AF 1.09e-04 (0.0109%) (source_registry key gnomad_v4)
• BS1 criterion definition from Richards et al. 2015 ACMG/AMP (PMID 25741868), applied via generic_acmg_combination_rules; project non-VCEP gnomAD BS1 threshold AF >0.3% (max observed 0.0109%, not met)

Still missing

• gnomAD-Canada (HostSeq) data not retrieved; not limiting for BS1 given v2.1/v4.1 frequencies are far below threshold.

Sources

Reasoning

BS2 requires observation of the variant in a healthy adult for a recessive (homozygous), dominant (heterozygous), or X-linked (hemizygous) disorder with full penetrance expected at an early age. POLE-associated cancer predisposition (polymerase proofreading-associated polyposis / colorectal cancer susceptibility) is autosomal dominant, adult-onset, and incompletely penetrant, so the 'full penetrance expected at early age' precondition is not satisfied. No phenotyped healthy-adult carrier observations are documented for this variant: gnomAD carrier counts (57/1,614,138 alleles in v4.1; 6/251,478 in v2.1) lack phenotype and age-at-ascertainment documentation and cannot substitute for BS2 evidence, and zero homozygotes were observed in both datasets. BS2 not met.

Rule

BS2: observed in a healthy adult for a recessive (homozygous), dominant (heterozygous), or X-linked (hemizygous) disorder with full penetrance expected at an early age (Richards et al. 2015, PMID 25741868).

Evidence considered

• gnomAD v2.1: 0 homozygotes among 6/251,478 alleles (source_registry key gnomad_v2)
• gnomAD v4.1: 0 homozygotes among 57/1,614,138 alleles; carrier observations without phenotype/age documentation (source_registry key gnomad_v4)
• BS2 criterion definition from Richards et al. 2015 ACMG/AMP (PMID 25741868), applied via generic_acmg_combination_rules; POLE phenotype is dominant, adult-onset, incompletely penetrant per gene_context.json

Still missing

• No phenotyped healthy-adult cohort data (with documented ages and penetrance follow-up) is available for this variant; BS2 would require such observations, which are absent.
• gnomAD-Canada (HostSeq) lookup registered but no variant data retrieved; would not resolve BS2 because population databases lack phenotype documentation.

Sources

Reasoning

BP1 (generic ACMG/AMP 2015; not customized by the governing POLE framework) applies to missense variants in a gene for which primarily truncating variants are known to cause disease. POLE does not meet this precondition: missense is a bona fide and prominent disease mechanism. The governing framework package (León-Castillo et al. 2020) treats five recurrent exonuclease-domain missense substitutions (p.P286R, p.V411L, p.S297F, p.A456P, p.S459F) as established pathogenic hotspot mutations, and OncoKB's gene-level curation states that select POLE mutations lead to ultra-high mutation rates in endometrial and colorectal cancer. Because missense variants are a known POLE disease mechanism, BP1 cannot be applied to a POLE missense variant; BP1 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 via generic_acmg_combination_rules); not customized in the governing framework.

Evidence considered

• vcep_path_250_323 (León-Castillo et al. 2020): five recurrent exonuclease-domain missense substitutions (P286R, V411L, S297F, A456P, S459F) are established pathogenic hotspot mutations - direct evidence that missense is a POLE disease mechanism, not primarily truncating-only.
• oncokb: 'Select POLE mutations lead to ultra-high mutation rates, most frequently in endometrial and colorectal cancer' - gene-level support for missense-mediated pathogenicity.
• generic_acmg_combination_rules: BP1 is the generic rule applied because the custom framework does not customize BP1.

Sources

Reasoning

The custom POLE framework pre-assigns BP4_Supporting only when the exact missense variant appears in Supplementary Table S2 or S3 with a REVEL class of 'Likely benign'/'Likely-benign' and >=4 benign in silico results. The queried variant p.R1111Q (c.3332G>A) is absent from both lookup spreadsheets (verified by direct read of PATH-250-323-s003.xlsx and PATH-250-323-s004.xlsx), so the custom BP4 rule does not trigger and the generic in-silico fallback governs. Under the generic fallback, missense variants are evaluated with REVEL only: REVEL = 0.36 is above the ClinGen SVI BP4 supporting threshold of <=0.016 (Pejaver et al. 2022, PMID 36413997), so BP4 is not met (gray zone). BayesDel (-0.0290682) has no verified published threshold and is not usable for BP4. SpliceAI max delta 0.02 indicates no predicted splice impact, but the splice path is not applied to missense variants under the generic rule (a missense variant's SpliceAI score does not clear its REVEL-assessed protein-level effect), and a low splice score is not a benign-protein-effect prediction. ClinVar single-submitter submissions (Ambry) describe the in silico prediction as 'inconclusive', consistent with no BP4.

Rule

Custom POLE framework (León-Castillo et al. 2020) BP4 rule: BP4_Supporting if exact missense variant is in Supplementary Table S2/S3 with REVEL class 'Likely benign'/'Likely-benign' and >=4 benign in-silico results. Not triggered (variant absent from both tables); generic fallback for missense variants uses REVEL only, BP4_Supporting if REVEL <= 0.016 (ClinGen SVI REVEL calibration, Pejaver et al. 2022, PMID 36413997).

Evidence considered

• REVEL score 0.36 for the exact variant (source_registry: revel); above the ClinGen SVI BP4 supporting threshold REVEL <= 0.016 (Pejaver et al. 2022, PMID 36413997) -> BP4 not met.
• Custom POLE framework BP4 rule requires the exact missense variant in Supplementary Table S2 or S3 with REVEL class 'Likely benign'/'Likely-benign' and >=4 benign in-silico results; p.R1111Q is absent from both tables (vcep_path_250_323_s003, vcep_path_250_323_s004) -> rule not triggered, generic fallback applies.
• BayesDel score -0.0290682 (source_registry: bayesdel) not used for BP4: no verified published threshold can be named.
• SpliceAI max delta 0.02 (source_registry: spliceai) not applied to BP4 for this missense variant: generic rule restricts missense variants to the REVEL path.
• Generic fallback rule set (source_registry: generic_acmg_combination_rules, based on Richards et al. 2015, PMID 25741868) applied because the custom rule did not trigger.

Still missing

• Custom POLE BP4 lookup rule could not be applied because the exact variant p.R1111Q is absent from Supplementary Tables S2 and S3.
• Only REVEL (0.36) was usable for the missense in-silico evaluation; BayesDel could not contribute because no verified published calibration threshold is available to the pipeline.
• No additional calibrated benign-leaning missense predictors (e.g., REVEL <= 0.016) were available for this variant.

Sources

Reasoning

BP6 is applied only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign. The ClinVar submission audit for VCV000405761 shows expert_panel_submissions = 0; all three submissions are single-submitter clinical laboratories classifying the variant as Uncertain significance. No expert-panel Benign/LB classification exists for this exact variant, so BP6 is not triggered. Aggregate ClinVar labels and ordinary laboratory submissions are explicitly excluded from BP6 per the group's gate.

Rule

BP6 applies only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign (supporting strength); ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions never trigger BP6.

Evidence considered

• clinvar: VCV000405761 submission audit - 3 single-submitter laboratory submissions, all Uncertain significance, 0 expert-panel submissions; no exact-variant expert-panel Benign/LB classification

Sources

Reasoning

PS2 (de novo occurrence, paternity and maternity confirmed) cannot be assessed: the evidence bundle contains no proband-level clinical data, no parental testing results, and no de novo occurrence reports for NM_006231.4:c.3332G>A (p.Arg1111Gln). ClinVar variation 405761 (3 clinical laboratory submissions, all 'Uncertain significance', no expert panel) was reviewed and none of the submissions report de novo occurrence or parental testing; the only submission evidence signals present are PP3 (GeneDx, Ambry) and PM2 (Labcorp). The local POLE custom framework (León-Castillo et al. 2020) does not customize PS2, and no ClinGen CSPEC exists for POLE, so the generic ACMG/AMP 2015 definition applies; no evidence satisfies it.

Rule

Generic ACMG/AMP 2015 PS2 (de novo, paternity and maternity confirmed) per Richards et al. 2015 (PMID 25741868); not customized by the local POLE custom framework

Evidence considered

• ClinVar variation 405761 submission audit: 3 single-submitter clinical laboratory submissions (GeneDx, Ambry, Labcorp), all classified Uncertain significance; none report de novo occurrence or parental testing for c.3332G>A
• Literature packet contains no segregation/de novo-themed paper for this variant (only PMID:25394175 referral guideline and PMID:28492532 Sherloc criteria paper, neither variant-specific)

Still missing

• No parental testing (paternity/maternity confirmed) data for the proband is available
• No report of de novo occurrence of c.3332G>A in any ClinVar submission or publication
• No proband phenotype or family history information is present in the case bundle to contextualize a de novo event

Sources

Reasoning

PS3 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

PM3 is not specified by the governing local León-Castillo et al. 2020 custom POLE framework (it customizes only PM1, PS4, PP3, BP4), so the generic ACMG/AMP 2015 definition applies per the framework's own fallback note. PM3 requires, for a recessive disorder, detection of the variant in trans with a pathogenic variant. The case bundle contains no allele-specific evidence whatsoever: ClinVar VCV000405761 (3 germline submissions, all Uncertain significance, single-submitter) describes only a single heterozygous variant with no second POLE allele, no phase/parental testing results, no biallelic (homozygous or compound-heterozygous) observations, and no affected proband carrying this variant in trans with a pathogenic variant. gnomAD v2.1 (6/251,478 alleles) and v4.1 (57/1,614,138 alleles) report 0 homozygotes, but population homozygote counts are not a clinical trans-phase observation and cannot satisfy PM3. The PM5 comparator harvest returned 0 same-residue/allelic candidates, and the only available full-text paper (PMID 25394175) does not mention this variant or POLE. Because the required trans/biallelic observation is absent from the evidence base, PM3 can be neither met nor refuted and is recorded as not assessed with the gaps identified.

Rule

PM3 (generic ACMG/AMP 2015, PMID 25741868): For recessive disorders, detected in trans with a pathogenic variant.

Evidence considered

• Generic ACMG/AMP 2015 PM3 definition (Richards et al. 2015, PMID 25741868): for recessive disorders, detected in trans with a pathogenic variant
• ClinVar VCV000405761: 3 germline submissions, all Uncertain significance (single-submitter review status); no second POLE allele, no phase, no biallelic observation described (clinvar)
• gnomAD v2.1 AF 2.39e-05 (6/251,478 alleles), 0 homozygotes; gnomAD v4.1 AF 3.53e-05 (57/1,614,138 alleles), 0 homozygotes; no homozygous carriers that could represent biallelic observations (gnomad_v2, gnomad_v4)
• PM5 comparator harvest: 0 same-residue/allelic candidates collected; no allelic comparator evidence available (pm5_candidates)
• PMID 25394175 full text checked: 0 matches for c.3332G>A / p.Arg1111Gln / POLE; provides no allelic evidence

Still missing

• No evidence that c.3332G>A is in trans with a pathogenic POLE variant in any affected proband
• No parental/phase testing or segregation data in the case bundle
• No biallelic clinical observations (homozygous or compound heterozygous) reported for this variant; gnomAD homozygote count is 0 but is population data, not a clinical trans observation
• No recessive-phenotype proband (e.g., FILS syndrome, biallelic POLE disease) carrying this variant in trans with a second variant

Sources

Reasoning

PM6 (assumed de novo without confirmation of paternity and maternity) cannot be assessed: there is no evidence in the bundle that this variant arose de novo in any proband. No ClinVar submission, publication, or case record reports an assumed de novo occurrence for NM_006231.4:c.3332G>A (p.Arg1111Gln), and no parental testing data (confirmed or unconfirmed) is available. POLE germline disease (autosomal dominant cancer predisposition context) can in principle present with de novo variants, so the criterion is applicable but unassessable with current evidence.

Rule

Generic ACMG/AMP 2015 PM6 (assumed de novo, without confirmation of paternity and maternity) per Richards et al. 2015 (PMID 25741868); not customized by the local POLE custom framework

Evidence considered

• ClinVar variation 405761 reviewed; no submission cites de novo occurrence or unconfirmed parental testing
• No segregation/de novo literature identified for this exact variant in the literature packet

Still missing

• No assumed de novo occurrence of c.3332G>A reported in any source
• No parental testing results (with or without paternity/maternity confirmation) available
• No proband identification or family structure information in the case bundle

Sources

Reasoning

PP1 (co-segregation with disease in multiple affected family members) cannot be assessed: the case bundle contains no family segregation data for NM_006231.4:c.3332G>A (p.Arg1111Gln). There are no affected relatives tested, no meiosis/segregation counts, no pedigree information, and no publication reporting co-segregation of this exact variant. The ClinVar record (405761) contains only laboratory classifications without segregation detail. No segregation-themed paper was surfaced in the literature pass (criterion_screened empty; only PP5/PS4 candidate paper PMID:25394175).

Rule

Generic ACMG/AMP 2015 PP1 (co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease) per Richards et al. 2015 (PMID 25741868); not customized by the local POLE custom framework

Evidence considered

• ClinVar variation 405761 reviewed; submissions contain no segregation or family-study data
• Literature packet screened for segregation themes returned no variant-relevant papers for PP1

Still missing

• No segregation analysis of c.3332G>A in affected family members is reported
• No number of informative meioses available
• No pedigree or affected-relative testing data in the case bundle; targeted familial testing would be required

Sources

Reasoning

PP2 (generic ACMG/AMP 2015; not customized by the governing POLE framework, which customizes only PM1, PS4, PP3, BP4) requires BOTH (a) a gene with a low rate of benign missense variation and (b) missense as a common mechanism of disease. Component (b) is satisfied for POLE in the sense that exonuclease-domain missense is an established pathogenic mechanism (five established hotspots in the León-Castillo package), but component (a) cannot be evaluated: no gene-level missense constraint metric (e.g., gnomAD missense Z-score or missense o/e ratio) for POLE is present anywhere in the case bundle (case_summary contains only variant-level allele-frequency data; no constraint/LOEUF fields were found). Without a named, verified missense-constraint source, PP2 cannot be met or refuted, so it 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 via generic_acmg_combination_rules); not customized in the governing framework.

Evidence considered

• generic_acmg_combination_rules: PP2 falls back to generic ACMG/AMP 2015 (PMID 25741868) because the custom POLE framework customizes only PM1, PS4, PP3, BP4.
• vcep_path_250_323: framework documents exonuclease-domain missense as an established POLE pathogenic mechanism (relevant to the 'missense common mechanism' limb) but provides no benign-missense-rate/constraint evidence.
• Bundle-wide search: no gnomAD gene-level constraint (Z-score, o/e, LOEUF) or equivalent missense-rate metric for POLE is present in case_summary.json, evidence.json, or vcep_materials.json.

Still missing

• Gene-level missense constraint data for POLE (gnomAD missense Z-score / o/e, or an equivalent named publication) are absent from the case bundle and would be required to evaluate the 'low rate of benign missense variation' component.
• No POLE-specific PP2 specification exists in the governing custom framework.

Sources

Reasoning

PP4 requires a phenotype or family history highly specific for a gene with a single genetic etiology. The evidence bundle contains no proband phenotype data: ClinVar submissions for VCV000405761 carry condition labels 'Not Provided' or empty, and no case-level clinical description, tumor phenotype, or family history is present in case_summary.json, prefetch.json, or evidence.json. Without any phenotypic information, phenotype specificity for POLE-associated disease (polymerase proofreading-associated polyposis, early-onset colorectal/endometrial cancer, POLE-ultramutated tumors) cannot be evaluated. Marked not_assessed pending case-level phenotype data.

Rule

Generic ACMG/AMP PP4: phenotype or family history highly specific for a gene with a single genetic etiology; requires proband clinical data.

Evidence considered

• clinvar: VCV000405761 condition labels ('Not Provided'/empty) confirm no phenotype information is recorded for this variant's submissions

Still missing

• No proband phenotype, clinical history, tumor phenotype, or family history data in the evidence bundle; ClinVar condition labels are 'Not Provided'/empty.

Sources

Reasoning

BS3 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

BS4 (lack of segregation in affected family members) cannot be assessed: no non-segregation evidence exists in the case bundle for NM_006231.4:c.3332G>A (p.Arg1111Gln). There are no family studies showing the variant in unaffected relatives or its absence in affected relatives, no segregation counts, and no publication reporting non-segregation of this exact variant. Note the variant is present at low frequency in gnomAD v4.1 (AF 3.53e-05, 57/1,614,138 alleles, 0 homozygotes); population frequency is assessed under population-frequency criteria (e.g., BS1/BA1/PM2) by other groups and does not by itself constitute BS4 non-segregation evidence.

Rule

Generic ACMG/AMP 2015 BS4 (lack of segregation in affected family members) per Richards et al. 2015 (PMID 25741868); not customized by the local POLE custom framework

Evidence considered

• ClinVar variation 405761 reviewed; no submission reports segregation or non-segregation in family members
• No literature source for this exact variant reports lack of segregation

Still missing

• No family segregation (or non-segregation) study of c.3332G>A is available
• No testing of affected/unaffected relatives reported in ClinVar or literature
• Targeted segregation analysis in affected family members would be required to assess BS4

Sources

Reasoning

BP2 is not specified by the governing local León-Castillo et al. 2020 custom POLE framework, so the generic ACMG/AMP 2015 definition applies (framework fallback note). BP2 requires either (a) the variant observed in trans with a pathogenic variant for a fully dominant disease/gene, or (b) the variant observed in cis with a pathogenic variant in any inheritance pattern. The case bundle contains no phase or co-occurrence data: ClinVar VCV000405761 reports only a single heterozygous VUS with no co-observed pathogenic variant and no family studies; no parental testing, no cis/trans phasing, and no segregation results are available. The only full-text paper (PMID 25394175) does not mention the variant or POLE. Without any trans/cis observation relative to a pathogenic variant, BP2 can be neither met nor refuted and is recorded as not assessed with the gaps identified.

Rule

BP2 (generic ACMG/AMP 2015, PMID 25741868): Observed in trans with a pathogenic variant for a fully dominant disease/gene; OR observed in cis with a pathogenic variant in any inheritance pattern.

Evidence considered

• Generic ACMG/AMP 2015 BP2 definition (Richards et al. 2015, PMID 25741868): observed in trans with a pathogenic variant for a fully dominant disease/gene; OR observed in cis with a pathogenic variant in any inheritance pattern
• ClinVar VCV000405761: 3 germline submissions, all Uncertain significance; no co-observed pathogenic variant, no phase, no family co-occurrence data (clinvar)
• PMID 25394175 full text checked: 0 matches for c.3332G>A / p.Arg1111Gln / POLE; provides no cis/trans co-occurrence evidence

Still missing

• No observation of the variant in trans with a pathogenic variant in a fully dominant POLE disease context
• No observation of the variant in cis with a pathogenic variant on the same allele
• No phasing (parental testing, long-read, or linkage-based) or family co-occurrence data in the case bundle
• No segregation studies reporting the variant together with another POLE variant

Sources

Reasoning

BP5 requires the variant to be found in a case with an alternate molecular basis of disease (e.g., a pathogenic variant in another gene that explains the phenotype). The evidence bundle contains no proband-level molecular or clinical data: no alternate pathogenic variant, no co-occurrence information, no family or segregation context. BP5 cannot be assessed from the available evidence and is marked not_assessed pending case-level data.

Rule

Generic ACMG/AMP BP5: variant found in a case with an alternate molecular basis of disease; requires proband-level co-occurrence data.

Evidence considered

• No source in the bundle contains co-occurrence or alternate-molecular-basis information for this variant

Still missing

• No proband-level data on an alternate molecular basis of disease (pathogenic variant in another gene) is present anywhere in the evidence bundle.