RB1 c.1390-17T>A, p.?

Reasoning

The intronic variant has a maximum SpliceAI delta score of 0.024, below the calibrated BP4 threshold of 0.1. Apply BP4 at supporting strength. This is the sole computational prediction code applied for this variant; PP3 is not also applied.

Rule

Generic ACMG non-missense computational pathway: SpliceAI maximum delta <0.1 supports BP4; score 0.024 meets this threshold.

Evidence considered

• SpliceAI Lookup for NM_000321.3:c.1390-17T>A reports maximum delta 0.024 (DS_AG 0.001, DS_AL 0.019, DS_DG 0.005, DS_DL 0.024).
• Generic ACMG operating rules direct intronic non-canonical splice-position variants to one SpliceAI-based PP3/BP4 pathway and assign BP4 supporting for maximum delta below 0.1.
• Walker et al., 2023, Using the ACMG/AMP framework to capture evidence related to predicted and observed impact on splicing (PMID:37352859), calibrated SpliceAI >=0.2 for PP3 and <=0.1 for BP4 for variants outside donor/acceptor ±1,2 positions.

Sources

Reasoning

RB1 loss of function is an established germline disease mechanism, but NM_000321.3:c.1390-17T>A is a non-canonical intronic SNV rather than a predicted null variant. It is 17 bases upstream of the exon boundary, has no defined protein consequence, and SpliceAI maximum delta is 0.024; therefore, there is no evidence that it causes aberrant splicing or a loss-of-function transcript eligible for PVS1.

Rule

Generic ClinGen SVI PVS1 framework: apply only when the variant is predicted to cause loss of function in a gene for which loss of function is an established disease mechanism.

Evidence considered

• No RB1 VCEP/CSPEC or local gene-specific PVS1 specification was retrieved, so the generic ClinGen SVI PVS1 framework was used.
• The MANE/RefSeq transcript annotation identifies c.1390-17T>A as intronic (exon 14i) with predicted protein consequence p.?, not a nonsense, frameshift, or canonical ±1/2 splice-site variant.
• The generic PVS1 assessment identifies the variant as bucket 'other' and does not apply the generic PVS1 framework to it.
• SpliceAI scores are DS_AG 0.001, DS_AL 0.019, DS_DG 0.005, and DS_DL 0.024 (maximum 0.024), indicating no predicted significant splice impact.

Still missing

• No variant-specific RNA study is available to directly assess splicing; RNA evidence would be required to reconsider PVS1 if a clinically consequential aberrant transcript were demonstrated.

Sources

Reasoning

The variant is not absent from population databases: it is observed in gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada. Thus it does not satisfy the literal generic ACMG/AMP PM2 requirement of absence from population controls. The low frequency may still be compatible with disease risk, but no RB1-specific rare-frequency threshold is available to replace the generic absence rule.

Rule

Generic ACMG/AMP PM2: variant absent from population databases/controls; rare presence without a validated disease-specific threshold is insufficient for the literal generic rule.

Evidence considered

• gnomAD v2.1 identifies the variant at 24/142264 alleles (AF 0.0001687).
• gnomAD v4.1 identifies the variant at 137/1277350 alleles (AF 0.000107253).
• gnomAD-Canada identifies the variant at 8/18356 alleles (AF 0.0004358248).
• All three datasets report 0 homozygotes, but presence of the variant in population controls means the generic PM2 absence criterion is not met.
• The gnomAD-Canada record flags the locus as low complexity region, so the observed frequency should be interpreted with technical caution; this does not make the variant absent.

Still missing

• No applicable RB1-specific VCEP/ClinGen PM2 threshold was available.

Sources

Reasoning

NM_000321.3:c.1390-17T>A is an intronic single-nucleotide variant with protein consequence p.?; it is not an in-frame deletion/insertion or stop-loss variant and no protein-length change is demonstrated. PM4 is therefore not met.

Rule

Generic ACMG/AMP PM4: protein length changes caused by in-frame deletions/insertions or stop-loss variants support PM4; this variant has neither demonstrated consequence.

Evidence considered

• Transcript normalization reports the protein consequence as NP_000312.2:p.? and the variant is intronic, 17 bases upstream of the exon boundary.
• SpliceAI maximum delta is 0.024, with no predicted significant splice impact; no altered transcript or protein-length change is reported.

Still missing

• No RNA study is available to exclude an unpredicted splice effect; however, no observed or predicted protein-length change supports PM4.

Sources

Reasoning

The variant is intronic (c.1390-17T>A), so the generic non-missense SpliceAI pathway applies. Its maximum SpliceAI delta score is 0.024, below the calibrated PP3 threshold of 0.2; PP3 is therefore not met.

Rule

Generic ACMG non-missense computational pathway: SpliceAI maximum delta >0.2 supports PP3; score 0.024 does not meet this threshold.

Evidence considered

• SpliceAI Lookup for NM_000321.3:c.1390-17T>A reports maximum delta 0.024 (DS_AG 0.001, DS_AL 0.019, DS_DG 0.005, DS_DL 0.024).
• Generic ACMG operating rules direct intronic non-canonical splice-position variants to one SpliceAI-based PP3/BP4 pathway, without combining splice and missense predictors.
• Walker et al., 2023, Using the ACMG/AMP framework to capture evidence related to predicted and observed impact on splicing (PMID:37352859), calibrated SpliceAI >=0.2 for PP3 and <=0.1 for BP4 for variants outside donor/acceptor ±1,2 positions.

Sources

Reasoning

ClinVar confirms the exact variant but records no expert-panel submission. Its reported assertions are non-expert single-submitter Uncertain significance and Likely benign; therefore PP5 cannot be applied.

Rule

Apply PP5 only to an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification; do not use laboratory submissions or aggregate ClinVar labels.

Evidence considered

• The ClinVar submission audit confirms an exact transcript-HGVS match and zero expert-panel submissions; both available submissions are non-expert single-submitter records.

Still missing

• An exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification would be required.

Sources

Reasoning

The variant does not meet the generic BA1 stand-alone benign threshold. Its maximum observed allele frequency is 0.0004358 (0.04358%) in gnomAD-Canada, with 0.0001073 (0.01073%) in gnomAD v4.1 and 0.0001687 (0.01687%) in gnomAD v2.1, all far below 5%.

Rule

Generic ACMG/AMP BA1: stand-alone benign when population allele frequency is greater than 5%.

Evidence considered

• gnomAD v2.1 reports 24/142264 alleles, AF 0.0001687, with 0 homozygotes.
• gnomAD v4.1 reports 137/1277350 alleles, AF 0.000107253, with 0 homozygotes.
• gnomAD-Canada reports 8/18356 alleles, AF 0.0004358248, with 0 homozygotes; the record flags the locus as low complexity region.
• Under the generic ACMG/AMP framework, BA1 requires an allele frequency greater than 5%; the observed frequencies are well below that threshold.

Sources

Reasoning

BP3 concerns in-frame deletions or insertions in a repetitive region without known function. This is an intronic SNV, not a protein-coding in-frame deletion or insertion; no qualifying repeat-region protein event is present.

Rule

Generic ACMG/AMP BP3: an in-frame deletion or insertion in a repetitive region without known function supports BP3; the molecular event here is an intronic SNV.

Evidence considered

• The variant-level assessment records NM_000321.3:c.1390-17T>A as a non-canonical intronic variant with unknown consequence and no defined protein change.
• The bundle provides no protein-coding in-frame indel or repeat-region annotation applicable to BP3.

Sources

Reasoning

ClinVar confirms the exact variant but records no expert-panel submission. The non-expert Likely benign laboratory assertion cannot trigger BP6.

Rule

Apply BP6 only to an exact-variant ClinVar expert-panel Benign or Likely benign classification; do not use laboratory submissions or aggregate ClinVar labels.

Evidence considered

• The ClinVar submission audit confirms an exact transcript-HGVS match and zero expert-panel submissions; the Likely benign assertion is from a non-expert single submitter.

Still missing

• An exact-variant ClinVar expert-panel Benign or Likely benign classification would be required.

Sources

Reasoning

No case-specific proband phenotype, parental genotypes, validated de novo status, or maternity/paternity confirmation is provided in the evidence bundle. The variant is therefore not assessable for PS2.

Rule

Generic ACMG/AMP PS2 requires a confirmed de novo occurrence in a patient with the relevant phenotype, with maternity and paternity confirmed; the available case evidence does not establish these elements.

Evidence considered

• The consolidated case-evidence bundle contains no case-specific segregation, parental-testing, meiosis, or de novo observations for NM_000321.3:c.1390-17T>A.
• No fetched paper extraction reports this exact variant in a de novo case or provides parental testing for it.

Still missing

• Proband phenotype and diagnostic context
• Genotypes of both biological parents
• Confirmation that the variant is absent in both parents
• Maternity/paternity confirmation or equivalent family-relationship evidence

Reasoning

No variant-specific, validated functional assay demonstrating a damaging effect was identified for NM_000321.3:c.1390-17T>A. The variant is intronic, but the available SpliceAI result is computational prediction rather than functional evidence and cannot establish PS3. The extracted literature includes only a general splicing-validation methodology paper that does not mention this RB1 variant.

Rule

Generic ACMG/AMP 2015: PS3 requires well-established functional studies supportive of a damaging effect; strength calibration requires validated methodology, appropriate controls, and biologically relevant quantitative evidence.

Evidence considered

• The case bundle reports no ClinVar criterion-lead submission, expert-panel submission, or validated PMID providing functional evidence for this exact variant.
• SpliceAI reports a maximum delta score of 0.024, but this is an in-silico prediction and is not used as PS3 functional evidence.
• PMID:31275557 was reviewed through its pre-computed extraction; it describes a general RNA-seq splicing-validation method but explicitly does not report RB1:c.1390-17T>A or any variant-specific result.

Still missing

• Variant-specific RNA or protein assay is needed to establish abnormal RB1 splicing or loss of function.
• Assay controls, replicate data, quantitative effect size, and biological relevance for this variant are unavailable.

Sources

Reasoning

No exact-variant case-control, cohort-enrichment, or statistically evaluable affected-versus-control evidence was identified. The two literature sources screened for PS4 do not report RB1 c.1390-17T>A.

Rule

Generic ACMG/AMP fallback: PS4 requires statistically increased prevalence in affected individuals compared with controls; no RB1-specific VCEP specification was retrieved.

Evidence considered

• PMID:25356965 was screened for PS4; its extraction states that it contains no variant-specific RB1 clinical or case information.
• PMID:35802134 was screened for PS4; its extraction states that RB1 c.1390-17T>A is not reported and that no variant-specific case data are provided.

Still missing

• An independent, phenotype-defined case series or case-control study reporting RB1 c.1390-17T>A counts in affected individuals and appropriate controls is needed.

Sources

Reasoning

No case report or family study in the evidence bundle establishes this variant as apparently de novo without parental testing. PM6 cannot be applied.

Rule

Generic ACMG/AMP PM6 is reserved for an apparently de novo variant when parental testing is not performed or unavailable; no such variant-specific evidence is present.

Evidence considered

• The consolidated case-evidence bundle contains no variant-specific report of apparent de novo occurrence for NM_000321.3:c.1390-17T>A.
• The available paper extraction for the fetched literature does not provide a case-specific de novo observation for this variant.

Still missing

• A case-specific report of the variant in an affected proband
• Evidence that the variant was apparently de novo
• Documentation of whether parental testing was performed or unavailable

Reasoning

No affected or unaffected relatives, informative meioses, variant-positive affected relatives, or variant-negative unaffected relatives are documented for this case. Segregation strength cannot be assessed.

Rule

Generic ACMG/AMP PP1 requires consistent cosegregation of the variant with disease in affected family members, interpreted using informative meioses and phenotype specificity; none are available here.

Evidence considered

• The consolidated case-evidence bundle contains no family pedigree, affected-relative testing, unaffected-relative testing, or segregation count for NM_000321.3:c.1390-17T>A.
• No paper extraction provides variant-specific cosegregation evidence for this variant.

Still missing

• Pedigree and number of informative meioses
• Genotypes of affected relatives
• Phenotype status and genotypes of unaffected relatives
• Phenotype consistency and segregation analysis across the family

Reasoning

No proband phenotype, diagnostic evaluation, or evidence that the phenotype is highly specific for RB1-related disease is available in the consolidated evidence bundle.

Rule

Generic ACMG/AMP fallback: PP4 requires a patient phenotype or family history highly specific for a disease with a single genetic etiology; no RB1-specific VCEP specification was retrieved.

Still missing

• Proband-level phenotype and diagnostic work-up, including evidence of a highly specific RB1-related presentation and exclusion of relevant alternatives, are needed.

Reasoning

Population data show the variant is rare, with maximum observed AF 0.0004358, but no applicable RB1-specific VCEP framework or validated disease-specific maximum credible allele frequency is available in this case. Disease prevalence, penetrance, and the relevant inheritance/phenotype model are therefore insufficiently specified to determine whether the observed frequency is greater than expected for the disorder.

Rule

Generic ACMG/AMP BS1: allele frequency is greater than expected for the disorder, using a disease-appropriate prevalence/penetrance threshold.

Evidence considered

• The highest reported overall frequency is 0.0004358248 in gnomAD-Canada (8/18356 alleles); gnomAD v4.1 reports AF 0.000107253 (137/1277350 alleles), and gnomAD v2.1 reports AF 0.0001687 (24/142264 alleles).
• The highest ancestry-specific gnomAD v4.1 frequency is 0.000279174 in the Middle Eastern group (1/3582 alleles); this is a single allele in a small subgroup and is not a validated disease-specific threshold.
• The gnomAD-Canada record flags the locus as low complexity region, a potential population-data quality artifact requiring caution.
• Generic BS1 requires a population frequency greater than expected for the disorder; the case bundle does not provide the RB1 prevalence/penetrance model or a validated maximum credible AF needed for that comparison.

Still missing

• No applicable RB1-specific VCEP/ClinGen specification was available.
• No validated RB1 disease-specific BS1 threshold incorporating prevalence and penetrance was provided.

Sources

Reasoning

No homozygotes are reported in gnomAD v2.1, gnomAD v4.1, or gnomAD-Canada, and the population records do not establish that any carriers are healthy adults with adequate age-related phenotype assessment. Therefore, the required observation of the variant in healthy individuals cannot be evaluated from the available data.

Rule

Generic ACMG/AMP BS2: observed in an appropriate number of healthy adult individuals, including homozygous/hemizygous observations where applicable.

Evidence considered

• gnomAD v2.1 reports 0 homozygotes among 24 variant alleles.
• gnomAD v4.1 reports 0 homozygotes among 137 variant alleles.
• gnomAD-Canada reports 0 homozygotes among 8 variant alleles.
• Population allele counts alone do not demonstrate that carriers are unaffected healthy adults, particularly for a condition with possible age-dependent or incomplete penetrance.

Still missing

• No documented healthy adult carrier phenotype or age/penetrance assessment was provided.
• No homozygous or hemizygous observations are present in the available population datasets.

Sources

Reasoning

No variant-specific, validated functional assay demonstrating preserved RB1 function was identified for NM_000321.3:c.1390-17T>A. Absence of a damaging computational prediction is not equivalent to a well-established benign functional assay, and the available literature does not provide variant-specific experimental results. Therefore BS3 cannot be assigned.

Rule

Generic ACMG/AMP 2015: BS3 requires well-established functional studies showing no damaging effect on protein function or splicing; strength calibration requires validated methodology, appropriate controls, and biologically relevant evidence.

Evidence considered

• The case bundle contains no variant-specific functional assay, validated benign functional result, or expert functional assessment for this exact variant.
• SpliceAI reports a low predicted splice-impact score, but computational prediction cannot satisfy BS3.
• PMID:31275557 was reviewed through its pre-computed extraction; it provides general splicing-assay methodology only and does not report a benign functional result for RB1:c.1390-17T>A.

Still missing

• Variant-specific functional testing with appropriate positive and negative controls is needed to establish preserved RB1 function.
• Quantitative assay performance, replicate results, and assay relevance to RB1 biology are unavailable.

Sources

Reasoning

No unaffected, appropriately evaluated relatives who are confirmed not to carry the variant are documented. There is no informative non-segregation evidence for BS4.

Rule

Generic ACMG/AMP BS4 requires lack of segregation in affected families, typically demonstrated by appropriately assessed unaffected relatives who do not carry the variant; this evidence is absent.

Evidence considered

• The consolidated case-evidence bundle contains no variant-specific non-segregation data, unaffected-relative testing, or family-based exclusion evidence for NM_000321.3:c.1390-17T>A.
• No paper extraction reports appropriately phenotyped unaffected relatives who lack this exact variant.

Still missing

• Unaffected relatives with reliable phenotype assessment
• Genotypes demonstrating absence of the variant in those relatives
• Adequate age and clinical evaluation to regard relatives as unaffected
• Family structure and informative meioses

Reasoning

Under generic ACMG/AMP, BP2 may apply when a variant is observed in trans with a pathogenic variant for a dominant disorder, or in cis with a pathogenic variant for a recessive disorder. RB1 retinoblastoma is documented as autosomal dominant, but the case bundle contains no second RB1 pathogenic variant and no phase information for NM_000321.3:c.1390-17T>A. BP2 therefore cannot be assessed.

Rule

Generic ACMG/AMP BP2: observed in trans with a pathogenic variant for a dominant disorder, or in cis with a pathogenic variant for a recessive disorder.

Evidence considered

• PMID:35802134 lists RB1-associated retinoblastoma as autosomal dominant (AD), establishing the relevant inheritance context for BP2.
• No source in the case bundle reports NM_000321.3:c.1390-17T>A with another pathogenic RB1 variant, in trans or in cis.
• No affected-proband genotype, parental testing, family segregation, or phase evidence was available for this criterion.

Still missing

• A second pathogenic or likely pathogenic RB1 variant in the same individual is not documented.
• Phase information establishing trans configuration is not documented.
• Affected-proband and family segregation data are not documented.

Sources

Reasoning

No affected individual with this exact variant and a separate, independently explanatory molecular diagnosis is documented in the consolidated evidence bundle.

Rule

Generic ACMG/AMP fallback: BP5 requires observation of the variant in an affected individual with an alternate molecular basis for disease; no RB1-specific VCEP specification was retrieved.

Still missing

• A documented affected individual carrying RB1 c.1390-17T>A together with an independently sufficient molecular diagnosis explaining the phenotype is needed.