FGFR3 c.1172C>T, p.(Ala391Val)

Reasoning

PM1 (Richards et al. 2015, PMID:25741868) is met for location in a critical and well-established functional domain. Residue Ala391 lies within the transmembrane domain of FGFR3: PMID:11426459 states the recurring Ala391Glu substitution is 'in the transmembrane domain of FGFR3' and that nearby transmembrane substitutions Gly380Arg and Gly375Cys cause achondroplasia; PMID:8880573 localizes ala391glu 'within the transmembrane region'. The transmembrane domain is a critical functional domain of this receptor tyrosine kinase (transmembrane signaling/activation), and it harbors multiple established disease-causing missense mutations (Gly375Cys, Gly380Arg, Ala391Glu). The variant is not in a statistically significant cancer hotspot (cancerhotspots.org query 'FGFR3 A391' returned no result), but PM1 is met on the functional-domain arm rather than the hotspot arm. No benign variation is documented in the transmembrane domain: the case variant is ultra-rare in gnomAD (v4.1 AF 7.44e-6, 0 homozygotes), and no common/population-frequency benign allele at this residue is reported in the reviewed sources. Note: no VCEP-defined FGFR3 domain map was available (generic ACMG fallback), so domain boundaries rest on the published literature.

Rule

acmg_pm1_critical_functional_domain

Evidence considered

• PMID:11426459: 'Ala391Glu substitution in the transmembrane domain of FGFR3' and 'Nearby substitutions in the transmembrane domain of FGFR3 (Gly380Arg and Gly375Cys) cause achondroplasia'
• PMID:8880573: ala391glu 'within the transmembrane region' of FGFR3
• cancerhotspots.org (evidence.json, hotspots): no statistically significant hotspot at FGFR3 A391 (PM1 met via functional-domain arm, not hotspot arm)
• gnomAD v4.1: AF 7.44e-6, 0 homozygotes - no evidence of benign population variation at this residue

Still missing

• No VCEP/CSPEC-defined FGFR3 domain map was available; transmembrane-domain boundaries inferred from literature (PMID:11426459, PMID:8880573)
• Benign variant load of the FGFR3 transmembrane domain not quantitatively assessed beyond the case variant's ultra-rare gnomAD frequency

Sources

Reasoning

PM2 (extremely low/absent frequency in population databases) is met at supporting strength. gnomAD v4.1 total AF is 7.44e-06 (0.00074%; 12/1,612,512 alleles, 0 homozygotes) and gnomAD v2.1 total AF is 3.56e-06 (0.00036%; 1/281,004 alleles, 0 homozygotes), both well below the non-VCEP 0.1% PM2 threshold. grpmax FAF in gnomAD v4.1 is 1.83e-06 (0.00018%), and the highest subpopulation AF (MID, 0.0165%, single allele in 6,052) remains below 0.1% despite being a small-sample artifact. For a rare autosomal dominant FGFR3 craniosynostosis phenotype, this frequency is consistent with a pathogenic allele. PM2 applied at supporting strength per ClinGen SVI recommendation; consistent with ClinVar lab submissions (GeneDx, Labcorp) that invoked gnomAD absence/low-frequency for this variant.

Rule

PM2: Absent in controls (or at extremely low frequency if recessive) (Richards 2015, PMID:25741868); operationalized as AF <0.1% for non-VCEP gnomAD in this pipeline; strength set to supporting per ClinGen SVI recommendation

Evidence considered

• gnomAD v4.1: total AF 7.44e-06 (0.00074%), grpmax FAF 1.83e-06, 0 homozygotes
• gnomAD v2.1: total AF 3.56e-06 (0.00036%), 0 homozygotes
• gnomAD v4.1 subpopulations: MID 1/6,052 (0.0165%), EAS 1/44,884, SAS 1/90,940, NFE 6/1,179,488; all <0.1%

Still missing

• gnomAD-Canada v1.0 unavailable (HTTP 500); frequency not corroborated in Canadian genomes
• No African/African American, Admixed American, Finnish, or Ashkenazi Jewish alleles observed; large AFrican ancestry sample (75,044) supports absence

Sources

Reasoning

PM5 (Richards et al. 2015, PMID:25741868) is met: the case variant is a novel missense change (p.Ala391Val) at an amino acid residue where a different missense change, p.Ala391Glu, has been determined to be pathogenic. PMID:8880573 (Wilkes et al. 1996) demonstrates that a recurrent mutation, ala391glu, causes Crouzon syndrome with acanthosis nigricans (CAN) in three unrelated patients, resulting from a gcg-to-gag transversion (C to A at the middle base of the Ala391 codon) within the transmembrane region. PMID:11426459 (Schweitzer et al. 2001) reports the recurring heterozygous C-to-A transversion at nucleotide 1,172 producing Ala391Glu identified in 10 patients with CAN. PMID:20199409 (Sharda et al. 2010) confirms Ala391Glu by sequencing in a newborn with CAN. The comparator is therefore established as pathogenic/recurrent and syndrome-specific (same residue 391, different amino acid Glu vs Val). The case change is a different missense (Val) at the same codon/residue and is not established benign (ClinVar VUS; gnomAD v4.1 AF 7.44e-6, no homozygotes). All three conditions of PM5 (established pathogenic different change at same residue; novel change; no established benign status) are satisfied.

Rule

acmg_pm5_different_missense_at_residue_with_established_pathogenic_change

Evidence considered

• PMID:8880573: 'A recurrent mutation, ala391glu, causes Crouzon syndrome and acanthosis nigricans' - 'a specific gcg to gag transversion, resulting in an amino acid substitution ala391glu within the transmembrane region' (3 unrelated patients)
• PMID:11426459: 'A recurring mutation (heterozygous C to A transversion at nucleotide 1,172) resulting in an Ala391Glu substitution in the transmembrane domain of FGFR3 has been identified in 10 patients with CAN'
• PMID:20199409: 'Molecular analysis by sequencing confirmed Ala391Glu substitution' in a newborn with CAN
• ClinVar VCV000809609: case variant p.Ala391Val = Uncertain significance (not established benign)
• gnomAD v4.1: AF 7.44e-6, 0 homozygotes - no established benign status

Still missing

• Pathogenicity of the comparator Ala391Glu rests on recurrent syndrome-specific literature rather than a retrieved ClinVar expert-panel (3-star) record for the comparator itself

Sources

Reasoning

PP2 (Richards et al. 2015, PMID:25741868) is met at supporting strength. Missense variants are a common, indeed the dominant, mechanism of FGFR3 germline disease: the reviewed literature documents disease-causing missense mutations across multiple FGFR3 domains and phenotypes, including Lys650Met (thanatophoric dysplasia type I, PMID:11426459), Gly380Arg and Gly375Cys (achondroplasia, PMID:11426459), and Ala391Glu (Crouzon syndrome with acanthosis nigricans, PMID:8880573, PMID:11426459, PMID:20199409). FGFR3 germline disorders (achondroplasia, thanatophoric dysplasia, CAN, Muenke syndrome) are gain-of-function and caused by activating missense variants rather than truncating variants. Benign missense variation in FGFR3 appears low: the case variant is ultra-rare (gnomAD v4.1 AF 7.44e-6, 0 homozygotes, grpmax FAF 1.83e-6) and no common benign missense allele at this residue is documented in the reviewed sources. Caveat: no quantitative missense constraint metric (e.g., gnomAD missense Z-score) was retrieved for FGFR3 in this case, so the low-benign-missense-rate component is inferred from the missense-dominant disease mechanism and population frequency evidence.

Rule

acmg_pp2_missense_common_mechanism_low_benign_missense

Evidence considered

• PMID:11426459: multiple disease-causing FGFR3 missense mutations - Lys650Met (TD1), Gly380Arg and Gly375Cys (achondroplasia), Ala391Glu (CAN)
• PMID:8880573: recurrent missense ala391glu causes Crouzon syndrome with acanthosis nigricans
• PMID:20199409: FGFR3 Ala391Glu missense confirmed in CAN case
• gnomAD v4.1: AF 7.44e-6, 0 homozygotes, grpmax FAF 1.83e-6 - ultra-rare, consistent with low benign missense rate

Still missing

• No quantitative missense constraint metric (e.g., gnomAD missense Z-score / missense observed-expected ratio) was retrieved for FGFR3; low benign missense rate inferred from mechanism and allele frequency evidence

Sources

Reasoning

Generic ACMG/AMP 2015 (PMID:25741868) BP4: multiple lines of computational evidence suggest no impact on the gene or gene product. Three independent calibrated computational lines are concordant in predicting no impact for FGFR3 c.1172C>T (p.Ala391Val): (1) splice path - SpliceAI max delta score 0.01 (DS_AG 0.0, DS_AL 0.0, DS_DG 0.01, DS_DL 0.0), well below the 0.2 threshold for splice-altering; (2) missense path - REVEL 0.162, clearly in the benign range; (3) missense path - BayesDel -0.317094, clearly in the benign/deleterious-negative range. All three lines agree on no impact, meeting the 'multiple lines' requirement without double-counting any single prediction across PP3/BP4/BP7 (BP7 not applicable - variant is missense, not synonymous). BP4 applied at supporting strength, the standard strength for this criterion under the generic framework. The splice-impact path (SpliceAI) and missense in-silico path (REVEL/BayesDel) are both owned by this group and both point to no impact.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID:25741868) BP4: multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.). Applied at supporting strength.

Evidence considered

• SpliceAI max delta score = 0.01 (DS_AG 0.0, DS_AL 0.0, DS_DG 0.01, DS_DL 0.0) -> no predicted splice impact
• REVEL score = 0.162 -> low/benign range
• BayesDel noAF score = -0.317094 -> negative/benign range

Still missing

• No gene-specific VCEP-calibrated BP4 thresholds for FGFR3; generic ACMG/AMP thresholds applied (REVEL < 0.3-0.5 benign, BayesDel negative, SpliceAI < 0.2 no splice impact).

Sources

Reasoning

PS1 (Richards et al. 2015, PMID:25741868) requires the SAME amino acid change as a previously established pathogenic variant regardless of nucleotide change. The case variant is NM_000142.4:c.1172C>T (p.Ala391Val). The only well-established pathogenic change at residue 391 is p.Ala391Glu (recurrent Crouzon syndrome with acanthosis nigricans mutation, 10+ patients across PMID:8880573, PMID:11426459, PMID:20199409), which is a DIFFERENT amino acid change (Glu vs Val) at the same residue. No source establishes p.Ala391Val itself as pathogenic: ClinVar (VCV000809609) classifies it as Uncertain significance (3 clinical laboratories, no expert panel). Because the amino acid substitution differs from the established pathogenic change, PS1 is not met; the same-residue comparator relationship is instead captured by PM5. No evidence supports that the exact same amino acid change (Ala391Val) has been previously established as pathogenic.

Rule

acmg_ps1_same_amino_acid_change_established_pathogenic

Evidence considered

• ClinVar VCV000809609: p.Ala391Val classified Uncertain significance (3 clinical laboratories; no expert panel submissions)
• PMID:11426459: established recurring mutation at codon 391 is Ala391Glu (C to A transversion at nucleotide 1,172), not Ala391Val
• PMID:8880573: recurrent ala391glu within the transmembrane region of FGFR3

Sources

Reasoning

Variant FGFR3 NM_000142.4:c.1172C>T (p.Ala391Val) is a missense substitution in exon 9 (transmembrane domain). No ClinGen CSPEC or VCEP specification exists for FGFR3 (cspec found=false; vcep_materials empty; framework_mode=generic_acmg), so generic ACMG/AMP 2015 (PMID:25741868) applies. PP3 requires multiple lines of computational evidence supporting a deleterious effect. All calibrated computational evidence for this variant instead predicts NO impact: SpliceAI max delta score = 0.01 (DS_AG 0.0, DS_AL 0.0, DS_DG 0.01, DS_DL 0.0; far below the 0.2 splice-impact threshold); REVEL = 0.162 (benign range); BayesDel = -0.317094 (negative/benign range). No computational line supports a deleterious effect, so PP3 is not met. A GeneDx ClinVar submission (SCV005079325) likewise stated in silico analysis 'does not alter protein structure/function'; per instructions, criterion was adjudicated from raw tool scores, not from another submitter's classification. No publication retrieved mentions the exact variant, so no paper-based computational evidence exists for PP3.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID:25741868) PP3: multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.). Requires concordant in-silico evidence of impact; none present here.

Evidence considered

• SpliceAI max delta score = 0.01 (DS_AG 0.0, DS_AL 0.0, DS_DG 0.01, DS_DL 0.0) -> no predicted splice impact
• REVEL score = 0.162 -> low/benign range
• BayesDel noAF score = -0.317094 -> negative/benign range

Still missing

• No gene-specific VCEP-calibrated PP3 thresholds exist for FGFR3; generic ACMG/AMP thresholds applied. Irrelevant to outcome because all three tools are far in the benign range.
• aGVGD not available in this case's evidence bundle; REVEL and BayesDel provide the missense in-silico lines and are concordant.

Sources

Reasoning

Per governing instruction, PP5 applies only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic. The ClinVar submission audit for VCV000809609 (clinvar) shows expert_panel_submissions = 0; all three submissions are single-submitter clinical laboratories (GeneDx, Labcorp/Invitae, CeGaT) with 'criteria provided, single submitter' review status, and the aggregate ClinVar classification is Uncertain significance (2-star, 'criteria provided, multiple submitters, no conflicts'). Ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions must never trigger PP5. No expert-panel classification exists for this variant, so PP5 is not met. PP5 is not used as a substitute for direct evidence, per the expert-panel-gated policy.

Rule

acmg_amp_2015_pp5 (expert-panel-gated: exact-variant ClinVar expert-panel Pathogenic/Likely pathogenic only)

Evidence considered

• ClinVar submission audit (clinvar): 3 submissions, 0 expert panel, classification VUS, review status single-submitter/2-star
• prefetch.json clinvar.submission_audit_summary: expert_panel_submissions=0, usable_for_criterion_leads=1 (Labcorp submission, non-expert)

Sources

Reasoning

BA1 (allele frequency >1%, non-VCEP gnomAD threshold) is not met. Highest observed allele frequency for FGFR3 c.1172C>T (p.Ala391Val) is 0.0165% (Middle Eastern, gnomAD v4.1, 1/6,052 alleles) and total AF is 0.00074% (12/1,612,512) in gnomAD v4.1 and 0.00036% (1/281,004) in gnomAD v2.1. All frequencies are >3 orders of magnitude below the 1% BA1 threshold. No population database approaches a common-polymorphism frequency.

Rule

BA1: Allele frequency is >5% in ExAC/1000 Genomes/ESP (Richards 2015, PMID:25741868); operationalized as >1% for non-VCEP gnomAD in this pipeline

Evidence considered

• gnomAD v2.1: total AF 3.56e-06 (0.00036%), max subpopulation AF 7.81e-06 (NFE, 0.00078%)
• gnomAD v4.1: total AF 7.44e-06 (0.00074%), max subpopulation AF 1.65e-04 (MID, 0.0165%), grpmax FAF 1.83e-06
• gnomAD-Canada v1.0: not available (HTTP 500)

Still missing

• gnomAD-Canada v1.0 unavailable (HTTP 500); Canadian population frequency not confirmed

Sources

Reasoning

BS1 (allele frequency >0.3%, non-VCEP gnomAD threshold) is not met. The highest subpopulation AF is 0.0165% (Middle Eastern, gnomAD v4.1, driven by a single allele in 6,052 chromosomes), and total AF is 0.00074% (gnomAD v4.1) / 0.00036% (gnomAD v2.1) — all far below the 0.3% threshold. FGFR3 Crouzon syndrome with acanthosis nigricans is an autosomal dominant, very rare disorder (craniosynostosis syndromes ~1/25,000-1/60,000 births, CAN a small subset); 12 alleles with 0 homozygotes in gnomAD v4.1 is compatible with a rare, de novo-enriched dominant disease allele and does not support an allele frequency greater than expected for the disorder.

Rule

BS1: Allele frequency is greater than expected for disorder (Richards 2015, PMID:25741868); operationalized as >0.3% for non-VCEP gnomAD in this pipeline

Evidence considered

• gnomAD v2.1: max subpopulation AF 7.81e-06 (NFE, 0.00078%)
• gnomAD v4.1: max subpopulation AF 1.65e-04 (MID, 0.0165%)
• gnomAD v4.1: 12 total alleles, 0 homozygotes

Still missing

• gnomAD-Canada v1.0 unavailable (HTTP 500)
• No disease-specific expected allele frequency calculation available; assessment rests on operational >0.3% threshold

Sources

Reasoning

BS2 (observed in a healthy adult for a fully penetrant autosomal dominant disorder expected to manifest at an early age) is not met. No homozygotes are observed in gnomAD v2.1 (0/281,004) or gnomAD v4.1 (0/1,612,512). The 12 heterozygous carriers in gnomAD v4.1 (0.00074%) lack phenotype documentation; gnomAD is not a phenotype-verified healthy-adult cohort for craniosynostosis, so carrier presence alone does not satisfy BS2's requirement of documented observation in healthy adults. ClinVar contains no report of homozygous or healthy adult carriers. If phenotype-verified healthy adult carrier data became available, BS2 would merit re-evaluation.

Rule

BS2: Observed in a healthy adult individual for a recessive (homozygous), autosomal dominant (heterozygous), or X-linked (hemizygous) disorder, and full penetrance expected at an early age (Richards 2015, PMID:25741868)

Evidence considered

• gnomAD v2.1: 0 homozygotes in 281,004 alleles
• gnomAD v4.1: 0 homozygotes in 1,612,512 alleles; 12 heterozygous carriers without phenotype data
• ClinVar: 3 submissions (VUS), no homozygous/healthy-adult carrier reports

Still missing

• No phenotype-verified healthy adult carrier observations available for this variant
• gnomAD v4.1 carrier count (12) cannot be resolved for phenotype status

Sources

Reasoning

BP1 (Richards et al. 2015, PMID:25741868) requires a gene for which primarily truncating variants are known to cause disease. FGFR3 germline disease is instead caused predominantly by activating (gain-of-function) missense mutations - achondroplasia (Gly380Arg, Gly375Cys), thanatophoric dysplasia (Lys650Met), Crouzon syndrome with acanthosis nigricans (Ala391Glu), Muenke syndrome (Pro250Arg) - with no established truncating-variant mechanism for the FGFR3 germline skeletal/craniosynostosis phenotypes (PMID:11426459, PMID:8880573). Missense is the disease mechanism, so BP1 is not met. Note: the pvs1_gene_context.json artifact asserts 'lof_mechanism_supported: true', but its supporting citations (CHEK2/ATM germline cancer paper, sperm-sequencing paper, RAPGEF6 paper) are unrelated to FGFR3 disease biology and contradict the established FGFR3 gain-of-function missense mechanism; that artifact was disregarded for this criterion.

Rule

acmg_bp1_truncating_variant_mechanism_gene

Evidence considered

• PMID:11426459: FGFR3 germline disorders (ACH via Gly380Arg/Gly375Cys, TD via Lys650Met, CAN via Ala391Glu) are caused by activating missense mutations
• PMID:8880573: recurrent missense ala391glu causes Crouzon syndrome with acanthosis nigricans
• pvs1_gene_context.json reviewed and rejected as unreliable for mechanism: its LoF-support citations (PMID:35674998 CHEK2/ATM, PMID:41062690 sperm sequencing, PMID:42274819 RAPGEF6) are unrelated to FGFR3; FGFR3 germline disease is gain-of-function/missense

Sources

Reasoning

Per governing instruction, BP6 applies only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign. The ClinVar submission audit for VCV000809609 (clinvar) shows expert_panel_submissions = 0; all three submissions are single-submitter clinical laboratories and the aggregate classification is Uncertain significance. No expert-panel benign classification exists for this variant, and ordinary laboratory submissions/aggregate ClinVar labels must never trigger BP6. Therefore BP6 is not met.

Rule

acmg_amp_2015_bp6 (expert-panel-gated: exact-variant ClinVar expert-panel Benign/Likely benign only)

Evidence considered

• ClinVar submission audit (clinvar): 3 submissions, 0 expert panel, classification VUS
• prefetch.json clinvar.submission_audit_summary: expert_panel_submissions=0

Sources

Reasoning

PS2 (de novo occurrence with confirmed parentage) requires documented parental testing showing that c.1172C>T arose de novo in a proband. No such data exists for this exact variant in any retrieved source: the evidence bundle contains no proband-parent trio or parental-genotype information (no 'de novo', 'segregation', 'parental', 'meiosis', or 'pedigree' content anywhere in review_manifest/case_summary/prefetch/evidence/literature_pass/vcep_materials), and the ClinVar record VCV000809609 (3 clinical-lab submissions, all germline origin, VUS, criteria provided/single submitter, no expert panel) carries no de novo assertion (submission-audit evidence signals are only PP3 in SCV005079325 and PM2 in SCV004451705). The only de novo observation in the retrieved literature (PMID:8880573 case 1: 'The parents of case 1 showed the normal wild type pattern, confirming the de novo nature of his mutation') concerns the distinct c.1172C>A (p.Ala391Glu) allele at the same nucleotide, not c.1172C>T (p.Ala391Val), and per exact-variant citation grounding cannot be applied to this variant. Governing framework: generic ACMG/AMP 2015 (Richards et al., PMID:25741868) because no FGFR3 CSPEC/VCEP specification exists (cspec.found=false; framework_mode=generic_acmg_fallback). PS2 not assessed pending variant-specific parental-testing data.

Rule

ACMG/AMP 2015 (Richards et al. 2015): PS2 = de novo (paternity and maternity confirmed) in a patient with the disease and no family history. No gene-specific FGFR3 CSPEC/VCEP specification available; generic rule applies.

Evidence considered

• ClinVar VCV000809609: 3 germline VUS submissions (GeneDx SCV005079325, Labcorp/Invitae SCV004451705, CeGaT SCV001154148); no de novo/parental assertion; audit signals only PP3/PM2
• PMID:8880573: de novo observation in case 1 applies to Ala391Glu (c.1172C>A), a different allele at nucleotide 1172; not usable for this variant
• PMID:11426459, PMID:20199409, PMID:7493034: case reports of Ala391Glu (c.1172C>A) only; no parental testing for Ala391Val
• grep sweep of evidence bundle: zero de novo/segregation/parental/meiosis/pedigree content for this case

Still missing

• No parental testing (trio/proband-parent) results reported for any proband carrying c.1172C>T (p.Ala391Val); a confirmed de novo observation for this exact variant would be required to meet PS2.

Sources

Reasoning

No well-established functional study of FGFR3 p.(Ala391Val) (c.1172C>T) exists in the case evidence, so PS3 cannot be met. Governing framework: generic ACMG/AMP 2015 (no FGFR3 CSPEC/VCEP available; framework_mode=generic_acmg_fallback). OncoKB curates the biological-effect context of A391V as 'Inconclusive' and cites two functional papers, but neither contains experimental data for this variant: PMID:34272467 (MANO-based evaluation of 160 COSMIC-recurrent FGFR1-4 mutations plus fusions) has zero occurrences of 391/A391/Ala391/1172 in the full text (FGFR3 transmembrane variants it tested were G370C, S371C, Y373C, G380E/R), and PMID:29533785 (pooled Ba/F3 + MCF10A viability functional screen) does not assay FGFR3 at all (zero FGFR3 mentions). COSMIC has no somatic record for c.1172C>T, consistent with A391V not being a recurrent cancer mutation. ClinVar-cited papers (PMID:7493034, PMID:8880573, PMID:11426459, PMID:20199409) concern the distinct Ala391Glu allele and contain no functional assay of Ala391Val (no in-vitro/transfection/assay content). In-silico predictors (REVEL 0.162, BayesDel -0.317094, SpliceAI max delta 0.01) are predictions, not functional studies, and cannot satisfy PS3. PS3 therefore not assessed pending variant-specific functional data.

Evidence considered

• OncoKB FGFR3/A391V page: biological effect 'Inconclusive', curated oncogenicity 'Inconclusive' (no assay result)
• PMID:34272467 full text: A391V absent from the 160 tested FGFR mutations
• PMID:29533785 full text: FGFR3 not among genes assayed
• COSMIC: no somatic report for FGFR3 c.1172C>T
• ClinVar-cited A391E papers (PMID:7493034, PMID:8880573, PMID:11426459, PMID:20199409): different allele, no functional assay of A391V

Still missing

• No validated variant-specific functional assay for p.Ala391Val (e.g., FGFR3 phosphorylation/signaling readout, STAT1 activation, receptor dimerization, or cell-proliferation/transformation assay) is available in the evidence bundle; external literature would be needed to fill this gap.

Sources

Reasoning

PS4 (ACMG/AMP 2015, PMID 25741868) requires the prevalence of the variant in affected individuals to be significantly increased compared with controls, supported by a case-control or large cohort study of the exact variant. No such study exists for NM_000142.4:c.1172C>T (p.Ala391Val). The four ClinVar-cited case papers (PMID:7493034, PMID:8880573, PMID:11426459, PMID:20199409) were reviewed and all describe p.Ala391Glu (c.1172C>A, 'GCG to GAG transversion at nucleotide 1,172') in Crouzon syndrome with acanthosis nigricans, a different nucleotide and amino-acid change at the same codon; none mention c.1172C>T or p.Ala391Val, so they cannot provide enrichment or prevalence data for this variant (grounding rule: variant must be mentioned in the source). ClinVar (clinvar) reports only VUS submissions from 3 single-submitter clinical laboratories with no phenotype or case-count data, which cannot be used for prevalence. The OncoKB functional papers (PMID:29533785, PMID:34272467) were screened and contain no variant-specific phenotype/prevalence data (PMID:29533785 has no FGFR3 mention; PMID:34272467 has no A391 mention). gnomAD allele counts (gnomad_v2, gnomad_v4) are population frequency data assessed by the population-frequency group, not case-control enrichment. Evidence is insufficient to assess PS4.

Rule

acmg_amp_2015_ps4 (generic fallback; no VCEP/CSPEC for FGFR3)

Evidence considered

• Full-text review of PMID:11426459, PMID:8880573, PMID:20199409 and abstract of PMID:7493034 establishing that all describe p.Ala391Glu (c.1172C>A), not c.1172C>T (p.Ala391Val)
• ClinVar submission audit (clinvar): 3 single-submitter VUS submissions, conditions not provided, expert_panel_submissions=0
• Literature pass criteria index flagged PMID:11426459 and PMID:7493034 as PS4 candidates; both concern the different variant p.Ala391Glu
• Screening of PMID:29533785 and PMID:34272467: no variant-specific phenotype/prevalence content

Still missing

• No case-control or large cohort prevalence study of the exact variant c.1172C>T (p.Ala391Val) is available; the ClinVar-cited literature concerns the different variant p.Ala391Glu (c.1172C>A) and cannot be applied.
• No proband case counts or control allele counts for this variant are documented anywhere in the case bundle.

Sources

Reasoning

PM6 (assumed de novo without confirmation of parentage) requires evidence of a de novo event for c.1172C>T in a proband even without formal maternity/paternity confirmation. No proband-parent sequencing or segregation data for this exact variant exists in the evidence bundle, and no ClinVar submission asserts a de novo origin for VCV000809609 (all three submissions: germline origin, VUS, no de novo signal in the submission audit). The de novo event documented in PMID:8880573 case 1 pertains to the different allele c.1172C>A (p.Ala391Glu), so it cannot support PM6 for p.Ala391Val under exact-variant citation grounding. Generic ACMG/AMP 2015 applies (no FGFR3 CSPEC/VCEP). PM6 not assessed pending any parental-testing data for this exact variant.

Rule

ACMG/AMP 2015 (Richards et al. 2015): PM6 = assumed de novo, but without confirmation of paternity and maternity. No gene-specific FGFR3 CSPEC/VCEP specification available; generic rule applies.

Evidence considered

• ClinVar VCV000809609: no de novo origin asserted in any of the 3 submissions
• PMID:8880573: de novo observation is for Ala391Glu (c.1172C>A), not Ala391Val
• PMID:20199409: single newborn case of Ala391Glu (c.1172C>A); no parental testing data

Still missing

• No assumed-de-novo evidence (unconfirmed-parentage trio data) for c.1172C>T (p.Ala391Val); parental genotypes are not reported for any carrier of this variant.

Sources

Reasoning

PP1 (co-segregation with disease in affected family members) requires the variant to segregate with the phenotype across multiple affected relatives. No segregation data for c.1172C>T (p.Ala391Val) exists in the evidence bundle: no affected family members were reported tested for this exact variant in any retrieved source, and the ClinVar record contains no segregation assertions. The familial/segregation observations in the retrieved literature are exclusively for the distinct allele c.1172C>A (p.Ala391Glu) — e.g., PMID:8880573 'the affected daughter of case 3 also shares his mutation' and PMID:11426459 'were sporadic except for two mother-daughter pairs and one mother-son pair' — and cannot be applied to p.Ala391Val under exact-variant citation grounding. Generic ACMG/AMP 2015 applies (no FGFR3 CSPEC/VCEP). PP1 not assessed pending segregation studies of this exact variant.

Rule

ACMG/AMP 2015 (Richards et al. 2015): PP1 = co-segregation with disease in affected family members in multiple family members. No gene-specific FGFR3 CSPEC/VCEP specification available; generic rule applies.

Evidence considered

• ClinVar VCV000809609: no segregation assertions in any submission
• PMID:8880573: affected daughter of case 3 shares the Ala391Glu (c.1172C>A) mutation — different allele, not c.1172C>T
• PMID:11426459: familial mother-daughter/mother-son pairs all carry Ala391Glu (c.1172C>A) — different allele
• PMID:7493034: first Ala391Glu report; full text not retrievable; no Ala391Val segregation data

Still missing

• No multi-generational family segregation study for c.1172C>T; no affected relatives tested for this exact variant, so the number of informative meioses is zero.

Sources

Reasoning

PP4 (ACMG/AMP 2015, PMID 25741868) requires the proband's phenotype or family history to be highly specific for a gene with a single genetic etiology. No proband phenotype or family-history data are present in this case bundle: all three ClinVar submissions for VCV000809609 list conditions as 'not provided' (clinvar, clinvar_submission_audit), the case summary contains no phenotype section, and no clinical report is available. FGFR3 is associated with multiple distinct phenotypes (achondroplasia, hypochondroplasia, thanatophoric dysplasia, Crouzon syndrome with acanthosis nigricans, etc.), so a highly specific phenotype match cannot be established without the proband's clinical findings. The ClinVar-cited papers describe the CAN phenotype of p.Ala391Glu (a different variant) and therefore cannot support PP4 for c.1172C>T (p.Ala391Val). Evidence is insufficient to assess PP4.

Rule

acmg_amp_2015_pp4 (generic fallback; no VCEP/CSPEC for FGFR3)

Evidence considered

• ClinVar submission audit (clinvar): conditions 'not provided' across all submissions
• Case literature reviewed for phenotype context; all describe p.Ala391Glu (c.1172C>A), a different variant

Still missing

• No proband phenotype or family-history data provided in the case bundle (ClinVar conditions recorded as 'not provided').
• No clinical report describing the phenotype of an individual carrying c.1172C>T (p.Ala391Val).

Sources

Reasoning

No well-established functional study demonstrates a lack of damaging effect for p.(Ala391Val), so BS3 cannot be met. BS3 requires well-established functional studies showing no damaging effect on gene/protein function; absence of functional evidence is not supportive. OncoKB 'Inconclusive' is not evidence of normal function, and in-silico-only signals (REVEL 0.162, BayesDel -0.317094, SpliceAI max delta 0.01) are predictions, not functional assays. The ultra-low population frequency (gnomAD v2.1 AF 3.6e-06; v4.1 AF 7.4e-06, grpmax FAF 1.83e-06) is a population observation evaluated under PM2/BS1-type criteria elsewhere and is not a functional readout. The retrieved functional papers (PMID:34272467, PMID:29533785) do not test A391V, and the ClinVar-cited A391E papers are a different substitution without functional assays. BS3 therefore not assessed pending variant-specific functional data.

Evidence considered

• OncoKB FGFR3/A391V page: biological effect 'Inconclusive' (no normal-function evidence)
• SpliceAI: max delta score 0.01 (in-silico splice prediction only; not a functional study)
• REVEL 0.162 and BayesDel -0.317094 (in-silico missense predictors; not functional studies)
• PMID:34272467 full text: A391V not tested
• PMID:29533785 full text: FGFR3 not assayed
• ClinVar-cited A391E papers: different allele, no functional assay of A391V

Still missing

• No functional assay data showing wild-type-like (normal) function of p.Ala391Val (e.g., signaling/proliferation readouts comparable to wild-type FGFR3), which would be required before BS3 could be applied.

Sources

Reasoning

BS4 (non-segregation with disease in affected family members) requires evidence that the variant does not segregate with disease — e.g., an affected family member who does not carry the variant, or (for recessive conditions) presence in trans with a pathogenic variant. No non-segregation data for c.1172C>T (p.Ala391Val) exists in any retrieved source: no affected family members were tested and found negative for this exact variant, and no unaffected-carrier segregation information is reported. The population presence of the variant (gnomAD v2.1 AF 3.6e-06, 1/281004 alleles; v4.1 AF 7.4e-06) is a population-frequency observation evaluated under BA1/BS1/PM2-type criteria elsewhere and does not constitute BS4 non-segregation evidence. The familial data in the ClinVar-cited papers concerns the different allele Ala391Glu (c.1172C>A) and is irrelevant to BS4 for this variant. Generic ACMG/AMP 2015 applies (no FGFR3 CSPEC/VCEP). BS4 not assessed pending non-segregation data for this exact variant.

Rule

ACMG/AMP 2015 (Richards et al. 2015): BS4 = non-segregation with disease in affected family members. No gene-specific FGFR3 CSPEC/VCEP specification available; generic rule applies.

Evidence considered

• ClinVar VCV000809609: no non-segregation assertion in any of the 3 submissions
• gnomAD v2.1 / v4.1: variant present at very low AF (3.6e-06 / 7.4e-06); population-frequency data, not BS4 segregation data
• PMID:8880573, PMID:11426459: familial observations all concern Ala391Glu (c.1172C>A), not c.1172C>T; not usable for BS4

Still missing

• No non-segregation evidence for c.1172C>T: no affected family member tested negative for this variant, and no segregation study reporting absence in affected relatives is available.

Sources

Reasoning

BP2 (Richards et al. 2015) requires an actual allelic observation: the variant observed in trans with a pathogenic variant for a fully dominant disease/mechanism, or observed in cis with a pathogenic variant in any inheritance pattern. The disease context is fully dominant (FGFR3 CAN/achondroplasia family, autosomal dominant), so the criterion is in principle applicable; however, no source in the evidence bundle provides any phase or co-occurrence information for c.1172C>T. ClinVar (VariationID 809609) shows 3 clinical-laboratory submissions (VUS, no expert panel) with no phase/family data and no allelic criteria cited (Labcorp cited PM2, GeneDx cited PP3); gnomAD allele counts (v2.1: 1/281004; v4.1: 12/1612512, 0 homozygotes) carry no phase information; and none of the six full-text publications mention this variant at all (all germline papers concern the different allele p.Ala391Glu). With zero observations of the variant in trans or in cis with a pathogenic variant, the criterion cannot be evaluated: insufficient evidence, returned not_assessed with the data gap identified.

Rule

BP2 (generic ACMG/AMP 2015, Richards et al., PMID:25741868, applied via generic_acmg_combination_rules fallback since no FGFR3 CSPEC/VCEP exists): 'Observed in trans with a pathogenic variant for a fully dominant disease/mechanism, or observed in cis with a pathogenic variant in any inheritance pattern.'

Evidence considered

• clinvar

Still missing

• No proband or family phase data: no observation of c.1172C>T in trans (or in cis) with a known pathogenic FGFR3 variant
• No co-occurrence/pedigree data in ClinVar submissions or any retrieved publication
• Requires targeted variant-level family testing or a clinical record documenting allele phase with a pathogenic FGFR3 allele

Sources

Reasoning

BP5 (ACMG/AMP 2015, PMID 25741868) applies when the variant is found in a case with an alternate molecular basis for disease. No proband-level genetic workup data are present in this case bundle: no alternate causative variant, no phenotype, and no family studies are documented; ClinVar conditions are 'not provided' (clinvar). Because there is no proband case data at all, it cannot be established whether an alternate molecular basis exists (or does not exist), so BP5 cannot be assessed. Evidence is insufficient.

Rule

acmg_amp_2015_bp5 (generic fallback; no VCEP/CSPEC for FGFR3)

Evidence considered

• ClinVar submission audit (clinvar): no phenotype/indication data, no alternate-diagnosis information
• Case bundle review: no proband workup artifact present

Still missing

• No proband-level genetic workup or case report documenting the presence or absence of an alternate molecular basis for disease.
• No proband phenotype or indication data in the case bundle (ClinVar conditions 'not provided').

Sources