SF3B1 c.2098A>G, p.Lys700Glu

Reasoning

PS3 is met at strong strength. The exact SF3B1 p.Lys700Glu (K700E) variant was tested in two independent functional studies using appropriate wild-type comparisons and multiple orthogonal assays. PMID:31474574 used CRISPR/Cas9-engineered K562 cells with mono-allelic tagged wild-type or K700E SF3B1, affinity purification, mass spectrometry, western blotting, RT-PCR, and HEK293T overexpression/rescue experiments. K700E caused upstream cryptic 3' splice-site usage, specifically impaired association with SUGP1, and the splicing defect was partially rescued by SUGP1 overexpression. PMID:27818134 used RNA-seq and splicing analyses of primary CLL samples, wild-type versus K700E expression in K562 cells, and endogenous K700E gene-edited Nalm-6 cells, with RT-PCR, Sanger sequencing, immunoblotting, and additional cellular assays; K700E increased mutation-associated aberrant splicing and altered downstream cellular phenotypes. These variant-specific, mechanistically coherent results with engineered endogenous models, wild-type controls, orthogonal readouts, and a partial rescue experiment are supportive of a damaging effect on SF3B1 function. The studies are research assays rather than validated clinical diagnostic-laboratory assays, so no greater-than-strong strength is assigned.

Rule

Well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product. Note: Functional studies that have been validated and shown to be reproducible and robust in a clinical diagnostic laboratory setting are considered the most well-established.

Evidence considered

• Wang et al. 2016 (PMID:27818134): SF3B1 K700E was evaluated in primary CLL samples, K562 overexpression models, and endogenous K700E gene-edited Nalm-6 cells; RNA-seq, RT-PCR, Sanger sequencing, immunoblotting, telomerase, DNA-damage, and Notch assays showed increased mutation-associated alternative splicing and downstream cellular effects.
• Zhang et al. 2019 (PMID:31474574): CRISPR/Cas9-engineered K562 cells with mono-allelic tagged wild-type or K700E SF3B1 showed upstream cryptic 3' splice-site usage and reduced SUGP1 association; HEK293T SUGP1 overexpression partially rescued the splicing defect.
• The evidence is variant-specific and includes wild-type comparisons, orthogonal functional readouts, endogenous gene editing, and a mechanistic rescue experiment; no clinical diagnostic-laboratory validation was reported.

Still missing

• The available studies are research assays and do not document validation and reproducibility in a clinical diagnostic laboratory setting.

Sources

Reasoning

SF3B1 p.Lys700Glu lies at codon 700, documented as a recurrent hotspot. PMID:25424858 reports a recurrent hotspot T>C substitution at codon 700 producing K700E, and COSMIC records the variant 811 times; the case hotspot source independently reports that the variant lies in a statistically significant hotspot. No VCEP domain table applies because this gene has no governing VCEP. The hotspot evidence supports PM1 under the generic ACMG/AMP definition; no separate functional-domain route is needed.

Rule

Located in a mutational hot spot and/or critical and well-established functional domain (e.g. active site of an enzyme) without benign variation.

Evidence considered

• The hotspot evidence states that the variant lies in a statistically significant hotspot.
• COSMIC records SF3B1 K700E as COSV59205318 with n = 811 occurrences.
• PMID:25424858 reports a recurrent hotspot T > C base-pair substitution at codon 700 leading to K700E.

Sources

Reasoning

PM2 is met at supporting strength under the supplied generic SVI threshold. The variant is observed at extremely low frequency in the default all-comers population sources: AF=8.31417e-05 (23/276636 alleles) in gnomAD v2.1 and AF=4.48853e-05 (72/1604090 alleles) in gnomAD v4.1, both at or below 0.0001. There are zero observed homozygotes in both datasets. The non-cancer subsets are not required because no VCEP specifies a non-cancer or exome-only source for this generic interpretation.

Rule

PM2 (supporting): <= 0.0001 — SVI recommendation to downgrade PM2 to Supporting (ClinGen SVI 2020) (PMID:25741868)

Evidence considered

• gnomAD v2.1 all-comers: 23/276636 alleles, AF=8.31417e-05, homozygotes=0.
• gnomAD v4.1 all-comers: 72/1604090 alleles, AF=4.48853e-05, homozygotes=0.
• The generic PM2 supporting threshold supplied for this case is <=0.0001, identified as the ClinGen SVI 2020 recommendation and supplied with PMID:25741868 as the case calibration citation.

Sources

Reasoning

The variant is a missense change, NM_012433.4:c.2098A>G (p.Lys700Glu), so the governing generic rule selects the REVEL path only; SpliceAI is not used for PP3/BP4 in this missense situation. REVEL is 0.619, below the generic supporting PP3 threshold of >=0.644 from the ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997), so PP3 is not met. BayesDel is not used because no generic ACMG-strength calibration is established. The SpliceAI max delta of 0.025 is not counted and does not establish PP3 for this missense variant.

Rule

Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.). ... As many in silico algorithms use the same or very similar input for their predictions, each algorithm should not be counted as an independent

Evidence considered

• Variant consequence: missense, NM_012433.4:c.2098A>G (NP_036565.2:p.(Lys700Glu)).
• Local REVEL lookup: score 0.619; the generic supporting PP3 threshold is >=0.644, using the supplied ClinGen SVI calibration (Pejaver et al. 2022, PMID:36413997).
• SpliceAI max delta 0.025 was excluded because the generic rule assigns missense variants to the REVEL path only; BayesDel was excluded because no generic ACMG-strength calibration is established.

Sources

Reasoning

The governing PP5 rule requires a reputable source to recently report the variant as pathogenic, with the evidence unavailable to the laboratory for independent evaluation. Exact-variant ClinVar records show no expert-panel submissions; the available Pathogenic and Likely pathogenic assertions are ordinary clinical-laboratory submissions or lack usable independent evidence, and the aggregate ClinVar label is conflicting. The literature_pass PP5 leads are not exact-variant expert-panel classifications. PP5 is therefore not met.

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 variation 376004: exact variant NM_012433.4:c.2098A>G (p.Lys700Glu) has zero expert-panel submissions; available classifications include conflicting Uncertain significance, Likely pathogenic, and Pathogenic laboratory submissions.
• The available ClinVar Pathogenic/Likely pathogenic assertions are non-expert clinical-laboratory submissions and cannot trigger PP5 under the specified generic rule.

Sources

Reasoning

BA1 is not met. The variant's all-comers gnomAD frequencies are 8.31417e-05 (23/276636 alleles) in gnomAD v2.1 and 4.48853e-05 (72/1604090 alleles) in gnomAD v4.1, with maximum reported population frequencies of 2.97147e-04 and 1.70207e-04, respectively; all are far below the generic stand-alone benign threshold of 0.05. No VCEP-specific population source overrides the generic default, and the variant is an SNV, so the indel calling caveat does not apply.

Rule

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

Evidence considered

• gnomAD v2.1 all-comers: 23/276636 alleles, AF=8.31417e-05; highest reported population AF was 2.97147e-04 in Ashkenazi Jewish individuals.
• gnomAD v4.1 all-comers: 72/1604090 alleles, AF=4.48853e-05; highest reported population AF was 1.70207e-04 in Ashkenazi Jewish individuals.
• Generic ACMG/AMP 2015 stand-alone BA1 calibration supplied for this case: >=0.05 (ACMG/AMP 2015, PMID:25741868).

Sources

Reasoning

BS1 is not met. The all-comers gnomAD AF is 8.31417e-05 in v2.1 and 4.48853e-05 in v4.1; the highest reported ancestry-specific AF is 2.97147e-04 in v2.1 and 1.70207e-04 in v4.1, all below the generic 0.01 default. These frequencies do not create a threshold artifact requiring a maximum-credible-frequency calculation, and there is no evidence here of a reduced-penetrance founder allele that would alter interpretation.

Rule

Allele frequency is greater than expected for disorder.

Evidence considered

• gnomAD v2.1 all-comers: AF=8.31417e-05 overall and maximum reported ancestry AF=2.97147e-04 in Ashkenazi Jewish individuals.
• gnomAD v4.1 all-comers: AF=4.48853e-05 overall and maximum reported ancestry AF=1.70207e-04 in Ashkenazi Jewish individuals.
• The generic BS1 default supplied for this case is >=0.01; Whiffin et al. 2017 (PMID:28518168) is identified in the governing prompt as the preferred source for a per-gene maximum-credible-frequency derivation, but the observed frequency is far below the default and no such calculation is needed to conclude BS1 is not met.

Sources

Reasoning

BS3 is not met. The available variant-specific functional studies do not show no damaging effect on protein function or splicing; instead, they show aberrant SF3B1 K700E splicing, impaired SUGP1 association, and downstream cellular effects. Therefore, the evidence supports PS3 and provides no basis for a benign functional criterion.

Rule

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

Evidence considered

• PMID:27818134 reports increased mutation-associated alternative splicing in SF3B1 K700E models and altered downstream cellular phenotypes.
• PMID:31474574 reports upstream cryptic 3' splice-site usage and reduced SUGP1 association for K700E, with partial rescue after SUGP1 overexpression.

Sources

Reasoning

SF3B1 p.Lys700Glu is a missense variant, but the reviewed SF3B1 evidence describes recurrent missense hotspot disease biology and experimentally demonstrated K700E-associated splicing defects. This is contrary to a gene for which primarily truncating variants are known to cause disease; the available evidence does not support applying BP1.

Rule

Missense variant in a gene for which primarily truncating variants are known to cause disease.

Evidence considered

• PMID:25424858 reports recurrent SF3B1 K700E hotspot mutations and associated splicing changes.
• PMID:27818134 reports K700E-associated alternative-splicing and cellular effects.
• PMID:31474574 reports K700E-induced cryptic 3-prime splice-site usage and impaired SUGP1 association.

Sources

Reasoning

The variant is a missense change, so the governing generic rule selects the REVEL path only; SpliceAI is not used for PP3/BP4 in this missense situation. REVEL is 0.619, which is above the generic supporting BP4 threshold of <=0.29 and therefore provides no benign computational support. The result is gray-zone for BP4, so the SpliceAI max delta of 0.025 cannot be substituted. BayesDel is not used because no generic ACMG-strength calibration is established.

Rule

Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc) ... As many in silico algorithms use the same or very similar input for their predictions, each algorithm cannot be counted as an independent criterion.

Evidence considered

• Variant consequence: missense, NM_012433.4:c.2098A>G (NP_036565.2:p.(Lys700Glu)).
• Local REVEL lookup: score 0.619; the generic supporting BP4 threshold is <=0.29, using the supplied ClinGen SVI calibration (Pejaver et al. 2022, PMID:36413997).
• SpliceAI max delta 0.025 was not substituted because the applicable missense path is REVEL only; BayesDel was excluded because no generic ACMG-strength calibration is established.

Sources

Reasoning

The governing BP6 rule requires a reputable source to recently report the variant as benign, with the evidence unavailable to the laboratory for independent evaluation. Exact-variant ClinVar records show no expert-panel submissions and no expert-panel Benign or Likely benign classification. The available record is conflicting and contains no qualifying benign expert-panel assertion. BP6 is therefore not met.

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 variation 376004: exact variant NM_012433.4:c.2098A>G (p.Lys700Glu) has zero expert-panel submissions and no qualifying expert-panel Benign or Likely benign assertion.
• The aggregate ClinVar record is conflicting and ordinary laboratory submissions cannot trigger BP6 under the specified generic rule.

Sources

Reasoning

The variant is SF3B1 p.Lys700Glu (K700E), and publications document the same amino-acid change as a recurrent somatic hotspot. However, the supplied evidence does not establish a previously established pathogenic variant under the PS1 rule: the ClinVar record is conflicting and the literature evidence is primarily somatic, so PS1 cannot be assigned without a validated pathogenic comparator record.

Rule

Same amino acid change as a previously established pathogenic variant regardless of nucleotide change.

Evidence considered

• The case variant is NP_036565.2:p.(Lys700Glu) / p.(K700E).
• ClinVar variation 376004 is reported as conflicting, with Uncertain significance, Likely pathogenic, and Pathogenic submissions but no expert-panel submission.
• The literature reports K700E as a recurrent hotspot and COSMIC reports the variant 811 times; these observations establish recurrence but do not by themselves establish a prior pathogenic germline comparator for PS1.

Still missing

• A validated prior pathogenic classification for the same amino-acid change, independent of the conflicting ClinVar submissions, is needed.

Sources

Reasoning

No proband-level phenotype, family-history assessment, or parental testing results documenting a de novo occurrence are provided. In particular, there is no evidence that both maternity and paternity were confirmed, so the PS2 definition cannot be evaluated.

Rule

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

Evidence considered

• The reviewed case evidence contains no proband-level segregation record, parental genotypes, maternity confirmation, paternity confirmation, or family-history assessment for NM_012433.4:c.2098A>G (p.Lys700Glu).

Still missing

• Document the proband's disease phenotype and absence or presence of relevant family history.
• Provide variant testing results for both biological parents, including confirmation of maternity and paternity.

Reasoning

The governing ACMG/AMP PS4 rule requires a case-control prevalence comparison, with relative risk or odds ratio >5.0 and a confidence interval that does not include 1.0. The literature_pass.json PS4 leads (PMID:22918138, PMID:24976289, PMID:34131312) were checked against their recorded titles and available case materials: they are general sequencing, CML-management, and chromosomal-microarray guidance papers, not variant-specific affected-versus-unaffected case-control studies. PMID:25424858 reports SF3B1 K700E in 17 of 23 SF3B1-mutant tumors among 1,293 breast tumors, but provides no unaffected/control comparison for this exact variant. gnomAD v4.1 provides a population control frequency of 72/1,604,090 alleles (AF 4.48853e-05), but no affected-case count is available. Therefore an OR/RR and 95% CI cannot be calculated and PS4 cannot be assessed.

Rule

The prevalence of the variant in affected individuals is significantly increased compared to the prevalence in controls. ... Note 1: 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:25424858 paper_extracts entry: SF3B1 K700E was found in 17 of 23 SF3B1-mutant cases among 1,293 breast tumors; this is a case series/reanalysis without an unaffected control comparison for the exact variant.
• gnomAD v4.1: 72/1,604,090 alleles, AF 4.48853e-05; this supplies a population control frequency only, not the affected-case frequency required for OR/RR.
• The literature_pass PS4 leads PMID:22918138, PMID:24976289, and PMID:34131312 correspond respectively to a diagnostic genome-sequencing report, CML management recommendations, and an ACMG chromosomal-microarray technical standard; none is a variant-specific case-control study in the available case materials.

Still missing

• A variant-specific affected cohort and matched unaffected/control counts are needed to calculate OR or RR and its 95% CI.

Sources

Reasoning

PM3 cannot be assessed because the available evidence does not establish a recessive SF3B1 disorder, an affected proband carrying NM_012433.4:c.2098A>G, a second pathogenic variant, or parental/offspring testing demonstrating that the variants are in trans. The reviewed SF3B1 K700E literature is functional/somatic and explicitly does not provide variant-specific segregation, population-frequency, or patient-level inheritance data.

Rule

PM3: For recessive disorders, detected in trans with a pathogenic variant...Note: This requires testing of parents (or offspring) to determine phase.

Evidence considered

• Wang et al. (PMID:27818134) studies SF3B1 K700E in chronic lymphocytic leukemia and functional models but states that the paper does not provide variant-specific segregation or patient-level inheritance data for c.2098A>G.
• No available case evidence establishes a recessive disorder, an affected-proband allelic observation, a second pathogenic variant, or phase by parental/offspring testing.

Still missing

• Establish the relevant SF3B1 disorder and whether it is recessive.
• Document an affected proband carrying NM_012433.4:c.2098A>G and a pathogenic variant in the same gene.
• Obtain parental or offspring testing establishing phase as trans.

Sources

Reasoning

The variant is a missense change at residue 700, but the supplied same-residue comparator collection contains no candidates and no different missense variant at Lys700 is documented in the reviewed evidence as determined pathogenic. The recurrent K700E observations are the same amino-acid change, not PM5 evidence for a different missense change.

Rule

Novel missense change at an amino acid residue where a different missense change determined to be pathogenic has been seen before.

Evidence considered

• The PM5 candidate collection reports zero same-residue candidates and zero expert-panel or non-expert-majority pathogenic/likely pathogenic candidates.
• PMID:25424858 and COSMIC document K700E itself, which is not a different missense change required by PM5.

Still missing

• A documented different missense substitution at Lys700 with an established pathogenic determination is needed.

Sources

Reasoning

No assumed-de-novo case observation or parental testing result is documented. The available evidence therefore does not establish that the variant arose de novo without confirmation of maternity and paternity.

Rule

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

Evidence considered

• The reviewed case evidence contains no parental genotypes, inheritance description, or case-level statement that NM_012433.4:c.2098A>G (p.Lys700Glu) was assumed de novo.

Still missing

• Provide case-level evidence that the variant is absent from the tested parents or otherwise assumed de novo.
• Document whether maternity and paternity were confirmed.

Reasoning

No affected family members, variant-positive relatives, variant-negative relatives, or informative meioses are documented. Consequently, co-segregation with disease cannot be assessed.

Rule

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

Evidence considered

• The reviewed case evidence contains no pedigree, affected-family-member descriptions, relative genotypes, or meiosis count for NM_012433.4:c.2098A>G (p.Lys700Glu).

Still missing

• Provide a pedigree identifying affected relatives and their clinical phenotypes.
• Provide variant-testing results and the number of informative meioses for each relevant relative.

Reasoning

The evidence documents recurrent SF3B1 missense mutations and functional effects of K700E, but it does not provide a gene-wide assessment showing both a low rate of benign missense variation and that missense variants are a common disease mechanism under the PP2 rule. A single variant's gnomAD frequency and cancer-literature recurrence cannot establish both gene-level requirements.

Rule

Missense variant in a gene that has a low rate of benign missense variation and where missense variants are a common mechanism of disease.

Evidence considered

• gnomAD v4.1 reports this variant at AF 4.48853e-05, but this is a variant-level frequency and not a gene-wide benign missense rate.
• PMID:27818134 and PMID:31474574 document recurrent K700E-associated splicing effects, but they do not establish the required gene-wide low benign missense rate.

Still missing

• A gene-wide benign missense variation analysis and evidence defining the relevant disease mechanism are needed.

Sources

Reasoning

The governing PP4 rule requires that the patient's phenotype or family history be highly specific for a disease with a single genetic etiology. The available evidence describes somatic SF3B1 K700E in cancer cohorts and functional studies, but does not provide this patient's phenotype, family history, or a phenotype-specific diagnosis establishing a single genetic etiology. PP4 therefore cannot be assessed.

Rule

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

Evidence considered

• The available literature concerns SF3B1 K700E in somatic cancer cohorts or experimental models and does not document this patient's phenotype or family history.

Still missing

• Patient phenotype, family history, and a defined disease diagnosis with appropriate specificity are required.

Sources

Reasoning

BS2 cannot be assessed from the available evidence. gnomAD provides population allele counts but does not identify an individual carrier as a healthy adult with documented phenotype, age, penetrance, and ascertainment suitable for this criterion. The available records show zero homozygotes, but that does not establish or refute the BS2 requirement for an observed healthy adult carrier; the governing rule also requires full penetrance at an early age, which is not documented here for the relevant SF3B1-related disorder.

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 0 homozygotes among 276636 alleles; gnomAD v4.1 reports 0 homozygotes among 1604090 alleles.
• Population allele counts do not provide the individual-level healthy-adult, age, phenotype, or penetrance documentation required to apply BS2.

Still missing

• A documented healthy adult heterozygous carrier with age and phenotype information is unavailable.
• Full penetrance at an early age for the relevant disorder is not established in the case evidence.

Sources

Reasoning

No affected family members with documented variant status are described. Therefore, there is no family-based non-segregation observation on which to assess BS4.

Rule

Lack of segregation in affected members of a family.

Evidence considered

• The reviewed case evidence contains no affected-relative genotypes or family observations demonstrating that the variant fails to segregate with disease.

Still missing

• Provide a pedigree and clinical status for affected relatives.
• Provide reliable variant-testing results for affected relatives and relevant family members.

Reasoning

BP2 cannot be assessed because the available evidence does not establish a fully penetrant dominant SF3B1 disorder or provide a pathogenic variant in the same individual with phase information showing either trans or cis positioning. The reviewed SF3B1 K700E literature is functional/somatic and does not provide patient-level inheritance or segregation data for c.2098A>G.

Rule

BP2: 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

• Wang et al. (PMID:27818134) reports functional and somatic SF3B1 K700E findings but does not provide patient-level inheritance, segregation, or phase data for c.2098A>G.
• No available case evidence establishes a fully penetrant dominant disorder or a cis/trans relationship between this variant and a pathogenic variant.

Still missing

• Establish whether the relevant SF3B1 disorder is fully penetrant and dominant.
• Document a pathogenic variant in the same affected individual.
• Obtain phase testing demonstrating trans or cis configuration.

Sources

Reasoning

The governing BP5 rule requires that the variant be found in a case with an alternate molecular basis for disease. The available records do not document an individual case with an alternate pathogenic molecular diagnosis, nor do the reviewed cancer cohort and functional papers establish such a case for this variant. BP5 cannot be assessed.

Rule

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

Evidence considered

• No available source documents a patient carrying NM_012433.4:c.2098A>G together with an alternate molecular basis for disease.

Still missing

• A case report or clinical record documenting an alternate molecular diagnosis in the same affected individual is required.

Sources