No variant summary was recorded for this case.
Five-step scale from benign to pathogenic. This variant is pathogenic. Select a step to read what it means.
Pathogenic.A pathogenic classification under the recorded framework.
Gene-level background from the recorded interpretation.
Words with a dotted underline have a short definition. Hover or tap them.RAD51D encodes a protein that works with other DNA-repair proteins to help accurately repair broken DNA through homologous recombination. Inherited changes in RAD51D increase the risk of ovarian cancer and may also increase breast cancer risk. RAD51D acts as a tumor-suppressor gene, and changes in it can contribute to chromosome instability and, rarely, resistance to chemotherapy in cancer cells.
The recorded framework is Not recorded. The criteria and strengths shown below come from the saved assessment; this page does not recalculate the classification.
The applied criteria and their strengths are shown as assessed under the stated framework.
Met, very strong: exon 4 frameshift Lys91IlefsTer13 truncates RAD51D at amino acid 103 of 329 and is expected to trigger NMD.
Met, strong: RAD51D K91fs reduced homologous-recombination repair and RAD51D stability across orthogonal assays with wild-type controls and three biological replicates.
Met at supporting strength: aggregate allele frequency was 1.35305e-05 in gnomAD v3.1 non-cancer genomes, below the 0.0001 PM2 threshold.
No benign criterion marked as applied in the recorded assessment.
Pathogenic: PVS1 (very strong) plus PS3 (strong) satisfy the generic ACMG/AMP Pathogenic combination; PM2 (supporting) is additionally met.
Evidence summaries and captured source records, where available. Missing entries should not be interpreted as negative findings.
gnomAD v4.1: 26 / 1,547,768 alleles. Recorded AF 1.67984e-05. Recorded homozygote count 0.
This variant is present in gnomAD v2.1 (AF= 5.56775e-05; MAF= 0.00557%, 14/251448 alleles, homozygotes = 0) and has highest observed frequency in the East Asian population (AF= 0.000761118; MAF= 0.07611%, 14/18394 alleles, homozygotes = 0); grpmax FAF= 0.00045935.
Not available in gnomAD-Canada v1.0.
Recorded SpliceAI maximum delta score: 0.151. Refer to the recorded PP3/BP4 rationale for interpretation.
13 Pathogenic · 2 Likely pathogenic · 1 pathogenic.
OncoKB identified variant-specific curated literature and context relevant to functional review; biological-effect context: Likely Loss-of-function; curated oncogenicity label: Likely Oncogenic.
This variant has not previously been reported in somatic cancers (COSMIC).
This variant does not lie in a statistically significant hotspot.
ACMG/AMP criteria under Not recorded, with the data and sources behind each one.
28 criteria with recorded statuses: 3 met · 6 not met · 9 not assessed · 10 not applicable. Open any row for the full reasoning, the rule and the evidence behind it.
No applicable RAD51D-specific VCEP specification was identified, so generic ACMG/AMP criteria were used.
The available evidence does not provide confirmed de novo testing, apparently de novo occurrence, affected-family cosegregation, or family-based non-segregation for this variant.
The exact variant-level observation in one cancer-naive control was considered but does not support PP1 or BS4.
Population assessment used the available non-cancer subsets where provided: gnomAD v2.1 non-cancer exomes and gnomAD v3.1 non-cancer genomes; gnomAD v4.1 all-comers remained the default comparison because no governing VCEP specified an alternative.
The variant is rare in aggregate population data, with no observed homozygotes; the maximum subgroup frequencies are ancestry-specific observations from small denominators and do not reach the supplied BA1 or BS1 thresholds.
No criteria match.
RAD51D loss of function is an established germline disease mechanism, and no RAD51D-specific CSPEC/VCEP or local PVS1 specification is available; therefore the ClinGen SVI generic PVS1 framework applies. NM_002878.4:c.270_271dup is a coding frameshift in exon 4 (exon c.264-345), producing NP_002869.3:p.(Lys91IlefsTer13). The predicted termination at amino acid 103 truncates the 329-amino-acid protein, eliminating approximately 69% of the normal sequence. Because the premature termination occurs in an internal exon, well upstream of the final exon-exon junction, the transcript is expected to be subject to nonsense-mediated decay rather than escape it. The affected exon is part of the established coding transcript, the consequence is not a canonical splice-site change, and no evidence indicates a non-critical distal region or repeat-region context that would warrant downgrading. Under PMC6185798, this supports full-strength PVS1, whose literal strength tier is very strong.
ClinGen SVI generic PVS1 framework (PMC6185798): established loss-of-function mechanism plus an early truncating frameshift in an internal, biologically relevant exon with predicted NMD supports full-strength PVS1.
PS3 is met at strong strength. No RAD51D-specific VCEP specification was available, so generic ACMG/AMP functional-evidence calibration applies. Ma et al. (PMID:33151324) evaluated the recurrent RAD51D K91fs variant, reported as 271_272insTA and equivalent to the case variant's normalized p.(Lys91IlefsTer13), in multiple orthogonal cell-based assays. Compared with RAD51D wild-type and empty-vector controls, K91fs reduced homologous-recombination repair in a DR-GFP reporter assay, impaired RAD51 recruitment to DNA-damage sites, failed to reduce gamma-H2AX foci after cisplatin treatment, reduced RAD51D protein stability, and increased olaparib sensitivity. The DR-GFP and gamma-H2AX experiments used three biological replicates with reported means and standard deviations and statistical testing; the assays directly interrogated RAD51D function and its established homologous-recombination role. This concordant, biologically relevant, multi-assay loss-of-function pattern supports strong PS3, although the publication uses the alternate insertion notation rather than the current transcript-level HGVS description.
Generic ACMG/AMP PS3: well-established functional studies supportive of a damaging effect; strong strength is justified here by concordant results from multiple orthogonal, biologically relevant assays with wild-type/empty-vector controls and replicated quantitative measurements.
No applicable gene-specific population specification was available, so the supplied generic PM2 supporting threshold of allele frequency <=0.0001 was used. The variant is rare but observed: gnomAD v2.1 non-cancer exomes AF 5.90942e-05, gnomAD v3.1 non-cancer genomes AF 1.35305e-05, and gnomAD v4.1 all-comers AF 1.67984e-05; all are below 0.0001. The maximum reported subgroup frequencies (0.00806452 in EAS_JPN for gnomAD v2.1 non-cancer and 0.00490196 in 1KG:JPT for gnomAD v3.1 non-cancer) are ancestry-specific small-subgroup observations and remain below the supplied PM2 threshold only for the aggregate dataset; therefore PM2 is supported by the aggregate population frequencies, with the subgroup values documented as a threshold artifact requiring caution. No homozygotes were observed.
Generic PM2: allele frequency <=0.0001, applied at supporting strength; governing VCEP precedence does not apply.
The exact variant is present in ClinVar as Pathogenic or Likely pathogenic in laboratory submissions, but the record has zero expert-panel submissions and is classified as criteria provided, single submitter. Under the specified rule, ordinary laboratory submissions, aggregate labels, and non-expert assertions cannot trigger PP5.
No applicable gene-specific population specification was available, so the supplied generic ACMG/AMP BA1 threshold of allele frequency >=0.05 was used. The highest reported frequency is below this threshold in every relevant dataset: gnomAD v2.1 non-cancer exomes total AF 5.90942e-05, maximum reported subpopulation AF 0.00806452 in EAS_JPN; gnomAD v3.1 non-cancer genomes total AF 1.35305e-05, maximum reported subpopulation AF 0.00490196 in 1KG:JPT; and gnomAD v4.1 all-comers total AF 1.67984e-05, maximum reported subpopulation AF 0.000583614. Therefore BA1 is not met.
Generic BA1: allele frequency >=0.05; governing VCEP precedence does not apply.
No applicable gene-specific population specification was available, so the supplied generic BS1 threshold of allele frequency >=0.01 was used. The highest reported frequency is 0.00806452 in the EAS_JPN subgroup of gnomAD v2.1 non-cancer exomes, below 0.01; the corresponding maximum in gnomAD v3.1 non-cancer genomes is 0.00490196 in 1KG:JPT, and gnomAD v4.1 all-comers has a maximum reported subpopulation AF of 0.000583614. The variant therefore does not meet BS1.
Generic BS1: allele frequency >=0.01; governing VCEP precedence does not apply.
BS2 requires observation of the variant in healthy adult individuals in a state inconsistent with the disease mechanism, classically homozygous for a dominant loss-of-function disorder. The variant has zero homozygotes in gnomAD v2.1, gnomAD v2.1 non-cancer exomes, gnomAD v3.1 non-cancer genomes, and gnomAD v4.1. A directly relevant study found the heterozygous truncating variant in one of 466 cancer-naive controls (0.21%), but this is not a homozygous observation and does not satisfy BS2. Accordingly, BS2 is not met.
Generic BS2: observation in healthy adults in a state inconsistent with the disease mechanism; no homozygous observations were found.
BS3 is not met. The available functional evidence does not demonstrate a normal or benign effect for this variant. Instead, PMID:33151324 directly assessed the equivalent RAD51D K91fs allele and found reduced homologous-recombination repair, impaired RAD51 recruitment, reduced RAD51D protein stability, and increased olaparib sensitivity relative to wild-type controls. No validated assay showing preserved RAD51D function with appropriate controls was identified; therefore BS3 cannot be applied.
Generic ACMG/AMP BS3: well-established functional studies showing no damaging effect on protein function or a validated functional assay; not satisfied when the available assay results show loss of function.
The exact variant has ClinVar laboratory Pathogenic and Likely pathogenic submissions, not an expert-panel Benign or Likely benign classification. The record has zero expert-panel submissions. Under the specified rule, ordinary submissions, aggregate labels, and non-expert assertions cannot trigger BP6.
PS2 cannot be assessed because the available evidence does not document a confirmed de novo occurrence in the proband, including verified maternity/paternity and parental testing results. No proband phenotype, parental genotype, or de novo confirmation details are provided for this variant.
Generic ACMG/AMP PS2: de novo occurrence in a patient with the disease and no family history, with maternity/paternity confirmed; strength depends on the quality and extent of de novo evidence.
No applicable RAD51D VCEP framework was available, so generic ACMG/AMP guidance was used. The literature includes gene-level case-control enrichment of inactivating RAD51D mutations (8/911 breast-ovarian cancer families versus 1/1,060 controls, P=0.01; PMID:21822267), but the exact c.270_271dup variant is not identified in that study. The exact variant was reported in 1/466 cancer-naive controls (0.21%; PMID:23372765), while a K91fs report found the variant in 8/325 Chinese triple-negative breast cancer patients but did not provide a qualifying matched control comparison in the extracted evidence (PMID:33151324). Therefore, an exact-variant case-control enrichment meeting PS4 cannot be established from the available evidence.
No applicable RAD51D-specific VCEP or CSPEC specification was available, so the generic ACMG/AMP framework was used. PM3 requires evidence that the variant is present in trans with a pathogenic or likely pathogenic variant in an affected individual, with strength determined by the quality and number of biallelic observations. The available case evidence does not document an affected proband with a second pathogenic RAD51D allele, phase, compound heterozygosity, or a confirmed biallelic genotype. PMID:23372765 reports the exact variant p.(Lys91Ilefs*13) in one cancer-naive control sample but provides no trans/cis or affected-proband observation. Therefore PM3 cannot be assessed from the available evidence.
generic_acmg_amp_PM3: pathogenic_variant_in_trans_with_the_tested_variant_in_an_affected_individual; strength depends_on_biallelic_observation_quality_and_number
PM6 cannot be assessed because no credible clinical report documents this variant as apparently de novo without confirmed maternity and paternity. The available variant-specific literature does not provide an apparently de novo case with sufficient family-testing detail.
Generic ACMG/AMP PM6: assumed de novo occurrence in a patient with the disease, without confirmation of maternity and paternity; evidence must still be credible and phenotype-consistent.
PP1 cannot be assessed because no informative affected relatives, meioses, genotype-positive affected relatives, or quantitative cosegregation data are documented. PMID:23372765 reports the exact protein-level variant p.(Lys91Ilefs*13) in one of 466 cancer-naive controls (0.21%), but this is not evidence of cosegregation in an affected family and cannot establish PP1.
Generic ACMG/AMP PP1: cosegregation with disease in multiple affected family members; strength may increase with the number of informative meioses and the quality of phenotype and genotype data.
No patient phenotype, clinical presentation, or disease-specific diagnostic information is provided in the available case evidence. Consequently, the specificity of the patient's phenotype for a RAD51D-associated disorder cannot be evaluated under generic ACMG/AMP PP4 guidance.
BS4 cannot be assessed because no affected family members reported to carry or not carry the variant, and no informative non-segregation analysis, are documented. PMID:23372765 reports the exact variant-level protein consequence in a cancer-naive control, but an isolated control observation is not evidence that the variant fails to segregate with disease within a family.
Generic ACMG/AMP BS4: lack of segregation in affected members of a family; isolated observations in unrelated controls do not establish BS4.
No applicable RAD51D-specific VCEP or CSPEC specification was available, so the generic ACMG/AMP framework was used. BP2 requires a documented allelic configuration involving the tested variant and a pathogenic variant, such as occurrence in trans in a fully penetrant dominant disorder or in cis in a recessive disorder, as applicable to the disease model. The available evidence contains no family-based or proband-level genotype data establishing cis or trans phase with another pathogenic RAD51D variant and no documented affected or unaffected individual in whom that configuration can be evaluated. PMID:23372765 reports the exact variant in a cancer-naive control sample but does not report a second pathogenic variant or phase. Therefore BP2 cannot be assessed from the available evidence.
generic_acmg_amp_BP2: variant_observed_in_trans_with_a_pathogenic_variant_for_a_fully_penetrant_dominant_disorder_or_in_cis_with_a_pathogenic_variant_for_a_recessive_disorder
No applicable RAD51D VCEP framework was available. The available evidence does not document an alternate pathogenic molecular explanation in the patient, a confirmed pathogenic variant in another gene, or a qualifying BP5-specific likelihood-ratio or other numeric measurement. Therefore BP5 cannot be assessed.
deterministic_prefilter: requires_missense
deterministic_prefilter: requires_missense
deterministic_prefilter: requires_stop_loss_or_nonrepeat_indel
deterministic_prefilter: requires_missense
deterministic_prefilter: requires_missense
PP3 is not applicable because NM_002878.4:c.270_271dup is normalized as NP_002869.3:p.(Lys91IlefsTer13), a frameshift variant that changes protein length. PP3 is restricted here to missense or splice-region/intronic variants; therefore SpliceAI and missense-predictor evidence must not be evaluated for this criterion.
deterministic_prefilter: requires_missense
deterministic_prefilter: requires_repeat_region_indel
BP4 is not applicable because NM_002878.4:c.270_271dup is normalized as NP_002869.3:p.(Lys91IlefsTer13), a frameshift variant that changes protein length. BP4 is restricted here to missense or splice-region/intronic variants; therefore SpliceAI and missense-predictor evidence must not be evaluated for this criterion.
deterministic_prefilter: requires_synonymous
Every interpreted variant in RAD51D, mapped onto the exon structure of NM_002878.4. Select a row to find it on the plot.
| Protein | Location | Classification | Link |
|---|
| East Asian26 / 44,550 | 0.00058 |
| East Asian14 / 18,394 | 0.00076 |
This variant is present in gnomAD v4.1 (AF= 1.67984e-05; MAF= 0.00168%, 26/1547768 alleles, homozygotes = 0) and has highest observed frequency in the East Asian population (AF= 0.000583614; MAF= 0.05836%, 26/44550 alleles, homozygotes = 0); grpmax FAF= 0.00040821.
This variant is present in gnomAD v2.1 (AF= 5.56775e-05; MAF= 0.00557%, 14/251448 alleles, homozygotes = 0) and has highest observed frequency in the East Asian population (AF= 0.000761118; MAF= 0.07611%, 14/18394 alleles, homozygotes = 0); grpmax FAF= 0.00045935.
Not available in gnomAD-Canada v1.0.
PP3 is not applicable because NM_002878.4:c.270_271dup is normalized as NP_002869.3:p.(Lys91IlefsTer13), a frameshift variant that changes protein length. PP3 is restricted here to missense or splice-region/intronic variants; therefore SpliceAI and missense-predictor evidence must not be evaluated for this criterion.
BP4 is not applicable because NM_002878.4:c.270_271dup is normalized as NP_002869.3:p.(Lys91IlefsTer13), a frameshift variant that changes protein length. BP4 is restricted here to missense or splice-region/intronic variants; therefore SpliceAI and missense-predictor evidence must not be evaluated for this criterion.
deterministic_prefilter: requires_synonymous
SpliceAI predicts no significant splice impact for this variant (max delta score = 0.15).
7 paper records listed. 2 cited in the classification.
A third truncating mutation, p.(Lys91Ilefs*13), was identified in one of 466 control samples (0.21%). All three of these mutations have previously been reported [3,4].
Location Results and Discussion, paragraph 1; Table 2, Exon 4
Context RAD51D mutation screening in 466 cancer-naive control DNA samples, following analysis of 245 unselected ovarian cancer cases and 1,060 familial breast/ovarian cancer index cases; variant positions were referenced to NM_002878.3 and NP_002869.
In a direct repeat GFP (DR-GFP) reporter assay, expression of both RAD51D V200X and K91fs variant resulted in a decrease in GFP-positive cell numbers compared to its WT counterpart. ... In summary, these results unambiguously show that RAD51D recurrent variants negatively impact DNA homologous recombination.
Location Results, 'RAD51D recurrent variants and DNA homologous recombination'; Figure 5C-F; Results, 'Effect of RAD51D recurrent variants on the expression level and stability of the RAD51D protein'; Figure 6A-B
Context FUSCCTNBC cohort of 325 Chinese patients with triple-negative breast cancer; RAD51D K91fs was tested in U2OS DR-GFP homologous-recombination assays, gamma-H2AX and RAD51-recruitment assays in HEK293T-derived systems, cycloheximide protein-stability assays, and olaparib viability assays in MDA-MB-231 and BT549 TNBC cell lines. Quantitative DR-GFP and gamma-H2AX results used three biological replicates with wild-type and empty-vector controls.
We identified eight inactivating RAD51D mutations in unrelated individuals from 911 breast-ovarian cancer families compared with one inactivating mutation identified in 1,060 controls (P = 0.01).
Abstract
Screenshots represent source records captured during evidence collection. Select one to view the recorded source content.
The evidence statements recorded by the interpretation pipeline. They have not been reinterpreted by the page.
This variant is present in gnomAD v4.1 (AF= 1.67984e-05; MAF= 0.00168%, 26/1547768 alleles, homozygotes = 0) and has highest observed frequency in the East Asian population (AF= 0.000583614; MAF= 0.05836%, 26/44550 alleles, homozygotes = 0); grpmax FAF= 0.00040821.
This variant is present in gnomAD v2.1 (AF= 5.56775e-05; MAF= 0.00557%, 14/251448 alleles, homozygotes = 0) and has highest observed frequency in the East Asian population (AF= 0.000761118; MAF= 0.07611%, 14/18394 alleles, homozygotes = 0); grpmax FAF= 0.00045935.
Not available in gnomAD-Canada v1.0.
This variant has been reported in ClinVar as Pathogenic (13 clinical laboratories) and as Likely pathogenic (2 clinical laboratories) and as pathogenic (1 clinical laboratory). (ClinVarID = 239394)
SpliceAI predicts no significant splice impact for this variant (max delta score = 0.15).
OncoKB identified variant-specific curated literature and context relevant to functional review; biological-effect context: Likely Loss-of-function; curated oncogenicity label: Likely Oncogenic.
This variant does not lie in a statistically significant hotspot. This variant has not previously been reported in somatic cancers (COSMIC).
This variant does not lie in a statistically significant hotspot.