ASXL1 c.2148dup, p.(Arg717Ter)

Reasoning

NM_015338.5:c.2148dup is a duplication causing a frameshift-equivalent premature stop, reported as NP_056153.2:p.(Arg717Ter) (nonsense at the normalized cDNA level per mutalyzer/VariantValidator). No official ASXL1 ClinGen VCEP CSPEC was found (cspec.found=false in case_summary and prefetch), so the generic ClinGen SVI PVS1 framework (PMC6185798) governs, as scaffolded in pvs1_variant_assessment.json (suggested_default_strength='PVS1', framework_cite='pvs1_generic_framework'). Gene-level gating supports application: pvs1_gene_context.json establishes germline loss-of-function as a plausible ASXL1 disease mechanism (pvs1_gene_gate='eligible', lof_mechanism_supported=true), grounded in literature identifying germline ASXL1 truncating variants as the cause of Bohring-Opitz syndrome. However, transcript/exon structure data from VariantValidator/mutalyzer show this gene has 13 exons and the variant maps to exon 13 of 13 (variant_exonic_positions start_exon=end_exon='13'), i.e. the final exon. Under the ClinGen SVI PVS1 decision tree, premature termination codons located in the final exon are not predicted to trigger nonsense-mediated decay, so the NMD-based full-strength (Very Strong) pathway does not apply; the variant instead falls into the 'NMD escape, protein truncation' branch. The predicted truncated protein retains only 716 of 1542 full-length residues (prefetch mutalyzer protein.predicted / position_last_predicted=716 vs position_last_original=1542), removing ~54% of the protein (well above the 10% threshold used by SVI to consider a truncation potentially consequential). This satisfies the 'meaningful protein truncation' condition for PVS1_Moderate under the SVI NMD-escape branch. Upgrading further to PVS1_Strong requires evidence that the specific truncated/removed C-terminal region is established as critical to protein function (e.g., multiple well-documented pathogenic truncating/pathogenic missense variants clustering downstream of this residue, or direct functional data on this region) -- this case's retrieved evidence bundle does not contain a grounded source demonstrating that. ClinVar has no entry for this variant (absent), and no PM5-type comparator analysis was performed (pm5_candidates_summary recommendation='not_applicable'), so no direct evidence of downstream pathogenic truncations near residue 717 could be confirmed from case materials. The OncoKB record labels ASXL1 R717* as 'Likely Oncogenic' with a curated 'Likely Loss-of-function' biological effect, consistent with, but not proof of, functional criticality of the truncated region for the germline context. SpliceAI predicts no meaningful splice disruption (max delta 0.018), so the variant's PTC-generating consequence is not confounded by an alternate splicing rescue mechanism. Net assessment: PVS1 criterion is met at Moderate strength (last-exon NMD-escape variant causing substantial, >10%, truncation of the ASXL1 protein in a gene with an established germline LOF disease mechanism), pending confirmation of downstream critical-region evidence that could support upgrading to Strong.

Rule

ClinGen SVI PVS1 generic decision tree (PMC6185798): nonsense/PTC variant in the final exon is not predicted to trigger NMD; truncation removing >10% of the protein in a gene with established germline LOF disease mechanism, without confirmed evidence that the specific truncated region is critical, is scored PVS1_Moderate.

Evidence considered

• mutalyzer/VariantValidator normalization confirms NM_015338.5:c.2148dup produces NP_056153.2:p.(Arg717Ter), a nonsense/premature-termination-codon variant.
• VariantValidator variant_exonic_positions places the variant in exon 13 of 13 (start_exon=end_exon='13'), the final exon of the ASXL1 transcript, meaning NMD is not predicted to occur per the ClinGen SVI PVS1 decision tree (PMC6185798).
• Mutalyzer predicted protein data show the truncated protein length is 716 aa versus a full-length 1542 aa protein, a loss of approximately 54% of the protein, exceeding the 10% threshold used in the SVI framework's NMD-escape branch to justify at least PVS1_Moderate.
• pvs1_gene_context.json (citing PMID:24218140 abstract: 'Germline ASXL1 mutations occur in patients with Bohring-Opitz syndrome.') establishes that germline ASXL1 loss-of-function truncating variants are an accepted disease mechanism, satisfying the gene-level PVS1 eligibility gate.
• pvs1_variant_assessment.json explicitly flags as a downgrade consideration: 'Downgrade if the predicted truncation escapes NMD or affects only a non-critical distal region under PMC6185798,' which is directly applicable here given the last-exon location.
• OncoKB curated annotation for ASXL1 R717* states biological_effect='Likely Loss-of-function' and classification='Likely Oncogenic', consistent with loss-of-function consequence though derived from a somatic-oncology curation context rather than germline-specific data.
• SpliceAI predicts no significant splice disruption (max_delta_score=0.018), so the nonsense/PTC consequence is the operative mechanism without a confounding splice-rescue possibility.
• ClinVar search for this exact variant returned no results (absent), so no independent classification or same-region comparator truncating variant could be used to confirm criticality of the removed C-terminal region.
• pm5_candidates_summary recommendation is 'not_applicable' with 0 same-residue candidates identified, so no comparator pathogenic truncations near residue 717 were available from that analysis to support a Strong-level criticality determination.

Still missing

• No official ASXL1 ClinGen VCEP PVS1 specification is available; generic SVI framework used as fallback.
• Criticality of the specific C-terminal region removed by this last-exon truncation (residues 717-1542) is not directly confirmed by grounded case evidence (no ClinVar comparators, no PM5 analysis, no cited functional-domain-mapping publication specific to this residue range); this precludes upgrading beyond PVS1_Moderate to PVS1_Strong.
• ClinVar has no record for this variant, so no independent classification or comparator truncating-variant data could be leveraged.

Sources

Reasoning

PM2 is met at supporting strength. The normalized variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, providing population rarity evidence. Absence alone does not establish disease causality or a benign criterion.

Rule

Generic ACMG/AMP fallback with ClinGen SVI population-frequency recommendation: apply PM2 at supporting strength for a variant absent or extremely rare in population databases when no ASXL1 VCEP/ClinGen specification is available.

Evidence considered

• gnomAD v2.1 queried 20-31022662-C-CT and reported the variant absent.
• gnomAD v4.1 queried chr20-32434859-C-CT and reported the variant absent.
• gnomAD-Canada v1.0 queried 20-32434859-C-CT and reported the variant absent.

Still missing

• No ASXL1 VCEP/ClinGen gene-specific PM2 strength specification was retrieved; this assessment uses the generic ClinGen SVI recommendation to apply PM2 at supporting strength for rarity evidence.

Sources

Reasoning

SpliceAI predicts no splice-altering effect for NM_015338.5:c.2148dup, with all four delta scores (AG=0.001, AL=0.0, DG=0.0, DL=0.018; max=0.018) far below the 0.2 low-sensitivity cutoff from the SpliceAI validation study (Jaganathan et al. 2019, PMID 30661751), supporting BP4 at the standard supporting strength for the splice-impact computational path. The missense in-silico path (REVEL/BayesDel/aGVGD) is not applicable since this is a frameshift/nonsense duplication (p.(Arg717Ter)) rather than a missense substitution, and no VCEP-specific PP3/BP4 lookup material exists for this gene (framework_mode=generic_acmg, cspec.found=false) to override this generic calibration.

Rule

generic_acmg_pp3_bp4

Evidence considered

• case_summary.json compact_evidence.spliceai: SpliceAI predicts no significant splice impact for this variant (max delta score = 0.018); scores DS_AG=0.001, DS_AL=0.0, DS_DG=0.0, DS_DL=0.018.
• SpliceAI delta scores well below the 0.2 threshold established in the original SpliceAI validation publication (Jaganathan KM et al., Cell 2019, PMID 30661751) support a benign/no-effect prediction for splicing, satisfying BP4 at supporting strength.

Still missing

• No gene-specific ASXL1 VCEP splicing lookup table is available to confirm or override the generic SpliceAI 0.2 threshold used here.

Sources

Reasoning

NM_015338.5:c.2148dup (p.(Arg717Ter)) is a frameshift/nonsense duplication, so the missense in-silico path (BayesDel/REVEL/aGVGD) is not applicable - REVEL and BayesDel scores are both absent from prefetch (score: null, found: null for both), and even if a BayesDel value existed it could not be used because no verified published BayesDel threshold is available to this pipeline. The variant lies 3 bp before the 3' end of a large coding exon (exon spans c.1518-2151) but is not at the canonical splice consensus (canonical_splice_consensus: false). SpliceAI, the only calibrated computational predictor with data for this variant, shows no meaningful splice-altering signal: DS_AG=0.001, DS_AL=0.0, DS_DG=0.0, DS_DL=0.018 (max delta = 0.018), all far below the 0.2 low-sensitivity threshold established by the SpliceAI validation study (Jaganathan et al. 2019, PMID 30661751). No evidence therefore supports a pathogenic splicing or missense computational effect for PP3.

Rule

generic_acmg_pp3_bp4

Evidence considered

• case_summary.json compact_evidence.spliceai: SpliceAI predicts no significant splice impact (max delta score = 0.018): DS_AG=0.001, DS_AL=0.0, DS_DG=0.0, DS_DL=0.018.
• SpliceAI delta scores are far below the 0.2 threshold established for splice-altering predictions in the original SpliceAI validation publication (Jaganathan KM et al., Cell 2019, PMID 30661751), so no PP3 splice signal is present.
• prefetch/case_summary compact_evidence.revel and compact_evidence.bayesdel both show score:null, found:null - no missense in-silico predictor data is available, and this is moot since the variant is a frameshift/nonsense change (p.(Arg717Ter)), not a missense substitution.
• BayesDel: no verified published calibration threshold is available to this pipeline, so any BayesDel value (none present here) could not be used for PP3 regardless.

Still missing

• REVEL and BayesDel scores were not computed/available in prefetch for this variant (moot, since the variant is not a missense change).
• No gene-specific VCEP PP3/BP4 lookup material was found (cspec.found=false, framework_mode=generic_acmg), so no VCEP-specific computational threshold override applies.

Sources

Reasoning

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

Rule

Generic ACMG/AMP fallback: BA1 requires an allele frequency sufficiently high to independently establish benignity. No ASXL1 VCEP/ClinGen gene-specific specification was retrieved.

Evidence considered

• gnomAD v2.1 queried 20-31022662-C-CT and reported the variant absent.
• gnomAD v4.1 queried chr20-32434859-C-CT and reported the variant absent.
• gnomAD-Canada v1.0 queried 20-32434859-C-CT and reported the variant absent.

Sources

Reasoning

BS1 is not met. The variant is absent from the queried population datasets; therefore, there is no allele frequency greater than expected for an ASXL1-related germline disorder.

Rule

Generic ACMG/AMP fallback: BS1 requires an allele frequency greater than expected for the relevant disorder. No ASXL1 VCEP/ClinGen gene-specific threshold was retrieved.

Evidence considered

• gnomAD v2.1 queried 20-31022662-C-CT and reported the variant absent.
• gnomAD v4.1 queried chr20-32434859-C-CT and reported the variant absent.
• gnomAD-Canada v1.0 queried 20-32434859-C-CT and reported the variant absent.

Sources

Reasoning

De novo occurrence cannot be assessed because the consolidated evidence bundle contains no proband-level phenotype information, parental genotypes, parentage confirmation, or testing method for NM_015338.5:c.2148dupT (p.Arg717Ter).

Rule

Generic ACMG/AMP de novo evidence framework; PS2 requires a confirmed de novo occurrence in a patient with a phenotype consistent with the gene-disease relationship.

Evidence considered

• The case framework is generic ACMG/AMP because no ASXL1 CSPEC/VCEP or local gene-specific framework was retrieved.
• No de novo case observation, parental testing result, parentage information, or phenotype description is provided in the consolidated case-evidence bundle.

Still missing

• Obtain molecular testing results for both biological parents using a method capable of detecting the variant.
• Confirm biological parentage when required by the laboratory policy.
• Document the proband phenotype and its consistency with the relevant ASXL1-associated germline disorder.

Sources

Reasoning

No governing VCEP/CSPEC exists for ASXL1 germline interpretation (cspec.found=false; framework_mode=generic_acmg), so generic ACMG/AMP PS3 criteria apply, which require a validated functional assay demonstrating an abnormal effect of THIS specific variant (NM_015338.5:c.2148dup / p.(Arg717Ter)) in an appropriately controlled study. The literature packet (oncokb citations) contains only gene-level ASXL1 loss-of-function mechanistic studies: PMID:22897849 used patient-derived cell lines with homozygous frameshift/nonsense ASXL1 mutations (not the specific R717* variant) and siRNA/shRNA knockdown of wild-type ASXL1 to show H3K27me3 loss and HOXA dysregulation; PMID:26095772 studied specific truncating mutants p.Y591X, p.R404X, and p.G646Wfs*12 (distinct variants) in a BAP1 deubiquitinase gain-of-function assay; PMID:19388938 and PMID:21455215 are mutation-discovery cohort studies without functional assays of this variant. None of the reviewed full-text sources report a validated functional assay performed on the exact c.2148dup/p.Arg717Ter allele. Per ClinGen SVI PS3/BS3 recommendations, evidence must be variant-specific (or from a validated proxy assay of the identical variant) to apply PS3, so this criterion cannot be scored from available evidence.

Evidence considered

• Searched all full-text publications retrieved for this case (PMID:19388938, PMID:21455215, PMID:22897849, PMID:26095772) for the variant string c.2148dup, R717*, Arg717Ter, and residue 717; no matches were found in any of the four full-text articles.
• PMID:22897849 full text: functional assays (western blot protein expression, siRNA/shRNA knockdown, ChIP-seq for H3K27me3, HOXA gene expression, ASXL1 add-back growth suppression) were performed using homozygous ASXL1-null leukemia cell lines (NOMO1, KBM5) and knockdown of wild-type ASXL1, not the p.Arg717Ter allele specifically.
• PMID:26095772 full text: functional deubiquitinase gain-of-function assays were performed on FLAG-tagged ASXL1 truncation mutants p.Y591X and p.G646Wfs*12 (and reference to p.R404X), none of which correspond to p.Arg717Ter.
• PMID:19388938 and PMID:21455215 are mutation-discovery/cohort sequencing studies (MDS/CMML/MPN patient cohorts) without variant-specific functional assay data.

Still missing

• No published functional assay (western blot, deubiquitinase activity, HOXA/H3K27me3 ChIP, colony/proliferation assay, or other validated model) has directly tested NM_015338.5:c.2148dup / p.(Arg717Ter); only other ASXL1 truncating variants (p.Y591X, p.R404X, p.G646Wfs*12) or pooled/knockdown constructs have been functionally characterized.
• No ClinGen ASXL1 VCEP specification available to define an approved PS3 functional assay or strength calibration for this gene.

Sources

Reasoning

No case-control or cohort enrichment data for the exact variant were retrieved.

Evidence considered

• No applicable ASXL1 VCEP/ClinGen specification was retrieved; the case framework is generic ACMG/AMP.
• The literature bundle's retrieved papers are not indexed to PS4 and a targeted search of the available full-text files found no mention of c.2148dup, p.Arg717Ter, or R717*.

Still missing

• Well-ascertained affected and control counts, or a validated enrichment analysis, for NM_015338.5:c.2148dupT (p.Arg717Ter).

Reasoning

No affected-proband observation with this variant and a pathogenic or likely pathogenic ASXL1 variant in confirmed trans was provided. The retrieved case material also does not establish an applicable recessive ASXL1 disease context.

Rule

Generic ACMG/AMP PM3: requires a pathogenic or likely pathogenic variant observed in trans with the assessed variant in an affected individual with a recessive disorder.

Evidence considered

• No ASXL1 ClinGen VCEP/CSPEC or local gene-specific framework was retrieved; the case is designated for generic ACMG/AMP fallback.
• The consolidated evidence bundle contains no proband-level genotype, parental phase, compound-heterozygous observation, or biallelic disease evidence for NM_015338.5:c.2148dupT.

Still missing

• Affected-proband data for this variant.
• A second pathogenic or likely pathogenic ASXL1 variant and laboratory-confirmed trans phase.
• An applicable autosomal-recessive ASXL1 disease association, if one is being considered.

Sources

Reasoning

Assumed de novo occurrence cannot be assessed because the consolidated evidence bundle provides no parental testing data or proband phenotype information for this variant.

Rule

Generic ACMG/AMP de novo evidence framework; PM6 may be considered only for an assumed de novo occurrence with unavailable or unconfirmed parental testing and a phenotype consistent with the gene-disease relationship.

Evidence considered

• The case framework is generic ACMG/AMP because no ASXL1 CSPEC/VCEP or local gene-specific framework was retrieved.
• No reported or assumed de novo observation, parental genotype, parentage information, or proband phenotype is provided in the consolidated case-evidence bundle.

Still missing

• Obtain parental variant testing results.
• Document whether parentage was confirmed and the proband phenotype's consistency with the relevant ASXL1-associated germline disorder.

Sources

Reasoning

Segregation cannot be assessed because no family pedigree, affected or unaffected relative genotypes, phenotypes, or informative meioses are provided for this variant.

Rule

Generic ACMG/AMP segregation evidence framework; PP1 requires co-segregation of the variant with disease in multiple affected family members and must be weighted from documented informative observations.

Evidence considered

• The case framework is generic ACMG/AMP because no ASXL1 CSPEC/VCEP or local gene-specific framework was retrieved.
• No segregation data, pedigree, relative genotypes, relative phenotypes, or meiosis count is provided in the consolidated case-evidence bundle.

Still missing

• Obtain a pedigree with phenotypes and variant testing results for informative relatives.
• Document the number of informative meioses and whether the variant segregates with the relevant ASXL1-associated germline phenotype.

Sources

Reasoning

No phenotype or diagnostic context for the tested individual was provided, so phenotype specificity cannot be evaluated.

Evidence considered

• No applicable ASXL1 VCEP/ClinGen specification was retrieved; the case framework is generic ACMG/AMP.
• The evidence bundle contains variant annotation and literature triage but no individual-level phenotype or diagnostic findings.

Still missing

• Patient phenotype, age of onset, clinical diagnosis, and assessment for phenocopies or alternative causes.

Reasoning

An exact-variant ClinVar expert-panel Pathogenic/Likely pathogenic assertion was not identified.

Evidence considered

• ClinVar search for NM_015338.5:c.2148dupT / ASXL1 c.2148dup returned no exact result; therefore no expert-panel assertion is available.

Still missing

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

Sources

Reasoning

BS2 is not assessed because the available population evidence establishes absence, not observations of the variant in well-phenotyped healthy adult individuals. No homozygote or healthy-carrier evidence is available for this absent variant.

Rule

Generic ACMG/AMP fallback: BS2 requires observation in healthy adult individual(s) when full penetrance is expected at an early age. No ASXL1 VCEP/ClinGen gene-specific BS2 rule was retrieved.

Evidence considered

• gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0 each reported the queried variant absent; consequently, none can provide healthy-carrier or homozygote observations.

Still missing

• Documented observations of the variant in appropriately phenotyped unaffected individuals at an age informative for ASXL1-related disease, with zygosity and ascertainment available.

Sources

Reasoning

No governing VCEP/CSPEC exists for ASXL1 (generic ACMG/AMP fallback). BS3 requires a validated functional assay showing no damaging effect of this specific variant. All available functional literature in this case's evidence packet supports a loss-of-function/damaging mechanism for ASXL1 truncating variants generally (protein loss, H3K27me3/HOXA dysregulation, growth/transformation effects), and none of it tested the exact c.2148dup/p.Arg717Ter allele or reported a benign/no-effect functional result for it. There is no evidence available, of any kind, suggesting this variant lacks functional impact, so BS3 cannot be assessed as met or not_met and is reported as not_assessed.

Evidence considered

• Same full-text search as for PS3 (PMID:19388938, PMID:21455215, PMID:22897849, PMID:26095772) found no variant-specific functional data, benign or damaging, for c.2148dup/p.Arg717Ter; all functional data in these papers pertains to other ASXL1 truncating variants or knockdown/knockout models and uniformly shows loss-of-function biology, not evidence of a normal/benign functional profile for this exact variant.

Still missing

• No functional assay data of any kind (damaging or benign) exists for the exact p.Arg717Ter allele.
• No ClinGen ASXL1 VCEP specification available to define an approved BS3 assay or strength calibration for this gene.

Sources

Reasoning

Lack of segregation cannot be assessed because no genotype-phenotype data for affected or unaffected relatives are available for this variant.

Rule

Generic ACMG/AMP segregation evidence framework; BS4 requires convincing non-segregation of the variant with disease in informative relatives after relevant confounders are addressed.

Evidence considered

• The case framework is generic ACMG/AMP because no ASXL1 CSPEC/VCEP or local gene-specific framework was retrieved.
• No family genotype-phenotype observations or non-segregation data are provided in the consolidated case-evidence bundle.

Still missing

• Obtain family genotype and phenotype data sufficient to evaluate whether the variant is absent from informative affected relatives or present in informative unaffected relatives.
• Assess phenocopies, age-dependent penetrance, variable expressivity, and parental mosaicism before using non-segregation evidence.

Sources

Reasoning

No observation of NM_015338.5:c.2148dupT in confirmed cis or trans with an independent pathogenic or likely pathogenic ASXL1 variant was provided. Therefore BP2 cannot be assessed.

Rule

Generic ACMG/AMP BP2: requires observation with a pathogenic or likely pathogenic variant in a phase and inheritance/disease context demonstrating that the combination is not causative.

Evidence considered

• No ASXL1 ClinGen VCEP/CSPEC or local gene-specific framework was retrieved; the case is designated for generic ACMG/AMP fallback.
• The consolidated evidence bundle contains no phase-resolved co-occurrence data for NM_015338.5:c.2148dupT and no second pathogenic or likely pathogenic ASXL1 variant.

Still missing

• An affected or unaffected individual carrying this variant with an independent pathogenic or likely pathogenic ASXL1 variant.
• Laboratory-confirmed cis/trans phase and phenotype information sufficient to determine whether the allele combination is non-causative under the relevant inheritance model.

Sources

Reasoning

No patient-level alternate molecular diagnosis or phenotype attribution was provided to assess whether this variant is incidental to another established cause.

Evidence considered

• The evidence bundle contains no patient-specific molecular diagnosis or phenotype information.

Still missing

• Patient phenotype and comprehensive molecular workup, including any independently established alternative diagnosis.

Reasoning

An exact-variant ClinVar expert-panel Benign/Likely benign assertion was not identified.

Evidence considered

• ClinVar search for NM_015338.5:c.2148dupT / ASXL1 c.2148dup returned no exact result; therefore no expert-panel assertion is available.

Still missing

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

Sources