NOTCH3 c.3278G>T, p.(Cys1093Phe)

Reasoning

p.(Cys1093Phe) alters a cysteine residue that falls within the region of NOTCH3 corresponding to the tandem EGF-like repeat domains (exon 20 of NM_000435.2, within the extracellular EGFr array where CADASIL disease-causing missense variants cluster). Case-gathered literature on NOTCH3 germline disease mechanism explicitly states that CADASIL is caused by pathogenic NOTCH3 variants, most commonly heterozygous cysteine-altering variants, establishing the cysteine residues of the EGF-like repeat domains as a well-characterized, mutation-dense functional region without benign missense variation typically reported there. No official ClinGen NOTCH3 VCEP specification (which formally codifies an 'odd number of cysteines within an EGFr domain' rule at Strong strength) was retrievable in this case bundle (cspec.found = false, gene_vcep_dir = null), so the criterion is applied at the generic ACMG Moderate default rather than any VCEP-specific strength.

Evidence considered

• PMID:41947304 (An in trans NOTCH3 loss-of-function variant modifies disease severity in CADASIL) abstract states: 'CADASIL is caused by pathogenic NOTCH3 variants, most commonly heterozygous cysteine-altering variants,' supporting cysteine residues within the EGF-like repeat domains as the principal disease-associated structural feature of NOTCH3.
• Variant p.(Cys1093Phe) (NP_000426.2) removes a cysteine residue, consistent with this established cysteine-altering pathogenic mechanism region.
• No official NOTCH3 CSPEC/VCEP document was retrieved in this case (cspec.found=false), so VCEP-specific strength escalation could not be confirmed.

Still missing

• No ClinGen NOTCH3 VCEP CSPEC document was available in this case bundle to confirm official domain boundaries or an upgraded (Strong) odd-cysteine rule for this specific EGFr repeat; strength is capped at generic-ACMG Moderate pending VCEP confirmation.
• Exact EGFr repeat number containing residue 1093 was not independently confirmed from a structural/domain-mapping source in this bundle.

Sources

Reasoning

PM2 is met at supporting strength under the generic ACMG/AMP framework. NM_000435.2:c.3278G>T (p.Cys1093Phe) is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, with no reported allele count, frequency, or ancestry-specific observation.

Rule

Generic ACMG/AMP PM2: absence or extreme rarity in population databases supports PM2 when the database has adequate representation and coverage; applied here as supporting strength because the variant is absent from multiple population datasets.

Evidence considered

• The case bundle reports the exact variant as absent from gnomAD v2.1.
• The case bundle reports the exact variant as absent from gnomAD v4.1.
• The case bundle reports the exact variant as absent from gnomAD-Canada v1.0.
• No allele count, allele frequency, homozygote count, or ancestry-specific observation was reported in these datasets.
• The variant is a rare missense substitution in NOTCH3; the available population evidence supports rarity but does not by itself establish disease mechanism or penetrance.

Sources

Reasoning

NOTCH3 is a gene in which missense variants are a common and well-established mechanism of disease. Case literature on NOTCH3 germline mechanism confirms that classic CADASIL is caused predominantly by heterozygous cysteine-altering missense variants in the EGF-like repeat domains, while loss-of-function (frameshift/homozygous LoF) NOTCH3 variants are reported as distinct, less common, and phenotypically divergent presentations rather than the predominant mechanism. This supports the generic PP2 criterion (missense variants are a common disease mechanism in this gene).

Evidence considered

• PMID:41947304 abstract: 'CADASIL is caused by pathogenic NOTCH3 variants, most commonly heterozygous cysteine-altering variants,' identifying missense (cysteine-altering) change as the predominant NOTCH3 disease mechanism.
• PMID:41196431 abstract describes homozygous loss-of-function NOTCH3 variants as 'expanding the spectrum beyond CADASIL,' i.e., a distinct and less common presentation relative to the classic heterozygous missense CADASIL mechanism, reinforcing that missense change is the predominant/common disease mechanism for this gene.

Still missing

• A gene-specific missense Z-score or formal ExAC/gnomAD constraint metric was not available in the case bundle to further quantify missense intolerance.

Sources

Reasoning

NM_000435.2:c.3278G>T (p.(Cys1093Phe)) is a missense variant with no gene-specific VCEP/CSPEC available for NOTCH3 (framework_mode=generic_acmg, cspec.found=false), so generic ACMG/AMP calibrated in-silico thresholds apply. REVEL score = 0.962, which exceeds the ClinGen SVI calibration threshold of >=0.932 for PP3_Strong (Pejaver et al., calibration of computational tools for missense pathogenicity classification and ClinGen recommendations for PP3/BP4, PMID:36413997). BayesDel score (0.586327) was also computed but has no verified published calibration threshold available to this pipeline and is therefore not used to support or oppose PP3/BP4. SpliceAI max delta score = 0.018, far below any splice-impact threshold, indicating no predicted splicing disruption for this exonic missense substitution; this splice-prediction result does not itself invoke PP3 (no splice gain/loss predicted) and is reported for completeness under this group's splice-impact remit.

Rule

generic_acmg_PP3_BP4_missense_computational

Evidence considered

• REVEL score 0.962 for NM_000435.2:c.3278G>T exceeds the >=0.932 PP3_Strong threshold from the ClinGen SVI computational-evidence calibration (Pejaver et al., PMID:36413997), supporting PP3 at Strong strength.
• BayesDel score 0.586327 was retrieved but excluded from PP3/BP4 adjudication because no named/verified publication with a calibrated BayesDel threshold is available to this pipeline.
• SpliceAI max delta score 0.018 (DS_AG 0.005, DS_AL 0.002, DS_DG 0.018, DS_DL 0.002) shows no predicted splice-site gain or loss for this variant, ruling out a splice-disruption mechanism as an alternative PP3 basis.

Sources

Reasoning

PP5 is not met. The exact variant NM_000435.2:c.3278G>T is reported as absent from ClinVar, with no exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic assertion. Ordinary or aggregate assertions would not qualify, and none are present here.

Rule

PP5 may be applied only for an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification; no such classification is available.

Evidence considered

• ClinVar exact-variant search returned no result for NM_000435.2:c.3278G>T; therefore there is no expert-panel Pathogenic/Likely pathogenic classification eligible to support PP5.

Sources

Reasoning

BA1 is not met. The variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, so no population allele frequency reaches a stand-alone benign threshold.

Rule

Generic ACMG/AMP population assessment: BA1 requires an allele frequency sufficiently high for a stand-alone benign assertion; absence from the available population datasets does not meet BA1.

Evidence considered

• The case bundle reports NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe) as absent from gnomAD v2.1.
• The case bundle reports the variant as absent from gnomAD v4.1.
• The case bundle reports the variant as absent from gnomAD-Canada v1.0.
• Because no allele frequency was observed, the BA1 stand-alone benign frequency threshold cannot be reached.

Sources

Reasoning

BS1 is not met. No population allele frequency was observed in gnomAD v2.1, gnomAD v4.1, or gnomAD-Canada v1.0, so the frequency is not higher than expected for the disease.

Rule

Generic ACMG/AMP BS1: apply only when the observed allele frequency is greater than expected for the disease; no observed frequency is available here.

Evidence considered

• The case bundle reports NM_000435.2:c.3278G>T as absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0.
• No ancestry-specific or overall allele count/frequency was reported for this variant in the available population datasets.
• An absent variant cannot satisfy BS1, which requires an allele frequency greater than expected for the relevant disease.

Sources

Reasoning

BP1 applies to genes where missense variation is not a typical/established mechanism of disease and truncating variants predominate. NOTCH3 is the opposite: missense (specifically cysteine-altering) variants are the predominant, well-established CADASIL disease mechanism per case literature. Therefore BP1 does not apply and is scored not_met for this missense variant.

Evidence considered

• PMID:41947304 abstract identifies heterozygous cysteine-altering missense variants as the most common cause of CADASIL, indicating missense variation is an established, not a benign/tolerated, disease mechanism for NOTCH3, so BP1 is not applicable/not met.

Sources

Reasoning

The dominant calibrated computational predictor for this missense variant (REVEL = 0.962) strongly supports pathogenicity (see PP3), so BP4 cannot be met on the missense path. SpliceAI predicts no splice-site alteration (max delta = 0.018), but this null splicing result is not treated as independent BP4 evidence for a variant whose primary mechanism under evaluation is missense substitution with a strong pathogenic REVEL score; using the splice-null result to invoke BP4 here would improperly counter the concordant, calibrated missense signal from the same variant rather than reflect a genuinely benign-leaning computational prediction.

Rule

generic_acmg_PP3_BP4_missense_computational

Evidence considered

• REVEL score 0.962 is far above any benign-leaning BP4 threshold, so the missense in-silico path does not support BP4.
• SpliceAI max delta score 0.018 shows no splice-altering effect, but is not independently applied as BP4 evidence given the strong concordant pathogenic missense signal (PP3) for this same variant.

Sources

Reasoning

BP6 is not met. The exact variant NM_000435.2:c.3278G>T is reported as absent from ClinVar, with no exact-variant ClinVar expert-panel Benign or Likely benign assertion. Ordinary or aggregate assertions would not qualify, and none are present here.

Rule

BP6 may be applied only for an exact-variant ClinVar expert-panel Benign or Likely benign classification; no such classification is available.

Evidence considered

• ClinVar exact-variant search returned no result for NM_000435.2:c.3278G>T; therefore there is no expert-panel Benign/Likely benign classification eligible to support BP6.

Sources

Reasoning

PS1 requires a prior pathogenic classification for a different nucleotide change producing the identical amino acid change (p.(Cys1093Phe)). ClinVar returned no result for this variant (search_status = no_result), and no other source in the bundle provides any prior classified variant at this exact codon/amino-acid change. No same-amino-acid-change comparator could be identified.

Evidence considered

• ClinVar scrape returned search_status: no_result / found: false for NOTCH3 c.3278G>T; evidence_sentence: 'This variant is absent from ClinVar.'

Still missing

• ClinVar has no record for this variant or codon; no source in this case bundle documents another nucleotide change producing the identical p.(Cys1093Phe) substitution with an established classification.

Sources

Reasoning

No proband-specific parental testing, maternity/paternity confirmation, or de novo observation is provided in the case-evidence bundle. PS2 cannot be assessed.

Rule

Generic ACMG/AMP PS2: de novo occurrence in a patient with the disease and no family history, with confirmed maternity and paternity; assess strength according to the quality and extent of de novo evidence.

Evidence considered

• The consolidated case-evidence bundle contains no proband, parental testing, maternity/paternity, or de novo evidence for NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe).
• No applicable NOTCH3-specific VCEP/CSPEC framework was retrieved; generic ACMG/AMP fallback therefore applies.

Still missing

• Documented parental testing showing the variant absent in both biologic parents
• Confirmation of maternity and paternity
• Relevant phenotype and family clinical context

Reasoning

No gene-specific VCEP/CSPEC framework is available for NOTCH3 (cspec.found=false, final_classification_framework.found=false; generic ACMG fallback in effect). The literature search returned zero PMIDs (literature_pass.json: total_pmids=0) and no functional/in vitro assay data (e.g., Notch signaling reporter, cleavage, trafficking, or aggregation assays) for NM_000435.2:c.3278G>T (p.Cys1093Phe) was identified in evidence.json, vcep_materials.json, or the PVS1 gene-context literature review. No validated functional assay evidence exists in the case bundle to support or refute PS3.

Still missing

• No published functional/in vitro assay (e.g., Notch3 signaling, ligand-binding, cleavage, or EGF-domain folding/aggregation assay) specific to p.Cys1093Phe was located in the available sources.
• No NOTCH3 VCEP-specific PS3 assay framework (validated assay list, OddsPath thresholds) was available in this case bundle to calibrate strength even if an assay were found.

Reasoning

PS4 was not assessed because the case-evidence bundle contains no affected-case count, case-control enrichment data, or variant-specific clinical cohort evidence for NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe).

Rule

Generic ACMG/AMP PS4: use statistically significant enrichment of the variant in affected individuals compared with controls; no qualifying enrichment evidence is available.

Evidence considered

• The consolidated case-evidence bundle reports no literature PMIDs and provides no affected-case or case-control dataset for this exact variant.

Still missing

• Variant-specific affected-case counts and an appropriate control cohort are unavailable.
• No case-control enrichment analysis or statistically interpretable phenotype prevalence data are available.

Reasoning

No affected-proband observation, second pathogenic or likely pathogenic allele, phase result, or family segregation data is present for NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe). The case bundle identifies no applicable NOTCH3 VCEP/CSPEC and provides only the generic ACMG/AMP framework. PM3 requires a recessive disease context with the tested variant demonstrated in trans with a pathogenic variant; the available case evidence does not establish that context or observation.

Rule

Generic ACMG/AMP PM3: a recessive disease criterion requiring observation of the variant in trans with a pathogenic or likely pathogenic variant in an affected individual; apply only when phase and disease context are established.

Evidence considered

• The governing-framework materials report no NOTCH3 CSPEC/VCEP specification and designate generic ACMG/AMP 2015 as the fallback framework.
• The consolidated case-evidence bundle contains no PM3, phase, trans, cis, affected-proband, or family-segregation evidence for this variant.
• The variant is a missense substitution, NM_000435.2:c.3278G>T, p.Cys1093Phe; no allelic observation relevant to PM3 is provided.

Still missing

• Disease-specific inheritance mode for the evaluated phenotype is not documented in the case evidence.
• No affected proband or affected-family observations involving this variant are documented.
• No second pathogenic/likely pathogenic variant and no cis/trans phase information are documented.

Reasoning

PM5 requires a different pathogenic missense change previously established at the identical codon (residue 1093). The dedicated PM5 comparator search in this case was explicitly not performed ('eligible_for_classic_pm5_search': false, pm5_mode: 'unknown') because classic same-residue PM5 semantics could not be safely confirmed for this case, and the recommendation explicitly instructs not to apply PM5 from that artifact alone. No other source in the bundle (ClinVar absent, no literature hits) documents any alternate amino-acid substitution at codon 1093.

Evidence considered

• pm5_candidates.json: eligible_for_classic_pm5_search=false, pm5_mode='unknown', candidate_summary all zero, recommendation.level='not_applicable' with explicit instruction not to apply PM5 from this artifact alone.
• ClinVar search for this gene/variant returned no_result, providing no comparator classified variants at this codon.

Still missing

• The pm5_candidates harvesting step was skipped for this case (mode=unknown); a manual/targeted ClinVar and literature search for other missense changes at residue 1093 has not been performed and is required before PM5 can be assessed.

Sources

Reasoning

No unconfirmed de novo observation or parental testing is provided in the case-evidence bundle. PM6 cannot be assessed.

Rule

Generic ACMG/AMP PM6: assumed de novo, without confirmation of paternity and maternity, in a patient with the disease and no family history.

Evidence considered

• The consolidated case-evidence bundle contains no proband, parental testing, maternity/paternity, or suspected de novo evidence for NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe).
• No applicable NOTCH3-specific VCEP/CSPEC framework was retrieved; generic ACMG/AMP fallback therefore applies.

Still missing

• A documented suspected de novo occurrence
• Parental testing or other available evidence regarding parental carrier status
• Relevant phenotype and family clinical context

Reasoning

No affected relatives, informative meioses, variant-positive relatives, or segregation results are provided. PP1 cannot be assessed.

Rule

Generic ACMG/AMP PP1: cosegregation with disease in multiple affected family members in a gene definitively known to cause the disease; consider increased strength with multiple informative meioses.

Evidence considered

• The consolidated case-evidence bundle contains no pedigree, affected-relative genotypes, informative meioses, or segregation data for NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe).
• No applicable NOTCH3-specific VCEP/CSPEC framework was retrieved; generic ACMG/AMP fallback therefore applies.

Still missing

• Pedigree and affected-relative information
• Genotype results for informative relatives
• Number of informative meioses and segregation concordance

Reasoning

PP4 was not assessed because no patient phenotype, diagnostic features, or variant-specific clinical presentation is provided in the evidence bundle. Gene-level disease context alone is insufficient to establish a highly specific phenotype in this case.

Rule

Generic ACMG/AMP PP4: apply only when the patient's phenotype or family history is highly specific for a disease with a single-gene etiology; the required phenotype evidence is unavailable.

Evidence considered

• The case summary identifies NOTCH3 and the variant but does not provide clinical phenotype information for the tested individual.

Still missing

• A documented patient phenotype and clinical diagnostic findings are unavailable.
• No evidence establishes that the phenotype is highly specific for a NOTCH3-related disorder.

Reasoning

BS2 cannot be assessed from the case bundle. No unaffected adult homozygotes or unaffected adult hemizygotes carrying the variant were reported, and no suitable population genotype-phenotype dataset was provided.

Rule

Generic ACMG/AMP BS2: requires observation of the variant in an appropriately evaluated unaffected adult homozygote or hemizygote; this evidence is not available.

Evidence considered

• The case bundle contains population absence data but no unaffected adult homozygote/hemizygote observations or phenotype follow-up for this exact variant.
• Variant absence from population databases is not evidence that an unaffected adult homozygote or hemizygote exists.

Still missing

• No genotype-level evidence for unaffected adult homozygotes or hemizygotes was available.
• No penetrance-adjusted cohort or family data were available to determine whether unaffected adult carriers would be informative for BS2.

Reasoning

No gene-specific VCEP/CSPEC framework is available for NOTCH3 (cspec.found=false; generic ACMG fallback in effect). The literature search returned zero PMIDs (literature_pass.json: total_pmids=0) and no functional/in vitro assay data showing normal/wild-type-like function for NM_000435.2:c.3278G>T (p.Cys1093Phe) was identified in evidence.json, vcep_materials.json, or the PVS1 gene-context literature review. No functional assay evidence exists in the case bundle to support BS3.

Still missing

• No published functional/in vitro assay data demonstrating normal (wild-type-comparable) function for p.Cys1093Phe was located in the available sources.
• No NOTCH3 VCEP-specific BS3 assay framework was available in this case bundle to calibrate strength even if an assay were found.

Reasoning

No unaffected relatives tested for the variant, age-dependent phenotype assessment, or informative non-segregation data are provided. BS4 cannot be assessed.

Rule

Generic ACMG/AMP BS4: lack of segregation in affected members of a family; apply only when the tested relatives are informative and the phenotype, age, and penetrance are adequately considered.

Evidence considered

• The consolidated case-evidence bundle contains no unaffected-relative testing or non-segregation evidence for NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe).
• No applicable NOTCH3-specific VCEP/CSPEC framework was retrieved; generic ACMG/AMP fallback therefore applies.

Still missing

• Genotype results for unaffected informative relatives
• Adequate age and clinical assessment of those relatives
• Pedigree structure and expected segregation pattern

Reasoning

No individual or family data shows NM_000435.2:c.3278G>T (NP_000426.2:p.Cys1093Phe) in cis with a pathogenic variant, or in trans with a pathogenic variant in a dominant disease context. The case bundle contains no phase, affected-proband, co-occurring pathogenic-variant, or segregation observation. Therefore BP2 cannot be evaluated from the available evidence.

Rule

Generic ACMG/AMP BP2: the variant is observed in trans with a pathogenic variant for a dominant disorder, or in cis with a pathogenic variant; phase and the relevant disease context must be established.

Evidence considered

• The governing-framework materials report no NOTCH3 CSPEC/VCEP specification and designate generic ACMG/AMP 2015 as the fallback framework.
• The consolidated case-evidence bundle contains no BP2, phase, trans, cis, affected-proband, or family-segregation evidence for this variant.
• The case bundle reports the queried variant as absent from ClinVar and gnomAD, but those population/database findings do not establish cis/trans allelic phase and cannot support BP2.

Still missing

• No co-occurring pathogenic or likely pathogenic variant is documented.
• No cis/trans phase information is documented.
• No affected-proband or family segregation data is documented to establish the relevant inheritance context.

Reasoning

BP5 was not assessed because the case-evidence bundle contains no patient-level molecular workup showing an alternative molecular explanation for the relevant phenotype and no clinical case data from which this criterion can be evaluated.

Rule

Generic ACMG/AMP BP5: consider when an affected individual has an alternative molecular explanation for the phenotype; no qualifying case evidence is available.

Evidence considered

• The consolidated case-evidence bundle does not document an alternative molecular explanation in an affected individual.

Still missing

• Patient-level phenotype and molecular testing results are unavailable.
• No established alternative pathogenic variant or alternative molecular diagnosis is documented.