RAD51C c.904+5G>T, p.?

Reasoning

NM_058216.3 is the preferred RAD51C transcript. Although c.904+5G>T is outside the canonical ±1/2 splice dinucleotides, variant-specific RNA and minigene evidence demonstrate exon 6 exclusion. The resulting transcript was reported as V280GfsX11, an early frameshift/premature-termination consequence expected to undergo nonsense-mediated decay. RAD51C loss of function is an established disease mechanism; therefore, under the generic ClinGen SVI PVS1 framework, the experimentally established null consequence meets PVS1 at very strong strength.

Rule

Generic ClinGen SVI PVS1 framework: a splice variant outside ±1/2 requires functional confirmation before use; confirmed aberrant splicing that yields a frameshift/premature termination in a gene with established loss-of-function disease mechanism supports PVS1.

Evidence considered

• The applicable RAD51C CSPEC identifies NM_058216.3 as the preferred transcript but contains no structured PVS1 rule; generic ClinGen SVI PVS1 guidance was therefore used for this criterion.
• PMID:20400964 reports c.904+5G>T as IVS 6 904+5G>T with predicted consequence V280GfsX11. RT-PCR from two carrier tumor samples yielded a product lacking exon 6, and a RAD51C exon-6 splicing reporter independently demonstrated exon 6 exclusion.
• The PVS1 gene-context record supports RAD51C loss of function as a germline disease mechanism and permits generic PVS1 assessment.

Sources

Reasoning

The exact RAD51C NM_058216.3:c.904+5G>T variant was evaluated in two complementary, variant-specific functional systems. Patient-derived tumor RNA from two carriers showed an RT-PCR product lacking exon 6 and no such product in controls. An independent RAD51C exon-6 splicing reporter/minigene assay, containing flanking intronic sequence, reproduced exon 6 exclusion caused by the variant. The assay findings are biologically relevant to the intron-6 donor-site mechanism and are supported by appropriate control comparisons; the evidence is sufficient for strong PS3 under generic ACMG/AMP functional-evidence calibration because the available RAD51C-specific VCEP payload does not provide a distinct PS3 strength rule.

Evidence considered

• The RAD51C VCEP specification is applicable to this case, but its extracted criterion payload is empty and does not provide a PS3-specific strength calibration; therefore the assay evidence was calibrated using generic ACMG/AMP principles.
• PMID:20400964 reports the exact 904+5G>T variant in two hereditary breast/ovarian-cancer pedigrees. RT-PCR of paraffin-embedded tumor RNA from two carriers produced an exon-6-lacking transcript that was absent from controls.
• PMID:20400964 independently tested the exact variant in a RAD51C exon-6 splicing reporter containing flanking intronic sequence and found variant-associated exon-6 exclusion, directly supporting loss of normal splicing.
• PMID:20400964 also reports loss of the wild-type allele in breast and ovarian tumor tissues from affected individuals in the relevant pedigree, providing biologically coherent tumor-level support, although the PS3 assignment is based on the direct RNA and reporter assays rather than this observation alone.

Sources

Reasoning

PM2 is met at supporting strength under the generic rare-variant interpretation used here. The variant is present only at ultra-rare frequency in gnomAD v2.1 and v4.1, with no observed homozygotes, and is absent from the supplied gnomAD-Canada result. The frequency is compatible with a rare RAD51C-associated disorder and does not show a population-frequency artifact that would preclude PM2.

Rule

Generic ACMG/AMP PM2 applied at supporting strength for an absent or ultra-rare variant in population databases; observed frequency is 8.73453e-06 in gnomAD v4.1 and 1.59658e-05 in gnomAD v2.1, with zero homozygotes.

Evidence considered

• The RAD51C ClinGen CSPEC record is present but marked incomplete with no usable PM2 population rule; generic ACMG/AMP guidance was therefore applied.
• Generic ACMG/AMP PM2 supports a pathogenic interpretation when a variant is absent or extremely rare in population databases, accounting for database quality and disease prevalence (Richards et al., PMID:25741868).
• gnomAD v4.1 reports 14/1,602,834 alleles, AF 8.73453e-06, and zero homozygotes; the highest broad population AF is 1.33955e-05.
• gnomAD v2.1 reports 4/250,536 alleles, AF 1.59658e-05, and zero homozygotes; the highest listed subpopulation AF is 3.52361e-05.
• The variant is absent from the supplied gnomAD-Canada v1.0 result.

Still missing

• The case bundle does not provide a RAD51C-specific maximum credible allele-frequency calculation or a CSPEC quantitative PM2 threshold.

Sources

Reasoning

RAD51C NM_058216.3:c.904+5G>T is a non-canonical intronic donor-region variant, so the splice-impact path is applicable. No usable RAD51C-specific PP3/BP4 rule or lookup is present in the supplied CSPEC materials; under the generic calibrated splice path, the SpliceAI maximum delta of 0.949 exceeds the PP3 threshold of >0.2.

Rule

Generic ACMG/AMP fallback, non-missense splice-impact path: SpliceAI maximum delta >0.2 supports PP3; no RAD51C-specific PP3/BP4 rule was available in the supplied CSPEC materials.

Evidence considered

• The supplied RAD51C CSPEC record identifies the governing framework but contains an empty criteria payload and no PP3/BP4-specific rule or lookup assignment.
• SpliceAI Lookup reports DS_AL=0.949, DS_DL=0.940, and a maximum delta score of 0.949 for NM_058216.3:c.904+5G>T. Applying the generic non-missense SpliceAI calibration (PP3 when maximum delta >0.2) yields PP3 at supporting strength.
• PMID:34299313 independently reports that c.904+5G>T was predicted to inactivate the RAD51C intron-6 wild-type donor site, but this paper-specific Human Splicing Factor prediction was not used to add or increase PP3 strength; the calibrated SpliceAI result alone supports this call.

Sources

Reasoning

No exact-variant ClinVar pathogenic or likely pathogenic expert-panel assertion is present. The record contains only non-expert laboratory, research, and curation submissions, which cannot trigger PP5.

Rule

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

Evidence considered

• ClinVar variation 142762 is an exact match for c.904+5G>T, but the submission audit records zero expert-panel submissions; all pathogenic/likely pathogenic submissions are non-expert.

Sources

Reasoning

BA1 is not met. The variant is far below the generic ACMG/AMP stand-alone benign population-frequency threshold of >5%: gnomAD v4.1 reports 14/1,602,834 alleles (AF 8.73453e-06; 0.000873%), and gnomAD v2.1 reports 4/250,536 alleles (AF 1.59658e-05; 0.00160%).

Rule

Generic ACMG/AMP BA1: allele frequency >5% is stand-alone benign; this variant is substantially below that threshold.

Evidence considered

• The RAD51C ClinGen CSPEC record is present, but its population rule payload is empty and the framework is marked incomplete; therefore generic ACMG/AMP population guidance was used for this criterion.
• The generic ACMG/AMP criterion BA1 is a stand-alone benign criterion when allele frequency is greater than 5% (Richards et al., PMID:25741868).
• gnomAD v4.1 reports 14 alternate alleles among 1,602,834 alleles, AF 8.73453e-06, with zero homozygotes; the highest broad population AF is 1.33955e-05 in African/African American individuals.
• gnomAD v2.1 reports 4 alternate alleles among 250,536 alleles, AF 1.59658e-05, with zero homozygotes; the highest listed subpopulation AF is 3.52361e-05 in European non-Finnish individuals.

Sources

Reasoning

BS1 is not met. The observed population frequency is extremely low and is not greater than expected for a rare RAD51C disease-associated variant. The available data do not show a frequency excess compatible with BS1.

Rule

Generic ACMG/AMP BS1: allele frequency greater than expected for the disorder; no frequency excess is demonstrated here.

Evidence considered

• The RAD51C ClinGen CSPEC record is present, but it contains no usable BS1 population rule or quantitative threshold and is marked incomplete; generic ACMG/AMP guidance was therefore used.
• Generic ACMG/AMP BS1 requires an allele frequency greater than expected for the disorder (Richards et al., PMID:25741868).
• gnomAD v4.1 shows AF 8.73453e-06 from 14/1,602,834 alleles and zero homozygotes; broad-population maximum AF is 1.33955e-05.
• gnomAD v2.1 shows AF 1.59658e-05 from 4/250,536 alleles and zero homozygotes; the highest listed subpopulation AF is 3.52361e-05.
• The variant is absent from the supplied gnomAD-Canada v1.0 result.

Still missing

• No explicit RAD51C disease-prevalence, penetrance, or maximum credible allele-frequency calculation is provided in the case bundle, so the assessment relies on the very low observed frequency and the generic BS1 definition rather than a gene-specific quantitative threshold.

Sources

Reasoning

The applicable generic splice-impact path permits BP4 only when SpliceAI maximum delta is <0.1. The observed maximum delta is 0.949; therefore BP4 is not met and the same SpliceAI prediction is not double-counted across PP3 and BP4.

Rule

Generic ACMG/AMP fallback, non-missense splice-impact path: SpliceAI maximum delta <0.1 supports BP4; values from 0.1 through 0.2 meet neither PP3 nor BP4.

Evidence considered

• No RAD51C-specific BP4 rule or lookup assignment is present in the supplied CSPEC materials.
• SpliceAI Lookup reports a maximum delta score of 0.949, which is incompatible with the generic BP4 threshold of <0.1 for non-missense splice-path variants.

Sources

Reasoning

No exact-variant ClinVar benign or likely benign expert-panel assertion is present; therefore BP6 cannot be applied.

Rule

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

Evidence considered

• ClinVar variation 142762 is an exact match for c.904+5G>T, and the submission audit records zero expert-panel submissions; no expert-panel Benign or Likely benign assertion is present.

Sources

Reasoning

No reliable evidence establishes that RAD51C c.904+5G>T arose de novo in the proband. The available exact-variant publication reports the variant in hereditary breast/ovarian cancer pedigrees, but the extracted report does not document parental genotypes, maternity/paternity confirmation, or a de novo event.

Rule

Applicable RAD51C CSPEC v1.0 was reviewed first; no structured PS2-specific rule is supplied in the case bundle. Generic PS2 requires a confirmed de novo variant in an affected proband with confirmed maternity and paternity; those facts are unavailable.

Evidence considered

• PMID:20400964 reports RAD51C c.904+5G>T in two breast/ovarian cancer pedigrees but does not report parental testing or classify the variant as de novo in the extracted finding.
• PMID:34299313 reports the variant in two cancer patients but provides no parental testing or de novo analysis.
• The RAD51C CSPEC version 1.0 is applicable, but the supplied framework has no structured PS2-specific rule or case-level parental data; therefore no positive PS2 application is made.

Still missing

• Proband and both parental genotypes are not documented.
• Maternity and paternity confirmation are not documented.
• A validated de novo observation in a phenotypically consistent proband is not documented.

Sources

Reasoning

Exact-variant observations in affected hereditary breast/ovarian cancer cohorts were identified, but no exact-variant case-control enrichment statistic or sufficiently specified case/control counts were provided to establish PS4.

Rule

RAD51C CSPEC was identified but supplies no structured PS4 rule; generic ACMG/AMP PS4 requires a statistically increased prevalence in affected individuals compared with controls.

Evidence considered

• PMID:20400964 reports c.904+5G>T in two breast/ovarian cancer pedigrees among 1,100 German hereditary gynecological-cancer families, without a control comparison for this exact variant.
• PMID:34299313 reports the exact variant in two Italian cancer patients but does not provide an exact-variant case-control analysis.

Still missing

• An exact-variant case-control analysis with numerator/denominator data, ancestry-matched controls, and statistical result is needed.

Sources

Reasoning

The applicable RAD51C ClinGen HBOC Expert Panel specification (Version 1.0) is identified, but its extracted ruleset contains no usable PM3 rule text. The variant is reported in affected breast/ovarian cancer pedigrees, including two carriers in Meindl et al., but the available evidence does not establish biallelic inheritance, a pathogenic variant in trans, or phase with another RAD51C variant. The reported loss of the wild-type allele in tumors is somatic loss of heterozygosity and is not evidence that the germline variant is in trans with a pathogenic allele. RAD51C has both autosomal-dominant breast/ovarian cancer and autosomal-recessive Fanconi anemia contexts; no affected individual with Fanconi anemia and confirmed compound heterozygosity or homozygosity for this variant is documented in the reviewed evidence.

Evidence considered

• The governing source is the ClinGen Hereditary Breast, Ovarian and Pancreatic Cancer Expert Panel specification for RAD51C, Version 1.0; the case bundle identifies autosomal-dominant familial breast/ovarian cancer and autosomal-recessive Fanconi anemia disease contexts, but provides no extracted PM3-specific rule text.
• Meindl et al. (PMID:20400964) reports c.904+5G>T in two breast/ovarian cancer pedigrees and reports tumor loss of the wild-type allele; this is tumor loss of heterozygosity, not proof of a second germline pathogenic RAD51C allele in trans.
• Guglielmi et al. (PMID:34299313) reports the variant in a patient with triple-negative breast cancer and in a patient with ovarian serous carcinoma, without germline phase, trans/cis, or biallelic Fanconi-anemia evidence.

Still missing

• No documented germline pathogenic RAD51C variant in trans with c.904+5G>T.
• No confirmed homozygous or compound-heterozygous affected individual for the Fanconi anemia context.
• No family-based phasing data establishing the allelic relationship of c.904+5G>T to another RAD51C variant.

Sources

Reasoning

No unconfirmed de novo occurrence is documented for this variant. The available reports describe inherited-cancer cohort or pedigree observations without parental genotypes or a stated de novo interpretation.

Rule

Applicable RAD51C CSPEC v1.0 was reviewed first; no structured PM6-specific rule is supplied. Generic PM6 requires a suspected de novo variant when parental testing is unavailable; the case evidence does not establish such an occurrence.

Evidence considered

• PMID:20400964 describes two carriers in hereditary breast/ovarian cancer pedigrees and does not provide an unconfirmed de novo case.
• PMID:34299313 lists the exact variant in two cancer patients but reports no parental testing or de novo conclusion.
• The RAD51C CSPEC version 1.0 is applicable, but no structured PM6-specific rule or qualifying case-level de novo evidence is present in the bundle.

Still missing

• No proband-level de novo observation without parental testing is documented.
• No phenotype-consistent case report supplies sufficient pedigree and parental information for PM6.

Sources

Reasoning

The exact variant was reported in two hereditary breast/ovarian cancer pedigrees, but the available extraction does not provide countable meioses, affected-relative genotypes, unaffected-relative genotypes, phase, or a segregation statistic. Tumor loss of the wild-type allele is tumor evidence and is not itself germline family segregation.

Rule

Applicable RAD51C CSPEC v1.0 was reviewed first; no structured PP1-specific rule is supplied. Generic PP1 requires cosegregation of the variant with disease in affected family members, weighted by informative meioses and phenotype specificity; the available evidence is insufficient to quantify this.

Evidence considered

• PMID:20400964 reports the exact variant in two breast/ovarian cancer pedigrees, but the extracted finding does not provide sufficient affected-relative and unaffected-relative genotype data or informative meioses to apply PP1.
• PMID:20400964 reports loss of the wild-type allele in breast and ovarian tumor tissues from affected individuals; this is not equivalent to germline cosegregation.
• PMID:34299313 reports two unrelated patient observations and no segregation analysis.
• The RAD51C CSPEC version 1.0 is applicable, but its supplied rule payload contains no structured PP1 thresholds.

Still missing

• Number of informative meioses is not reported.
• Genotypes and phenotypes of affected relatives are not reported in a form permitting segregation assessment.
• Genotypes and phenotypes of unaffected relatives and ages at last follow-up are not reported.
• No cosegregation statistic or VCEP-defined PP1 threshold can be applied.

Sources

Reasoning

The reported tumors (breast cancer, including triple-negative breast cancer, and ovarian serous carcinoma) are within RAD51C-associated hereditary cancer contexts but are not a highly specific phenotype for a single-gene diagnosis; no individual-level phenotype workup or exclusion of alternative causes is supplied.

Rule

RAD51C CSPEC was identified but supplies no structured PP4 rule; generic ACMG/AMP PP4 requires a phenotype or family history highly specific for a disease with a single genetic etiology.

Evidence considered

• The RAD51C CSPEC disease contexts include familial ovarian cancer and hereditary breast carcinoma.
• PMID:34299313 reports c.904+5G>T in a patient with triple-negative breast cancer and in another patient with ovarian serous carcinoma; these phenotypes alone do not establish a highly specific single-gene presentation.

Still missing

• Detailed personal and family phenotype, tumor pathology, and assessment of alternative hereditary cancer etiologies are needed to evaluate phenotype specificity.

Sources

Reasoning

BS2 cannot be assessed from the available evidence. Population databases show zero homozygotes, but they do not establish that a heterozygous carrier is an unaffected healthy adult at an appropriate age, nor do they provide clinical phenotype, penetrance, or ascertainment information for carriers. No reliable healthy-adult carrier or unaffected homozygote observation is documented.

Rule

Generic ACMG/AMP BS2: observation in healthy adults, with disease age of onset and penetrance considered; database homozygote absence alone is insufficient.

Evidence considered

• The RAD51C CSPEC framework is incomplete and supplies no usable BS2-specific rule.
• Generic ACMG/AMP BS2 requires observation of the variant in healthy adults, with consideration of disease age of onset and penetrance (Richards et al., PMID:25741868).
• gnomAD v4.1 reports zero homozygotes and gnomAD v2.1 reports zero homozygotes, but these database counts do not document unaffected clinical status or provide a valid BS2 observation.

Still missing

• No phenotype-confirmed healthy adult heterozygote data are provided.
• No unaffected homozygote data are provided; gnomAD zero homozygotes is a count, not a BS2 observation.
• No age, ascertainment, disease-prevalence, or penetrance information is available to determine whether apparently unaffected carriers would challenge the disease model.

Sources

Reasoning

No validated benign functional assay for the exact RAD51C c.904+5G>T variant was identified. The available variant-specific experiments demonstrate abnormal splicing rather than preserved normal function, so they cannot support BS3.

Evidence considered

• The RAD51C VCEP specification is applicable to this case, but its extracted criterion payload is empty and does not provide a BS3-specific rule.
• PMID:20400964 reports exon 6 exclusion in patient RNA and in a RAD51C splicing reporter for the exact variant; these are abnormal-function results and provide no benign functional evidence.

Still missing

• A validated assay showing normal splicing or preserved RAD51C function for the exact c.904+5G>T allele, with appropriate controls and biologically relevant context, is not available in the reviewed evidence.

Sources

Reasoning

No reliable non-segregation evidence is provided. The case bundle contains no tested unaffected relatives who clearly lack the variant despite adequate age and phenotype assessment, and no family-level analysis showing lack of cosegregation.

Rule

Applicable RAD51C CSPEC v1.0 was reviewed first; no structured BS4-specific rule is supplied. Generic BS4 requires lack of segregation in affected family members or appropriately informative unaffected relatives, with phenotype and age considered; those data are unavailable.

Evidence considered

• PMID:20400964 reports affected pedigree observations and tumor loss of heterozygosity, but does not report informative unaffected relatives who tested negative or a non-segregation analysis.
• PMID:34299313 reports patient-level observations without family testing or non-segregation data.
• The RAD51C CSPEC version 1.0 is applicable, but no structured BS4-specific rule is supplied and no qualifying non-segregation evidence is available.

Still missing

• No genotyped, phenotypically informative unaffected relatives are documented.
• No family-level non-segregation analysis or sufficient follow-up ages are documented.
• No evidence permits exclusion of reduced penetrance, phenocopy, or ascertainment effects.

Sources

Reasoning

The applicable RAD51C ClinGen HBOC Expert Panel specification (Version 1.0) is identified, but its extracted ruleset contains no usable BP2 rule text. The reviewed reports do not document c.904+5G>T in trans with a known pathogenic variant in a recessive disease context, nor in cis with a pathogenic variant in a dominant disease context. The affected-patient observations and tumor loss-of-heterozygosity findings do not establish cis/trans phase and cannot be used to satisfy BP2.

Evidence considered

• The governing source is the ClinGen Hereditary Breast, Ovarian and Pancreatic Cancer Expert Panel specification for RAD51C, Version 1.0; the case bundle provides no extracted BP2-specific rule text.
• Meindl et al. (PMID:20400964) reports affected pedigrees and tumor loss of the wild-type allele for c.904+5G>T, but does not provide germline cis/trans phase with a pathogenic RAD51C variant.
• Guglielmi et al. (PMID:34299313) reports the variant in two affected patients, but gives no cis/trans phase or linked pathogenic RAD51C variant.

Still missing

• No confirmed phase result showing c.904+5G>T in trans with a pathogenic variant.
• No confirmed phase result showing c.904+5G>T in cis with a pathogenic variant.
• No family segregation or molecular phasing data sufficient to evaluate BP2.

Sources

Reasoning

One reported ovarian-cancer patient also had a non-coding ATM variant, but the available report does not establish that the ATM finding is pathogenic or fully explains the phenotype. No confirmed alternate molecular diagnosis is available.

Rule

RAD51C CSPEC was identified but supplies no structured BP5 rule; generic ACMG/AMP BP5 requires an alternate molecular basis for the disease to be established.

Evidence considered

• PMID:34299313 lists c.904+5G>T in an ovarian serous carcinoma patient together with another non-coding ATM variant, without evidence that the ATM variant is disease-causing or sufficient to explain the phenotype.

Still missing

• A confirmed pathogenic alternate molecular diagnosis that fully explains the observed phenotype is needed.

Sources