RAD51C c.187A>T, p.(Ile63Phe)

Reasoning

PM2 is met: the variant is absent from large population cohorts - gnomAD v2.1 (exome, ~125.7k individuals) and gnomAD v4.1 (exome, ~730.9k individuals) both returned 'absent' (AF=0). Generic ACMG/AMP 2015 PM2 (Richards et al., PMID 25741868) is satisfied by absence in large population databases, and the local non-VCEP operating threshold (PM2 <0.1%) is trivially satisfied. PM2 is applied at Supporting strength, consistent with the SVI Bayesian re-calibration of PM2 as a supporting-strength criterion (Tavtigian et al. 2018, PMID 29300386). No threshold artifact applies since the variant is absent rather than near a cutoff. gnomAD-Canada v1.0 could not be queried (HTTP 500) but is not required for this determination.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID 25741868): PM2 met when variant is absent in large population cohorts (or at extremely low frequency if recessive); local non-VCEP operating threshold PM2 <0.1%. Absent from gnomAD v2.1 and v4.1 -> met at Supporting strength (SVI re-calibration, Tavtigian et al. 2018, PMID 29300386).

Evidence considered

• gnomAD v2.1 (exome, ~125.7k individuals): variant 17-56772333-A-T absent (search_status 'absent', found=false, AF=0) (registry key: gnomad_v2).
• gnomAD v4.1 (exome, ~730.9k individuals): variant chr17-58694972-A-T absent (search_status 'absent', found=false, AF=0) (registry key: gnomad_v4).
• Generic ACMG/AMP 2015 PM2: absent in large population cohorts (or extremely low frequency if recessive) (Richards et al. 2015, PMID 25741868); local non-VCEP operating threshold PM2 <0.1% (registry key: generic_acmg_combination_rules). Absence (AF=0) satisfies PM2.
• PM2 applied at Supporting strength consistent with SVI Bayesian re-calibration of the ACMG/AMP criteria (Tavtigian et al. 2018, PMID 29300386).
• Governing framework check: ClinGen CSPEC RAD51C v1.0 (doc 1535377178) retrieved but with empty criteria payload; no RAD51C-specific PM2 rule (e.g. PM2_Supporting wording or frequency cutoffs) retrievable, so generic ACMG/AMP 2015 governs (registry key: cspec).

Still missing

• gnomAD-Canada v1.0 (HostSeq genomes) lookup failed with HTTP 500 and contributed no data.
• gnomAD lookups returned 'absent' with empty source_data; per-population counts and coverage metrics were not retrieved, and genome-dataset frequencies were not separately queried (exome datasets only).

Sources

Reasoning

BP4 is met via the missense in-silico path: REVEL 0.084 falls within the ClinGen SVI-calibrated REVEL BP4 moderate interval (0.016 < score <= 0.183; Pejaver et al. 2022, PMID 36413997, Table 2), i.e. strongly in the benign-leaning range, and is applied at supporting strength per the governing generic ACMG/AMP fallback framework (Richards et al. 2015, PMID 25741868), which caps PP3/BP4 at supporting. Corroborating but NOT separately counted (single-tool framework for missense, no double counting): BayesDel noAF -0.3387 falls in the SVI-calibrated BayesDel BP4 supporting interval (-0.36 < score <= -0.18; PMID 36413997, Table 2), and SpliceAI max delta 0.01 is far below the 0.2 splice-altering threshold (PMID 30661751), consistent with no splice impact. The variant is not at a canonical splice position (exon 2, c.146-404).

Rule

generic ACMG/AMP fallback (generic_acmg_combination_rules; PMID 25741868): missense variants use the REVEL path only; REVEL score in the SVI-calibrated BP4 interval (PMID 36413997, Table 2) supports BP4 at supporting strength

Evidence considered

• REVEL score 0.084 falls within the ClinGen SVI-calibrated REVEL BP4 moderate interval (0.016 < score <= 0.183; Pejaver et al. 2022, Am J Hum Genet, PMID 36413997, Table 2); BP4 applied at supporting strength per the governing generic ACMG/AMP fallback framework (Richards et al. 2015, PMID 25741868), which caps PP3/BP4 at supporting.
• BayesDel noAF score -0.3387 falls within the SVI-calibrated BayesDel BP4 supporting interval (-0.36 < score <= -0.18; Pejaver et al. 2022, PMID 36413997, Table 2); concordant with BP4 but not counted as a separate line (single-tool framework for missense).
• SpliceAI max delta 0.01, below the 0.2 splice-altering threshold (Jaganathan et al. 2019, Cell, PMID 30661751); concordant with no splice impact, but the splice path does not independently govern for a missense variant; not counted as a separate line.
• Governing framework: no gene-specific RAD51C PP3/BP4 rule retrievable from CSPEC v1.0 (empty ruleset), so generic ACMG/AMP applies (Richards et al. 2015, PMID 25741868); generic missense rule evaluates REVEL only.

Still missing

• Gene-specific HBOPC/RAD51C CSPEC v1.0 PP3/BP4 rule could not be retrieved (retrieved ruleset payload is empty: criteria [], rule_payload {}, framework_complete false); if the live CSPEC page pre-assigns a gene-specific BP4 rule (e.g., a specific threshold or 'BP4 not applicable'), re-check this determination.
• aGVGD prediction not available for this variant (no score in evidence bundle).
• BayesDel (SVI-calibrated BP4 supporting interval, PMID 36413997 Table 2) and SpliceAI (max delta 0.01) are concordant but intentionally not counted as additional BP4 lines under the governing single-tool framework for missense variants.

Sources

Reasoning

No established pathogenic variant producing the same amino acid change p.(Ile63Phe) (regardless of nucleotide change) was found in the available evidence. ClinVar variation 1781854 contains only a single-submitter Benign entry (Ambry, criteria provided) for this exact change; per the anti-inference rule that final classification is not used as criterion-level evidence, but its presence confirms no pathogenic same-change submission is documented in ClinVar. The pm5_candidates artifact reports 0 candidates seen and that automated same-residue harvesting was skipped (pm5_mode unknown), so no alternative-nucleotide comparator (e.g., c.187A>C p.Ile63Leu, c.187A>G p.Ile63Val) was systematically collected. The only retrieved paper (PMID:25394175) is an ACMG/NSGC referral guideline that does not mention this variant. PS1 therefore cannot be met on available evidence.

Rule

PS1 (generic ACMG/AMP 2015, PMID 25741868): Same amino acid change as a previously established pathogenic variant regardless of nucleotide change. Fallback used because the ClinGen HBOC EP RAD51C VCEP v1.0 payload was not retrievable.

Evidence considered

• clinvar: variation 1781854, NM_058216.3(RAD51C):c.187A>T (p.Ile63Phe) - single-submitter Benign entry (SCV002717657, Ambry); used only to document absence of any pathogenic same-change submission, not as criterion evidence
• pm5_candidates: pm5_mode unknown, eligible_for_classic_pm5_search false, total_candidates_seen 0, same_residue_candidates 0 - no same-residue comparator variants collected
• PMID:25394175 (reviewed, full text): general referral-indications guideline; variant not mentioned; no support for PS1

Still missing

• No systematic ClinVar/literature search was performed for alternative nucleotide changes at c.187 producing p.Ile63Phe or other changes at residue 63 (harvesting skipped, pm5_mode unknown)
• ClinGen HBOC EP RAD51C VCEP v1.0 rule text not retrievable in this case (empty rule payload); a VCEP-specific PS1 rule may differ from the generic definition applied here
• Only one paper was retrieved for this case and it does not mention the variant; a dedicated same-residue literature search is outstanding

Sources

Reasoning

The hotspot prong of PM1 is negative: the Cancer Hotspots query for 'RAD51C I63' returned no statistically significant hotspot (evidence.json CANCERHOTSPOTS result, no_result), and the variant is absent from COSMIC. The functional-domain prong could not be assessed: no RAD51C functional-domain annotation is present in the case bundle (no domain map, no VCEP PM1 text), so no critical/well-established domain at or around residue 63 can be asserted from available evidence. Ile63 is an N-terminal-region residue (exon 2) outside the Walker A ATP-binding motif by primary-sequence inspection, but that observation is not backed by a citable source in this case and is not used as evidence. PM1 is therefore not met on the available evidence.

Rule

PM1 (generic ACMG/AMP 2015, PMID 25741868): Located in a mutational hot spot and/or critical and well-established functional domain (e.g., active site of an enzyme) without benign variation. Fallback used because the ClinGen HBOC EP RAD51C VCEP v1.0 payload was not retrievable.

Evidence considered

• Cancer Hotspots query 'RAD51C I63' returned no statistically significant hotspot (evidence.json CANCERHOTSPOTS, search_status no_result) - negative for the hotspot prong of PM1; no source_registry key exists for this query, so it is described here as case-evidence context
• COSMIC search for RAD51C c.187A>T: not reported in somatic cancers (evidence.json) - no hotspot/cancer-recurrence support
• cspec: ClinGen HBOC EP RAD51C v1.0 detected (doc 1535377178) but rule_payload empty - no VCEP-specific PM1 rule available
• pvs1_variant_assessment: variant classified as missense in exon 2, not in canonical splice consensus - no domain information provided

Still missing

• RAD51C functional-domain map (e.g., Walker A/B boxes, DNA-binding domains, protein-protein interaction regions) not available in the case bundle; needs external annotation source
• ClinGen HBOC EP RAD51C VCEP v1.0 PM1 definition not retrievable in this case (empty rule payload); VCEP may define specific critical domains/hotspots for RAD51C
• Cancer Hotspots query was for residue I63 specifically; no broader hotspot-region analysis was provided

Sources

Reasoning

No pathogenic missense variant at a different amino acid at residue 63 (e.g., p.Ile63Leu, p.Ile63Val, p.Ile63Ser) is documented in the available evidence. The pm5_candidates artifact explicitly reports pm5_mode 'unknown' with automatic same-residue candidate harvesting skipped (eligible_for_classic_pm5_search false, total_candidates_seen 0, same_residue_candidates 0), and ClinVar variation 1781854 contains no pathogenic comparator at this residue. PM5 therefore cannot be met on the available evidence; this is an absence-of-evidence determination, not a comprehensive negative search.

Rule

PM5 (generic ACMG/AMP 2015, PMID 25741868): Novel missense change at an amino acid residue where a different missense change determined to be pathogenic has been seen before. Fallback used because the ClinGen HBOC EP RAD51C VCEP v1.0 payload was not retrievable.

Evidence considered

• pm5_candidates: pm5_mode unknown, eligible_for_classic_pm5_search false, same_residue_candidates 0 - no same-residue pathogenic comparator collected; recommendation level not_applicable
• clinvar: variation 1781854 contains only the c.187A>T (p.Ile63Phe) benign submission; no different-missense comparator at residue 63 documented

Still missing

• Automated same-residue comparator harvesting was skipped (pm5_mode unknown); a dedicated ClinVar/literature query for other missense changes at RAD51C codon 63 is outstanding
• ClinGen HBOC EP RAD51C VCEP v1.0 PM5 rule text not retrievable in this case (empty rule payload)
• No ClinVar submissions for alternate changes at residue 63 were retrieved in this case bundle

Sources

Reasoning

PP2 requires both a low rate of benign missense variation AND missense as a common mechanism of disease in the gene. RAD51C is a homologous-recombination DNA-repair tumor suppressor whose germline disease mechanism is predominantly loss of function: pvs1_gene_context reports lof_mechanism_supported true with supporting germline literature (PMID:20400963 Fanconi anemia-like disorder, plus PMID:35353237, PMID:37149903, PMID:39149814, PMID:41070818). Missense variants are not a common mechanism of RAD51C disease, so the second prong fails and PP2 is not met. No assertion about the gene-wide benign missense rate is made because no such calibration is available in this case.

Rule

PP2 (generic ACMG/AMP 2015, 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. Fallback used because the ClinGen HBOC EP RAD51C VCEP v1.0 payload was not retrievable.

Evidence considered

• pvs1_gene_context: lof_mechanism_supported true; germline RAD51C disease mechanism is loss of function (supporting sources include PMID:20400963, PMID:35353237, PMID:37149903, PMID:39149814, PMID:41070818) - missense is not a common disease mechanism, so PP2 prong 2 fails
• generic_acmg_combination_rules: encodes Richards et al. 2015 (PMID 25741868) generic ACMG/AMP definitions used for PP2 in the absence of retrievable VCEP content

Still missing

• Gene-wide rate of benign missense variation for RAD51C not available in the case bundle (not needed for the determination, which fails on the mechanism prong)
• ClinGen HBOC EP RAD51C VCEP v1.0 PP2 rule text not retrievable in this case (empty rule payload)

Sources

Reasoning

Variant NM_058216.3:c.187A>T (p.Ile63Phe) is a missense substitution in exon 2 (c.146-404), so the missense in-silico path governs PP3/BP4 (REVEL only per the generic fallback framework; splice path and BayesDel are not counted as separate lines). No calibrated computational evidence supports a damaging prediction: REVEL 0.084 is far below the ClinGen SVI-calibrated REVEL PP3 supporting interval (score >= 0.644; Pejaver et al. 2022, PMID 36413997, Table 2); BayesDel noAF -0.3387 is below the SVI-calibrated BayesDel PP3 supporting interval (score >= 0.13; same table, corroborating only); SpliceAI max delta 0.01 is far below the 0.2 splice-altering threshold (Jaganathan et al. 2019, PMID 30661751). All available calibrated predictors are concordant against a damaging computational prediction, so PP3 is affirmatively not met.

Rule

generic ACMG/AMP fallback (generic_acmg_combination_rules; PMID 25741868): missense variants use the REVEL path only; PP3 requires REVEL score in the SVI-calibrated PP3 interval (PMID 36413997, Table 2)

Evidence considered

• REVEL score 0.084; ClinGen SVI-calibrated REVEL PP3 supporting interval is score >= 0.644 (Pejaver et al. 2022, Am J Hum Genet, PMID 36413997, Table 2); 0.084 is far below the interval, so REVEL provides no PP3 support.
• SpliceAI max delta 0.01 (DS_AG 0.0, DS_AL 0.01, DS_DG 0.0, DS_DL 0.0); SpliceAI splice-altering threshold is delta >= 0.2 (Jaganathan et al. 2019, Cell, PMID 30661751); score far below threshold, no splice-impact evidence toward PP3.
• BayesDel noAF score -0.3387; SVI-calibrated BayesDel PP3 supporting interval is score >= 0.13 (Pejaver et al. 2022, PMID 36413997, Table 2); score below interval, no PP3 support (corroborating only; framework counts REVEL only for missense).
• Governing framework: no gene-specific RAD51C PP3/BP4 rule retrievable from CSPEC v1.0 (empty ruleset), so generic ACMG/AMP applies (Richards et al. 2015, PMID 25741868); generic missense rule evaluates REVEL only.

Still missing

• Gene-specific HBOPC/RAD51C CSPEC v1.0 PP3/BP4 rule could not be retrieved (retrieved ruleset payload is empty: criteria [], rule_payload {}, framework_complete false); if the live CSPEC page pre-assigns a gene-specific PP3 rule, re-check this determination.
• aGVGD prediction not available for this variant (no score in evidence bundle); would only add to the missense in-silico line already covered by REVEL.

Sources

Reasoning

PP5 applies only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic. ClinVar variation 1781854 has no expert-panel submission (expert_panel_submissions: 0; is_expert_panel false on the sole SCV); the only submission is an ordinary clinical-laboratory assertion (Ambry Genetics, Benign, 'criteria provided, single submitter', 1 star). Per the governing instruction, ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions must never trigger PP5. Additionally, the sole ClinVar label is Benign, the opposite direction. The only ClinVar-linked publication (PMID 25394175) is a referral guideline that does not mention the variant and is not an expert-panel assertion. PP5 is therefore definitively not met.

Rule

Governing instruction for this group: PP5 triggered only by an exact-variant ClinVar expert-panel Pathogenic/Likely-pathogenic classification; no such classification exists for this variant.

Evidence considered

• ClinVar variation 1781854 submission audit: 0 expert-panel submissions, 0 criterion-lead submissions; single SCV002717657 from Ambry Genetics (clinical testing, germline, Benign, criteria provided single submitter).

Sources

Reasoning

Variant NM_058216.3:c.187A>T (p.Ile63Phe) is absent from gnomAD v2.1 (exome) and gnomAD v4.1 (exome) variant lookups (search_status 'absent', found=false for both 17-56772333-A-T and chr17-58694972-A-T), i.e. allele frequency = 0. BA1 requires a common-population allele frequency far above any reasonable threshold (generic ACMG/AMP 2015 MAF >5%, PMID 25741868; local non-VCEP operating threshold BA1 >1%). AF=0 cannot meet BA1 under any threshold, so BA1 is not met. No threshold artifact applies because the variant is absent rather than near a cutoff.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID 25741868): BA1 met when allele frequency >5% in a large population database; local non-VCEP operating threshold BA1 >1% in gnomAD. Variant absent (AF=0) -> not met.

Evidence considered

• gnomAD v2.1 (exome) variant page 17-56772333-A-T: search_status 'absent', found=false, AF=0 (registry key: gnomad_v2).
• gnomAD v4.1 (exome) variant page chr17-58694972-A-T: search_status 'absent', found=false, AF=0 (registry key: gnomad_v4).
• Generic ACMG/AMP 2015 BA1: AF >5% in a large population database (Richards et al. 2015, PMID 25741868); local non-VCEP operating threshold BA1 >1% (same framework, registry key: generic_acmg_combination_rules). AF=0 is below every threshold, so BA1 not met.
• Governing framework check: ClinGen CSPEC RAD51C v1.0 (doc 1535377178) retrieved but with empty criteria payload (criteria=[], rule_payload={}, framework_complete=false, references=[]); no RAD51C-specific BA1 rule retrievable, so generic ACMG/AMP 2015 governs (registry key: cspec).

Still missing

• gnomAD lookups returned 'absent' with empty source_data; no per-population allele counts or coverage metrics were retrieved, and only exome datasets were queried (genomes not separately checked). Absence in two large exome datasets is nonetheless sufficient to exclude BA1.

Sources

Reasoning

BS1 requires an allele frequency greater than expected for the disorder. The variant is absent from gnomAD v2.1 (exome) and gnomAD v4.1 (exome) (AF=0), far below the local non-VCEP operating threshold BS1 >0.3% and below any frequency-based BS1 threshold derived for a moderate-penetrance adult-onset ovarian/breast cancer gene. With zero observed alleles, BS1 cannot be met.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID 25741868): BS1 met when allele frequency is greater than expected for the disorder; local non-VCEP operating threshold BS1 >0.3% in gnomAD. Variant absent (AF=0) -> not met.

Evidence considered

• gnomAD v2.1 (exome) variant page 17-56772333-A-T: search_status 'absent', found=false, AF=0 (registry key: gnomad_v2).
• gnomAD v4.1 (exome) variant page chr17-58694972-A-T: search_status 'absent', found=false, AF=0 (registry key: gnomad_v4).
• Generic ACMG/AMP 2015 BS1: allele frequency greater than expected for disorder (Richards et al. 2015, PMID 25741868); local non-VCEP operating threshold BS1 >0.3% (registry key: generic_acmg_combination_rules). AF=0 is below any BS1 threshold, so BS1 not met.
• Governing framework check: ClinGen CSPEC RAD51C v1.0 (doc 1535377178) retrieved but with empty criteria payload; no RAD51C-specific BS1 rule retrievable, generic ACMG/AMP 2015 governs (registry key: cspec).

Still missing

• No per-population AF breakdown or coverage metrics retrieved (source_data empty); exome datasets only. Absence is sufficient to exclude BS1.

Sources

Reasoning

BS2 requires the variant observed in a healthy adult individual for a dominant disorder with full penetrance expected at an early age (homozygous observation for a recessive disorder). No such observation exists for this variant: it is absent from gnomAD v2.1 and v4.1 (zero carriers among ~856k exome individuals across the two datasets), and no curated unaffected-control or case-control observation of the variant is present in the case bundle. Additionally, RAD51C-associated ovarian/breast cancer is a moderate-penetrance adult-onset phenotype (not full penetrance at early age), and the recessive Fanconi anemia phenotype is early-onset/severe, so no generic BS2 scenario is plausibly met. BS2 is therefore not met.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID 25741868): BS2 met when the variant is observed in a healthy adult (dominant, full penetrance at early age) or homozygous (recessive). No healthy-adult/homozygote observations; variant absent from gnomAD -> not met.

Evidence considered

• gnomAD v2.1 (exome): variant absent (found=false), zero healthy-carrier observations (registry key: gnomad_v2).
• gnomAD v4.1 (exome): variant absent (found=false), zero healthy-carrier observations (registry key: gnomad_v4).
• Generic ACMG/AMP 2015 BS2: observed in a healthy adult for a dominant disorder with full penetrance expected at an early age, or homozygous in a recessive disorder (Richards et al. 2015, PMID 25741868); no such observation exists, so BS2 not met (registry key: generic_acmg_combination_rules).
• Governing framework check: ClinGen CSPEC RAD51C v1.0 (doc 1535377178) retrieved but with empty criteria payload; no RAD51C-specific BS2 rule retrievable, generic ACMG/AMP 2015 governs (registry key: cspec).

Still missing

• No curated unaffected-control or case-control observation data beyond gnomAD is available in the case bundle; the single ClinVar submission (Benign, Ambry, 1-star, no criteria/PMIDs) is not usable as criterion evidence and was not used.

Sources

Reasoning

BP1 applies only to genes in which primarily truncating variants are known to cause disease. This is not the case for RAD51C: established pathogenic missense variants exist. pvs1_gene_context documents PMID:20400963 (Vaz et al. 2010), which identified a homozygous RAD51C missense mutation causing a Fanconi anemia-like disorder, with direct abstract text: 'We have identified a homozygous missense mutation in the RAD51C gene in a consanguineous family with multiple severe congenital abnormalities characteristic of FA.' Because missense variants are established disease-causing mechanisms in RAD51C, the truncating-only precondition fails and BP1 is not met.

Rule

BP1 (generic ACMG/AMP 2015, PMID 25741868): Missense variant in a gene for which primarily truncating variants are known to cause disease. Fallback used because the ClinGen HBOC EP RAD51C VCEP v1.0 payload was not retrievable.

Evidence considered

• pvs1_gene_context: documents PMID:20400963 abstract 'We have identified a homozygous missense mutation in the RAD51C gene in a consanguineous family with multiple severe congenital abnormalities characteristic of FA' - established pathogenic missense in RAD51C, so BP1's truncating-only precondition fails
• generic_acmg_combination_rules: encodes Richards et al. 2015 (PMID 25741868) generic ACMG/AMP definitions used for BP1 in the absence of retrievable VCEP content

Still missing

• ClinGen HBOC EP RAD51C VCEP v1.0 BP1 rule text not retrievable in this case (empty rule payload); VCEP may define the truncating-variant proportion threshold for RAD51C
• Quantitative proportion of truncating vs missense pathogenic RAD51C variants not provided in the case bundle (the mechanism-level evidence is sufficient to fail BP1)

Sources

Reasoning

BP6 applies only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign. Although the ClinVar label for variation 1781854 is Benign, it originates from a single clinical laboratory (Ambry Genetics, SCV002717657, review status 'criteria provided, single submitter', 1 star) with zero expert-panel submissions. Per the governing instruction, ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions must never trigger BP6. The Benign assertion therefore does not meet the expert-panel precondition, and BP6 is definitively not met.

Rule

Governing instruction for this group: BP6 triggered only by an exact-variant ClinVar expert-panel Benign/Likely-benign classification; the sole ClinVar assertion is a non-expert laboratory submission and must never trigger BP6.

Evidence considered

• ClinVar variation 1781854: aggregate classification Benign from 1 clinical-laboratory submission (Ambry Genetics); expert_panel_submissions: 0; review_stars: 1 (criteria provided, single submitter).

Sources

Reasoning

No proband or parental genotype data is present anywhere in the case bundle. There is no observation of a de novo occurrence of NM_058216.3:c.187A>T (p.Ile63Phe) in a proband with confirmed maternity and paternity. The sole ClinVar submission (Ambry, SCV002717657) provides no criterion-level evidence (criteria_found: [], usable_for_criterion_leads: false, no validated PMIDs) and cannot be used to infer de novo status. The only full-text publication (PMID:25394175) is an ACMG/NSGC referral-indications practice guideline that does not mention the variant or any proband/family data. PS2 cannot be evaluated.

Rule

De novo (maternity and paternity confirmed) - strong evidence of pathogenicity (Richards et al. 2015, PMID:25741868); no de novo data available, so not assessable.

Evidence considered

• generic_acmg_combination_rules
• PMID:25394175

Still missing

• No proband phenotype or variant genotype data available for this case
• No parental genotype testing results (needed to establish de novo occurrence and to confirm maternity/paternity)

Sources

Reasoning

No retrievable VCEP-specific PS3 rule: the ClinGen Hereditary Breast, Ovarian and Pancreatic Cancer (HBOPC) RAD51C v1.0 CSPEC record (doc 1535377178) was found but its rule payload/criteria are empty in this workspace (criteria.json = [], rule_payload = {}, framework_complete = false), so the generic ACMG/AMP 2015 definition applies: PS3 = well-established functional studies show a deleterious effect (PMID 25741868). No functional assay evidence for NM_058216.3:c.187A>T (p.Ile63Phe) was identified in any consulted source: OncoKB reports no variant-specific reviewed functional evidence (not_reviewed, is_variant_specific = false); COSMIC and Cancer Hotspots have no record; the only literature paper (PMID:25394175) is an ACMG/NSGC referral-indications practice guideline whose full text does not mention this variant or RAD51C and contains no functional data. Evidence is therefore insufficient to assess PS3; the criterion is not met from available data.

Rule

PS3 (generic ACMG/AMP 2015, PMID 25741868): Well-established functional studies show a deleterious effect.

Evidence considered

• OncoKB: no variant-specific reviewed functional evidence for RAD51C p.Ile63F (evidence.json, oncokb result; is_variant_specific=false, not_reviewed=true).
• COSMIC: no record for RAD51C c.187A>T (evidence.json, cosmic result; negative).
• SpliceAI: max delta score 0.01, no predicted splice impact (evidence.json, spliceai result).
• ClinVar variation 1781854: single-submitter (Ambry) Benign, criteria provided but no criteria found, no validated PMIDs (prefetch.json clinvar submission_audit).
• PMID:25394175 full text (publications/25394175.txt): ACMG/NSGC referral indications practice guideline; variant not mentioned; no functional data.

Still missing

• No variant-specific functional assay data for p.Ile63Phe (e.g., homology-directed repair / RAD51 focus formation assays, mitomycin C sensitivity, or other validated RAD51C functional assays) identified in any consulted source.
• A VCEP-specific PS3 specification for the HBOPC RAD51C v1.0 framework could not be retrieved (empty rule payload); only generic ACMG/AMP 2015 (PMID 25741868) was available to apply.

Sources

Reasoning

No case-control enrichment data exists for RAD51C c.187A>T (p.Ile63Phe). The variant is absent from gnomAD v2.1 and v4.1, but population-frequency data cannot substitute for case/control counts. No case series, cohort study, or individual case report mentioning this exact variant was identified in the literature pass; the only retrieved publication (PMID 25394175) is an ACMG/NSGC referral-indications practice guideline whose full text does not mention the variant (verified by full-text search for c.187A>T, p.Ile63Phe, p.I63F, RAD51C). The ClinVar entry (variation 1781854) is a single laboratory submission (Ambry Genetics, Benign) with no affected-individual counts. Evidence is therefore insufficient to evaluate PS4; the gap is the complete absence of case-control or case-series data for this exact variant.

Rule

Generic ACMG/AMP 2015 PS4 (PMID 25741868): variant prevalence in affected individuals significantly increased versus controls; the RAD51C ClinGen HBOPC CSPEC (doc 1535377178) was detected but carried no usable criterion rules (rule_payload empty, framework_complete false), so generic rules govern.

Evidence considered

• ClinVar variation 1781854: one clinical-laboratory submission (Ambry Genetics, Benign, criteria provided single submitter); contains no case/control counts and no affected-individual observations usable for PS4.
• PMID:25394175 full text reviewed (2034 lines): general cancer-genetics referral practice guideline; exact variant c.187A>T / p.Ile63Phe / RAD51C not mentioned anywhere in the text.
• gnomAD v2.1 and v4.1 absence (variant not observed) noted as population-frequency context only, not case-control enrichment.

Still missing

• No case-control enrichment study or case series reporting RAD51C c.187A>T in affected individuals; any PS4 claim would require prevalence data in cases versus controls for this exact variant.

Sources

Reasoning

Governing framework: the ClinGen Hereditary Breast, Ovarian and Pancreatic Cancer EP RAD51C v1.0 CSPEC ruleset (cspec, doc 1535377178) was retrieved but contains no criterion rules (criteria [], rule_payload {}, framework_complete false); per instructions generic ACMG/AMP (Richards et al. 2015, PMID 25741868) is therefore used. Generic PM3: 'For recessive disorders, detected in trans with a pathogenic variant.' RAD51C has a bona fide recessive disease association (Fanconi anemia complementation group O, biallelic RAD51C variants; see cspec raw_rule_origin 'Autosomal recessive inheritance' for MONDO:0019391), so the criterion is theoretically applicable to a missense variant like c.187A>T (p.Ile63Phe). However, no evidence in this case supports an in-trans observation: (1) no affected proband with biallelic RAD51C variants including c.187A>T is reported in any case source; (2) no phase/parental testing, compound-heterozygosity, or segregation data exist in the case bundle; (3) the only clinical report is a single-submitter ClinVar entry (VCV001781854, Ambry Genetics, 'Benign', 1 star) whose submission (SCV002717657) carries no usable criteria or PMIDs and cannot be used as criterion evidence nor as a source of phase information (and its final classification must not be treated as criterion evidence); (4) gnomAD v2.1/v4.1 absence documents that no population homozygotes exist but provides no proband-phase data; gnomAD-Canada lookup failed (HTTP 500). With zero biallelic/in-trans observations, PM3 cannot be assessed; it is neither met nor refuted by available data.

Rule

Generic ACMG/AMP 2015 PM3: 'For recessive disorders, detected in trans with a pathogenic variant' (Richards et al. 2015, PMID 25741868); no applicable VCEP-specific PM3 rule retrieved (RAD51C HBOPC CSPEC v1.0 ruleset empty)

Evidence considered

• cspec (RAD51C v1.0 ruleset): establishes autosomal recessive Fanconi anemia disease context (MONDO:0019391) making PM3 theoretically applicable, but contains no PM3 rule text; generic PM3 (Richards et al. 2015, PMID 25741868) applied
• clinvar (VCV001781854): single submission (SCV002717657, Ambry Genetics) classified Benign, criteria provided but no criteria/PMIDs usable for criterion leads; provides no biallelic or in-trans observation for PM3
• gnomad_v2 / gnomad_v4: variant absent in population cohorts; no homozygous carriers observed, but absence does not constitute proband-phase evidence
• PMID:25394175 (full text reviewed): practice guideline on referral indications; does not mention c.187A>T or RAD51C and contains no allelic/phase data

Still missing

• No affected proband(s) reported with biallelic RAD51C variants where c.187A>T is one allele (no FA-O / compound-heterozygote observation)
• No phase information (trans vs cis) relative to any pathogenic RAD51C variant; no parental testing or segregation data in the case evidence
• No published case series or ClinVar criterion submissions documenting an in-trans pathogenic variant for this allele

Sources

Reasoning

No de novo occurrence of NM_058216.3:c.187A>T is reported in any source, and no proband or parental genotype data exists in the case bundle to establish even an assumed de novo event without parental confirmation. The ClinVar entry provides only a single-submitter Benign classification with no criteria or supporting family data, and the only retrieved full-text paper (PMID:25394175) contains no variant or family information. PM6 cannot be evaluated.

Rule

Assumed de novo, but without confirmation of paternity and maternity - moderate evidence of pathogenicity (Richards et al. 2015, PMID:25741868); no de novo observation available, so not assessable.

Evidence considered

• generic_acmg_combination_rules
• PMID:25394175

Still missing

• No proband genotype data and no parental testing results to identify or even assume a de novo occurrence
• No family history or parental age information to support assumed de novo status

Sources

Reasoning

No segregation data exists for NM_058216.3:c.187A>T. There are no affected family members genotyped, no reported meioses, and no co-segregation observations in any source. The case's literature packet contains only PMID:25394175 (an ACMG/NSGC referral-indications guideline) which does not mention the variant or segregation, and the ClinVar submission carries no family/segregation evidence (criterion_leads: []). PP1 cannot be evaluated.

Rule

Co-segregation with disease in affected family members in multiple family members - supporting evidence of pathogenicity (Richards et al. 2015, PMID:25741868); no segregation observations available, so not assessable.

Evidence considered

• generic_acmg_combination_rules
• PMID:25394175

Still missing

• No affected relatives genotyped for the variant (no co-segregation data, no number of informative meioses reported)

Sources

Reasoning

No phenotype data are available for any carrier of RAD51C c.187A>T (p.Ile63Phe). The only clinical assertion is the ClinVar laboratory submission, which lists the generic condition 'Hereditary cancer-predisposing syndrome' with no proband phenotype, tumor type, or family-history details. RAD51C is associated with hereditary breast/ovarian cancer and Fanconi anemia (FA-like) phenotypes; without any documented personal or family history for a carrier of this exact variant, phenotype specificity cannot be evaluated. Evidence is insufficient; the gap is the absence of any proband phenotype or family-history information.

Rule

Generic ACMG/AMP 2015 PP4 (PMID 25741868): phenotype or family history highly specific for a gene with a single genetic etiology; RAD51C ClinGen HBOPC CSPEC present but without usable criterion rules, so generic rules govern.

Evidence considered

• ClinVar variation 1781854: submission condition listed only as 'Hereditary cancer-predisposing syndrome'; no phenotype or family-history details provided.

Still missing

• No proband phenotype, tumor type, or family-history data for any individual carrying c.187A>T; PP4 requires a phenotype or family history highly specific for a gene with a single genetic etiology.

Sources

Reasoning

No retrievable VCEP-specific BS3 rule (HBOPC RAD51C v1.0 CSPEC record found but empty rule payload; criteria.json = []), so the generic ACMG/AMP 2015 definition applies: BS3 = well-established functional studies show no damaging effect on protein function or splicing (PMID 25741868). No well-established functional study of p.Ile63Phe is available. The SpliceAI max delta score of 0.01 is an in silico splicing prediction, not a functional assay, and cannot satisfy BS3 on its own. The ClinVar Benign classification (1 clinical laboratory, 1-star, criteria provided but no criteria/PMIDs attached, no expert panel) is another source's final classification with no functional evidence behind it and was not used to infer BS3. Evidence is therefore insufficient to assess BS3.

Rule

BS3 (generic ACMG/AMP 2015, PMID 25741868): Well-established functional studies show no damaging effect on protein function or splicing.

Evidence considered

• SpliceAI: max delta score 0.01 (in silico only; not a well-established functional study).
• OncoKB: no variant-specific reviewed functional evidence for RAD51C p.Ile63F (evidence.json, oncokb result).
• ClinVar variation 1781854: Benign, single submitter (Ambry, SCV002717657), 1-star; evidence_signals.criteria_found = [] and no validated PMIDs; final classification not inferred as BS3.
• PMID:25394175 full text (publications/25394175.txt): variant not mentioned; no functional assay data.

Still missing

• No well-established functional study demonstrating no damaging effect on RAD51C protein function or splicing for p.Ile63Phe (e.g., validated functional assays showing wild-type-like homologous recombination repair activity).
• ClinVar Benign call is single-submitter (1-star) with no criteria or validated PMIDs, and carries no functional assay data; cannot substitute for BS3 evidence.
• A VCEP-specific BS3 specification for the HBOPC RAD51C v1.0 framework could not be retrieved (empty rule payload); only generic ACMG/AMP 2015 (PMID 25741868) was available to apply.

Sources

Reasoning

No segregation data exists to evaluate lack of segregation. No affected family members have been genotyped, so there is no observation of an affected individual lacking the variant (non-segregation) and no informative meioses to count. The ClinVar single-submitter Benign classification (Ambry, SCV002717657) was not used as evidence for BS4 because no criteria were provided with the submission and a final classification from another source must not be converted into criterion-level evidence. BS4 cannot be evaluated.

Rule

Lack of segregation in affected family members - strong evidence of benign impact (Richards et al. 2015, PMID:25741868); no segregation data available, so not assessable.

Evidence considered

• generic_acmg_combination_rules
• PMID:25394175

Still missing

• No affected family member genotype data available to test for non-segregation
• No family segregation study reported in any retrieved source

Sources

Reasoning

Governing framework: generic ACMG/AMP (Richards et al. 2015, PMID 25741868) because the retrieved RAD51C CSPEC v1.0 ruleset contains no criterion rules. Generic BP2: 'Observed in trans with a pathogenic variant for a fully dominant disease/disorder; or observed in cis with a pathogenic variant in any inheritance pattern.' RAD51C is associated with autosomal dominant hereditary breast/ovarian cancer (cspec raw_rule_origin: MONDO:0016248 familial ovarian cancer, MONDO:0016419 hereditary breast carcinoma), so a trans-with-pathogenic observation would be a benign-direction signal for this variant in the dominant disease context. However, no phase observations of any kind exist for c.187A>T: no variant is reported in trans or in cis with a pathogenic RAD51C variant; there is no parental testing, family study, or segregation data in the case bundle; the single ClinVar submission (SCV002717657, Ambry Genetics, 'Benign', 1 star, no usable criteria/PMIDs) does not report phase and its final classification cannot be used as criterion evidence; gnomAD v2.1/v4.1 absence provides no phase information. Without any cis/trans observation relative to a pathogenic variant, BP2 cannot be assessed; it is neither met nor refuted by available data.

Rule

Generic ACMG/AMP 2015 BP2: 'Observed in trans with a pathogenic variant for a fully dominant disease/disorder; or observed in cis with a pathogenic variant in any inheritance pattern' (Richards et al. 2015, PMID 25741868); no applicable VCEP-specific BP2 rule retrieved (RAD51C HBOPC CSPEC v1.0 ruleset empty)

Evidence considered

• cspec (RAD51C v1.0 ruleset): establishes autosomal dominant HBOC disease context (MONDO:0016248, MONDO:0016419) in which a trans-with-pathogenic observation would support BP2; no BP2 rule text in retrieved ruleset
• clinvar (VCV001781854): single-submitter Benign entry without criteria or phase information; not usable as a cis/trans observation
• gnomad_v2 / gnomad_v4: variant absent from population cohorts; no allele-phase information derivable
• PMID:25394175 (full text reviewed): referral-indications guideline; does not mention the variant and contains no phase/allelic data

Still missing

• No observation of c.187A>T in trans with a pathogenic RAD51C variant in an affected/unaffected individual (dominant disease context)
• No observation of c.187A>T in cis with a pathogenic RAD51C variant in any inheritance pattern
• No parental testing, familial co-segregation, or haplotype/phase data available for this case

Sources

Reasoning

BP5 requires evidence of an alternative molecular basis for disease in a proband carrying this variant (e.g., a clearly pathogenic variant in another gene explaining the phenotype). No proband or family data exist for RAD51C c.187A>T (p.Ile63Phe), and no alternate-gene or second-variant findings were identified in any retrieved source. Evidence is insufficient; the gap is the complete absence of case data needed to evaluate whether an alternative cause has been demonstrated.

Rule

Generic ACMG/AMP 2015 BP5 (PMID 25741868): alternative molecular basis for disease found in a proband with this variant; no such evidence available.

Evidence considered

• No source retrieved contains any proband or family data for c.187A>T; ClinVar submission provides no individual-level findings.

Still missing

• No proband-level data (no individual with this variant and a documented alternative molecular diagnosis) with which to evaluate BP5.

Sources