ASXL1 c.1934dup, p.Gly646TrpfsTer12

Reasoning

NM_015338.6:c.1934dup is a frameshift variant in the clinically relevant MANE Select ASXL1 transcript, producing NP_056153.2:p.(Gly646TrpfsTer12) and truncating the 1541-amino-acid protein after approximately 657 residues. It lies in exon 13, the terminal coding exon, so the premature termination codon is predicted to escape nonsense-mediated decay; this is the specific named downgrade from the default very strong PVS1 strength. The variant therefore remains a null-effect truncating allele with loss of function established as a germline ASXL1 disease mechanism, but the NMD-escape rule limits PVS1 to strong. The variant occurs in an eight-guanine homopolymer repeat, a sequencing-confirmation concern documented for this allele, not a consequence-based reason to remove PVS1; independent Sanger confirmation is described in the reviewed literature.

Rule

PVS1: Null variant (nonsense, frameshift, canonical +/−1 or 2 splice sites, initiation codon, single or multi-exon deletion) in a gene where loss of function (LOF) is a known mechanism of disease.

Evidence considered

• Variant normalization identifies NM_015338.6:c.1934dup as a frameshift producing NP_056153.2:p.(Gly646TrpfsTer12), with the affected coding interval assigned to exon 13 of the MANE Select ASXL1 transcript.
• The ASXL1 gene-level context records loss of function as a supported germline disease mechanism, making generic PVS1 eligible.
• The terminal exon 13 location places the premature termination codon in the final coding exon; predicted escape from nonsense-mediated decay is the named downgrade applied here, from very strong to strong.
• PMID:31076447 reports the exact p.G646Wfs*12/c.1934dup allele and describes its occurrence in an eight-guanine homopolymer; the paper reports bidirectional Sanger sequencing for confirmation, supporting that repeat context as a technical reliability issue rather than a reason to negate the frameshift consequence.
• PMID:26095772 directly evaluates p.G646Wfs*12 as a truncated ASXL1 construct in a HEK293T ASXL1-BAP1 PR-DUB assay; this confirms the documented truncated protein context but is not used as an independent PVS1 functional assay criterion.

Sources

Reasoning

PS3 is met at supporting strength. PMID:26095772 directly tested the exact ASXL1 p.G646Wfs*12 product corresponding to NM_015338.6:c.1934dup in a HEK293T cell-based assay. The variant was expressed with BAP1 and compared with full-length ASXL1 and other truncations; all truncated ASXL1 proteins, including p.G646Wfs*12, enhanced BAP1-associated H2AK119Ub deubiquitinase activity substantially more than full-length ASXL1. The assay used a biologically relevant ASXL1-BAP1 PR-DUB readout and included full-length and empty/single-expression controls, supporting an altered functional effect. The evidence is not calibrated above supporting because it is a research cell assay rather than a validated, reproducible clinical diagnostic assay, and the reported effect is hyperactivity of the truncated complex rather than a direct measurement of loss of normal ASXL1 function.

Rule

Well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product.

Evidence considered

• Cancer-associated ASXL1 truncations, including the exact p.G646Wfs*12 product corresponding to c.1934dup, were tested in transiently transfected HEK293T cells with BAP1.
• The experiment compared full-length ASXL1 with truncated constructs and measured global H2AK119Ub by immunocytochemistry and western blotting; the paper reports that all truncated ASXL1 proteins enhanced BAP1 H2AK119Ub deubiquitinase activity substantially more than full-length ASXL1.
• The assay included biologically relevant ASXL1-BAP1 PR-DUB activity and full-length/single-expression comparator conditions, but was a research cell assay rather than a validated clinical diagnostic assay.

Still missing

• No validated, reproducible clinical diagnostic-laboratory assay was identified.
• The study demonstrates altered or hyperactive ASXL1-BAP1 activity, but does not directly quantify loss of normal ASXL1 function in a patient-derived system.

Sources

Reasoning

Generic ACMG/AMP PS4 is met by the exact-variant case-control comparison reported in PMID:31076447. The paper reports p.G646Wfs*12/c.1934dup in 33 of 333 patients with primary myelofibrosis. Using the gnomAD v4.1 aggregate frequency of 685/1,601,778 alleles as an approximate control carrier count of 685/800,889 individuals, the carrier-count table is 33 affected carriers, 300 affected noncarriers, 685 control carriers, and 800,204 control noncarriers. The calculated odds ratio is 128.50 with an approximate 95% CI of 89.01-185.51. The governing ACMG/AMP threshold is OR >5.0 with a confidence interval that does not include 1.0; therefore 128.50 > 5.0 and 89.01-185.51 excludes 1.0. Human review is needed because the cohort is a primary-myelofibrosis cohort and the variant has known somatic/artifact-detection context, while gnomAD is not explicitly ancestry-matched in the supplied evidence.

Rule

The prevalence of the variant in affected individuals is significantly increased compared to the prevalence in controls. ... Relative risk (RR) or odds ratio (OR), as obtained from case-control studies, is >5.0 and the confidence interval around the estimate of RR or OR does not include 1.0.

Evidence considered

• PMID:31076447, Results/Table 1: exact ASXL1 p.G646Wfs*12 (c.1934dup) variant in 33 of 333 primary-myelofibrosis patients.
• gnomAD v4.1: 685/1,601,778 alleles; for the approximate carrier comparison, treated the 685 observed alleles as 685 heterozygous control carriers among 800,889 individuals.
• Calculated from those counts: OR = (33 x 800,204)/(300 x 685) = 128.50; approximate 95% CI = 89.01-185.51.

Still missing

• The published case cohort and gnomAD controls are not explicitly ancestry-matched in the supplied evidence.
• The cohort is primary myelofibrosis and may represent somatic rather than constitutional occurrence; confirm applicability to the intended germline disease context.

Sources

Reasoning

PM6 is met at supporting strength because ClinVar records for the exact variant NM_015338.6:c.1934dup include seven clinical-testing submissions labeling its origin as de novo, while no submission documents confirmed maternity and paternity. This supports assumed de novo occurrence without the parental-confirmation required for PS2; the underlying proband and parental laboratory reports are not available for independent verification.

Rule

Assumed de novo, but without confirmation of paternity and maternity.

Evidence considered

• ClinVar variation 426927 submission rows: SCV001526547, SCV001737058, SCV001934374, SCV000803831, SCV004102730, SCV005871206, and SCV007600664 label origin as de novo; the record does not document confirmed maternity and paternity.

Still missing

• Underlying proband-level laboratory reports and parental testing records
• Clinical documentation confirming the affected patient's phenotype and disease diagnosis

Sources

Reasoning

PM2 is not met. The variant is present in multiple population datasets at frequencies above the supplied generic PM2 supporting threshold of 0.0001: gnomAD v2.1 all-comers AF 0.00049539, gnomAD v4.1 all-comers AF 0.00042765, and the non-cancer v2.1 exome subset AF 0.000436797. The variant is therefore neither absent from controls nor at or below the generic extremely-low-frequency cutoff.

Rule

Absent from controls (or at extremely low frequency if recessive) in Exome Sequencing Project, 1000 Genomes or Exome Aggregation Consortium...PM2 (supporting): <= 0.0001 — SVI recommendation to downgrade PM2 to Supporting (ClinGen SVI 2020) (PMID:25741868)

Evidence considered

• gnomAD v2.1 all-comers: AF 0.00049539, 121/244252 alleles, zero homozygotes.
• gnomAD v4.1 all-comers: AF 0.00042765, 685/1601778 alleles, zero homozygotes.
• When using the non-cancer exome source, GNOMAD_V2_1_NON_CANCER reports AF 0.000436797, 89/203756 alleles, zero homozygotes.
• All available AF values exceed the supplied generic PM2 supporting threshold of <= 0.0001.

Sources

Reasoning

PP5 is not met. Although literature_pass identifies PMID:30147881 as a case-report lead describing the exact variant as a probable cause of Bohring-Opitz syndrome, the exact-variant ClinVar record has zero expert-panel submissions. Its classifications are ordinary laboratory or other non-expert assertions, which do not satisfy the required exact-variant expert-panel condition for PP5.

Rule

Reputable source recently reports variant as pathogenic, but the evidence is not available to the laboratory to perform an independent evaluation.

Evidence considered

• ClinVar exact-variant audit for Variation ID 426927: 17 submissions, 0 expert-panel submissions; the available Pathogenic/Likely pathogenic assertions are not expert-panel classifications.
• literature_pass identifies PMID:30147881 only as an AI-classified lead; that lead is not sufficient to override the required expert-panel condition.

Sources

Reasoning

BA1 is not met. The governing frequency measure is the gnomAD filtering allele frequency (grpmax FAF), not the popmax point estimate. In the default all-comers datasets, grpmax FAF is 0.00055108 in gnomAD v2.1 and 0.00042259 in gnomAD v4.1, both far below the generic stand-alone threshold of 0.05. The governing generic definition requires an allele frequency above 5%, and this variant also has a pathogenic/likely pathogenic ClinVar consensus, so there is no basis for a stand-alone benign call from population frequency.

Rule

Allele frequency is above 5% in Exome Sequencing Project, 1000 Genomes or Exome Aggregation Consortium.

Evidence considered

• gnomAD v2.1 all-comers: 121/244252 alleles, AF 0.00049539, zero homozygotes; grpmax FAF 0.00055108.
• gnomAD v4.1 all-comers: 685/1601778 alleles, AF 0.00042765, zero homozygotes; grpmax FAF 0.00042259.
• The non-cancer exome subset is also well below BA1: GNOMAD_V2_1_NON_CANCER AF 0.000436797, 89/203756 alleles, zero homozygotes; its reported grpmax FAF is 0.00040074.
• ClinVar variation 426927 reports 13 Pathogenic, 2 Likely pathogenic, and 1 Uncertain significance clinical-laboratory submissions; no expert-panel submission is present.

Sources

Reasoning

BS1 is not met. The appropriate filtering frequency is grpmax FAF: 0.00055108 in gnomAD v2.1 all-comers and 0.00042259 in gnomAD v4.1 all-comers; the non-cancer v2.1 exome subset has grpmax FAF 0.00040074. Each is below the supplied generic BS1 default threshold of 0.01, so the observed frequency is not greater than expected for the disorder under the available generic rule. The frequency is not elevated in a small isolate population, and no maximum-credible-frequency calculation is needed to reject a value this far below the default threshold.

Rule

Allele frequency is greater than expected for the disorder...BS1 (strong): >= 0.01 — Generic SVI default; per-gene derivation preferred (Whiffin et al. 2017) (PMID:28518168)

Evidence considered

• gnomAD v2.1 all-comers: grpmax FAF 0.00055108; the highest observed subpopulation point estimate is only 0.000766636 from 5/6522 alleles.
• gnomAD v4.1 all-comers: grpmax FAF 0.00042259; the highest observed subpopulation point estimate is 0.000694826 from 43/61886 alleles.
• When the governing source is the non-cancer v2.1 exome subset, the reported AF is 0.000436797 and grpmax FAF is 0.00040074.
• The generic BS1 calibration supplied for this case is >= 0.01; all qualifying FAF values are more than an order of magnitude below it.

Sources

Reasoning

BS3 is not met. The exact-variant functional study in PMID:26095772 did not show no damaging effect or normal function; instead, p.G646Wfs*12 produced an altered ASXL1-BAP1 complex with substantially enhanced H2AK119Ub deubiquitinase activity compared with full-length ASXL1. Although the assay does not establish the precise disease mechanism or justify a stronger PS3 level, its result is inconsistent with the absence of a damaging functional effect required for BS3.

Rule

Well-established in vitro or in vivo functional studies show no damaging effect on protein function or splicing.

Evidence considered

• The exact p.G646Wfs*12 ASXL1 construct was tested in HEK293T cells and showed substantially enhanced BAP1-associated H2AK119Ub deubiquitinase activity relative to full-length ASXL1.
• This altered functional result does not support a conclusion of no damaging effect or normal protein function.

Sources

Reasoning

BP5 is not met because the supplied evidence contains no documented case in which this exact variant co-occurs with an alternate molecular basis for the disease. Somatic cancer-cohort observations and population occurrences do not establish an alternate molecular diagnosis for a disease case.

Rule

Variant found in a case with an alternate molecular basis for disease.

Reasoning

BP6 is not met. The exact-variant ClinVar audit reports zero expert-panel submissions and no exact-variant ClinVar expert-panel classification as Benign or Likely benign. Ordinary laboratory submissions, aggregate labels, and non-expert assertions cannot trigger BP6.

Rule

Reputable source recently reports variant as benign, but the evidence is not available to the laboratory to perform an independent evaluation.

Evidence considered

• ClinVar exact-variant audit for Variation ID 426927: expert_panel_submissions = 0; no expert-panel Benign or Likely benign assertion is present.

Sources

Reasoning

PS2 cannot be assessed because the available record does not document both maternity and paternity confirmation for the proband. Several ClinVar submissions label the variant origin as de novo, but the record provides no parental genotypes, maternity confirmation, paternity confirmation, or family-history details; therefore the PS2 definition is not met on auditable evidence.

Rule

De novo (both maternity and paternity confirmed) in a patient with the disease and no family history.

Evidence considered

• ClinVar variation 426927 submission rows: seven submissions label origin as de novo, but the available rows do not provide parental genotypes, maternity confirmation, paternity confirmation, or family-history information.

Still missing

• Proband-level parental genotypes with explicit maternity and paternity confirmation
• Documented absence of family history

Sources

Reasoning

PM3 cannot be assessed because the available exact-variant records report de novo or unspecified origins but provide no parent/offspring phase testing, no pathogenic variant documented in trans, and no established recessive disease context for this case.

Rule

For recessive disorders, detected in trans with a pathogenic variant.

Evidence considered

• ClinVar exact-match submission audit for NM_015338.6(ASXL1):c.1934dup (p.Gly646fs), Variation ID 426927: several submissions report de novo origin and others report germline or unknown origin, but the audit contains no parent-tested phase result or pathogenic variant documented in trans.

Still missing

• Parental or offspring testing establishing phase is unavailable.
• No affected-proband observation with a second pathogenic variant in trans is documented.
• The recessive disease context required for PM3 is not established in the available evidence.

Sources

Reasoning

PP1 cannot be assessed because the available evidence contains no documented affected relatives, informative meioses, family pedigree, or variant-status results in multiple affected family members. Reports of de novo origin in ClinVar are not segregation data and cannot establish co-segregation.

Rule

Co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease.

Evidence considered

• The available ClinVar record reports variant origin classifications but provides no pedigree, affected-relative testing, meioses, or familial co-segregation results.

Still missing

• Pedigree with clinically affected relatives
• Variant testing and phenotype data for multiple affected family members
• Informative meioses or segregation analysis

Sources

Reasoning

The supplied evidence identifies a Bohring-Opitz syndrome case report as a lead, but it does not provide a sufficiently detailed patient phenotype or family-history description to independently establish that the phenotype is highly specific for a single genetic etiology.

Rule

Patient’s phenotype or family history is highly specific for a disease with a single genetic etiology.

Still missing

• Obtain the patient's detailed clinical phenotype and family history from the full case report or clinical record.
• Document why the phenotype is highly specific for the relevant disease and its genetic etiology.

Reasoning

BS2 is not assessed. The population datasets show zero homozygotes, but they do not document that any individual carrying this indel was a clinically characterized healthy adult with the relevant phenotype assessed. Aggregate gnomAD presence cannot by itself establish the required healthy-adult observation, particularly because this homopolymer indel has a documented history of sequencing-call artifacts and the disorder associated with germline ASXL1 loss of function is an early-onset dominant disorder. A curated individual-level healthy-adult observation with reliable orthogonal confirmation would be required.

Rule

Observed in a healthy adult individual for a recessive (homozygous), dominant (heterozygous), or X-linked (hemizygous) disorder, with full penetrance expected at an early age.

Evidence considered

• gnomAD v2.1 reports 121 variant alleles and zero homozygotes; gnomAD v4.1 reports 685 variant alleles and zero homozygotes.
• The non-cancer v2.1 exome subset reports 89 variant alleles and zero homozygotes.
• PMID:31076447 reports that ASXL1 p.G646Wfs*12 was long considered a sequencing artifact because it occurs in a homopolymer of eight guanine nucleotides, and describes cancer-cohort observations rather than healthy adult controls.
• No source reviewed provides a confirmed individual-level healthy adult carrier observation with clinical phenotype assessment.

Still missing

• No individual-level, clinically phenotyped healthy adult carrier with orthogonal confirmation was identified.
• The available population records do not establish whether the reported carriers are true constitutional variants rather than sequencing artifacts or mosaic observations.

Sources

Reasoning

BS4 cannot be assessed because no affected family members with negative variant testing are documented. The available evidence does not provide a pedigree, phenotype assessments of relatives, or non-segregation results; therefore lack of segregation cannot be established.

Rule

Lack of segregation in affected members of a family.

Evidence considered

• The available ClinVar record contains no family pedigree, affected-relative genotypes, or non-segregation observations.

Still missing

• Pedigree and phenotype assessment of affected relatives
• Variant testing results for affected relatives
• Assessment for phenocopies or more than one pathogenic variant in the family

Sources

Reasoning

BP2 cannot be assessed because the available exact-variant records do not document a pathogenic variant in trans in an affected proband or a pathogenic variant in cis; reported de novo origin of this variant does not establish either allelic relationship.

Rule

Observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder or observed in cis with a pathogenic variant in any inheritance pattern.

Evidence considered

• ClinVar exact-match submission audit for NM_015338.6(ASXL1):c.1934dup (p.Gly646fs), Variation ID 426927: submissions include de novo, germline, and unknown-origin reports, but no cis/trans co-occurring pathogenic variant or phase result is documented.

Still missing

• No affected-proband co-occurrence data with a second pathogenic variant are documented.
• No phase result distinguishing cis from trans is available.
• No documented pathogenic variant in cis or in trans is available for BP2 evaluation.

Sources