FGFR3 c.746C>G, p.Ser249Cys

Reasoning

PS3 is met at moderate strength. Two independent, variant-specific functional studies directly evaluated FGFR3 p.Ser249Cys/c.746C>G using complementary systems and wild-type comparisons. In the human bladder cancer cell line 97-7, the mutant receptor formed stable dimers, had higher phosphorylation than wild type, and supported proliferation, colony formation, and anchorage-independent growth; FGFR3 knockdown reduced these phenotypes and mutant re-expression rescued them (PMID:17384684). In a quantitative FRET assay using transfected CHO-cell plasma-membrane vesicles, the variant increased ligand-independent dimerization (dimerization constant 500 ± 100 versus 290 ± 60 for wild type; stabilization −0.3 ± 0.1 kcal/mol) and altered dimer structure, although the effect was described as modest and the dimeric-fraction difference alone was not statistically significant (p = 0.15) (PMID:25606676). Together these reproducible, biologically relevant results support a damaging gain-of-function effect, but the evidence is not upgraded to strong because the studies were research assays rather than validated clinical diagnostic laboratory assays and one biophysical endpoint was modest.

Rule

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

Evidence considered

• Tomlinson et al. 2007 (PMID:17384684), Results: in FGFR3 S249C-positive 97-7 bladder cancer cells, the mutant receptor formed stable dimers and had higher phosphorylation than wild-type FGFR3b; shRNA knockdown reduced cell proliferation, colony formation, and anchorage-independent growth, while S249C re-expression rescued the phenotype.
• Del Piccolo et al. 2015 (PMID:25606676), Materials and Methods/Results: quantitative FRET in transfected CHO-cell plasma-membrane vesicles showed ligand-independent S249C dimerization of 500 ± 100 versus 290 ± 60 for wild type, with dimerization stabilization of −0.3 ± 0.1 kcal/mol and altered dimer structure; the effect was described as modest and dimeric fraction alone was not significant (p = 0.15).

Still missing

• No validated, reproducible clinical diagnostic laboratory assay was identified in the reviewed evidence.
• The quantitative dimerization study reported a modest effect, with dimeric fraction not significant by its reported test (p = 0.15).

Sources

Reasoning

The governing rule is: "The prevalence of the variant in affected individuals is significantly increased compared to the prevalence in controls." The ACMG/AMP note requires: "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." PMID:25614871 reports the exact variant in 11 of 173 affected TD I cases. gnomAD v2.1 reports 0 of 238624 alleles. Treating the reported case frequency and the available population frequency as a case-versus-population-control comparison, I computed a Haldane-Anscombe continuity-corrected OR because the control count is zero: OR = ((11.5 x 238624.5)/(161.5 x 0.5)) = 33983.67; approximate 95% CI = 1994.12-579146.64. The required comparison is OR > 5.0 and CI excluding 1.0: 33983.67 > 5.0? yes; 1994.12-579146.64 includes 1.0? no. PS4 is met. The literature_pass PS4-classified PMID:8845844 was independently checked against its PubMed abstract and confirms S249C in 2/26 TD1 cases, but was not needed for the calculation.

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:25614871, Results/Table 1: exact p.Ser249Cys/c.746C>G was present in 11 of 173 TD I cases (6.4%).
• gnomAD v2.1: 0 of 238624 alleles carried the variant.
• Computed the case-control comparison with a Haldane-Anscombe 0.5 continuity correction: OR = 33983.67; approximate 95% CI = 1994.12-579146.64.
• Numeric comparison: 33983.67 > 5.0? yes; 1994.12-579146.64 includes 1.0? no.
• The literature_pass ai-classified PS4 paper PMID:8845844 was checked against its PubMed abstract; it reports S249C in 2/26 TD1 cases and no controls.

Still missing

• The control count is zero, so the finite OR and CI depend on the explicitly stated Haldane-Anscombe continuity correction.
• Case ancestry is not reported, so the gnomAD comparison is population-based rather than demonstrably ancestry-matched.

Sources

Reasoning

PM2 is met at Supporting strength. The variant was absent from gnomAD v4.1 and absent from gnomAD v2.1 all-comers exomes (0/238,624 alleles; 0 homozygotes), including the reported African/African American subpopulation (0/14,544 alleles). The separately reported gnomAD v2.1 non-cancer exome subset also showed AF 0 (0/227,558 alleles; 0 homozygotes). These observations satisfy the supplied generic PM2 Supporting threshold of <=0.0001 and support rarity in reference populations.

Rule

Absent from controls (or at extremely low frequency if recessive) in Exome Sequencing Project, 1000 Genomes or Exome Aggregation Consortium.

Evidence considered

• gnomAD v2.1 all-comers exomes: 0 alternate alleles among 238,624 total alleles, AF 0, and 0 homozygotes; the highest reported subpopulation AF was 0 in African/African American individuals (0/14,544 alleles).
• gnomAD v4.1: variant absent.
• The separately reported gnomAD v2.1 non-cancer exome subset also showed AF 0 (0/227,558 alleles; 0 homozygotes); no governing VCEP or source-specific non-cancer requirement applies here, so the all-comers totals are the default evidence.
• Generic PM2 calibration supplied for this case: <=0.0001 at Supporting strength, identified as the ClinGen SVI recommendation to downgrade PM2 to Supporting (PMID:25741868).

Sources

Reasoning

PMID:30692697 identifies the exact FGFR3 c.746C>G (p.S249C) variant in subject P18 and describes the FGFR3 variants in subjects P9-P18 as de novo, with skeletal-dysplasia findings consistent with thanatophoric dysplasia I. The report confirms the fetal variant by a secondary assay and postnatal testing but provides no confirmation of maternity and paternity, supporting PM6 rather than PS2.

Rule

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

Evidence considered

• PMID:30692697 states that de novo FGFR3 pathogenic variants were identified in subjects P9-P18 and were consistent with fetal skeletal-dysplasia phenotypes.
• Table 1, subject P18 reports FGFR3 c.746C>G (p.S249C), fetal variant fraction 11.6%, thanatophoric dysplasia I, and confirmation by secondary assay and postnatal testing followed by neonatal demise; no maternity or paternity confirmation is reported.

Still missing

• Parental testing results confirming or excluding the variant in both parents are not reported.

Sources

Reasoning

The variant is a missense change, NM_000142.5:c.746C>G (p.Ser249Cys), so the governing generic rule selects the REVEL path only; SpliceAI is not used for PP3/BP4 in this missense situation. REVEL is 0.902, meeting the supplied moderate PP3 threshold of >=0.773 but not the strong threshold of >=0.932. BayesDel is not used because no generic ACMG-strength calibration is established.

Rule

Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.). ... PP3 (moderate): >= 0.773 — ClinGen SVI REVEL calibration (Pejaver et al. 2022, Am J Hum Genet) (PMID:36413997)

Evidence considered

• NM_000142.5:c.746C>G is a missense variant producing NP_000133.1:p.(Ser249Cys); therefore REVEL is the applicable computational path and SpliceAI is excluded.
• Local REVEL lookup reports score 0.902 for this variant.
• ClinGen SVI REVEL calibration (Pejaver et al. 2022, Am J Hum Genet; PMID:36413997) supplies the generic PP3 moderate threshold >=0.773 and strong threshold >=0.932; 0.902 meets moderate PP3 only.
• BayesDel score 0.356866 was not used because no generic ACMG-strength calibration is established.

Sources

Reasoning

The variant is FGFR3 p.Ser249Cys caused by c.746C>G, and the reviewed clinical and functional sources establish this exact variant. However, no previously established pathogenic variant producing the same Ser249Cys amino-acid change through a different nucleotide change was identified. Reports of the exact c.746C>G allele do not satisfy PS1's distinct-nucleotide requirement.

Rule

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

Evidence considered

• FGFR3 p.Ser249Cys caused by c.746C>G was reported in 11 of 173 thanatophoric dysplasia type I cases; the source describes the exact c.746C>G allele, not a different nucleotide change producing Ser249Cys.
• The functional studies directly used the Ser249Cys construct generated by the 746C-to-G transition and therefore document the same nucleotide-level allele rather than an alternate nucleotide change.
• ClinVar variation 16339 identifies the exact variant as NM_000142.5:c.746C>G (p.Ser249Cys); no alternate nucleotide representation for the same amino-acid change is provided in the reviewed record.

Sources

Reasoning

The exact FGFR3 c.746C>G (p.S249C) variant was reported as de novo in fetal subject P18 with thanatophoric dysplasia I and phenotype-consistent skeletal abnormalities. However, the report documents secondary and postnatal confirmation of the fetal variant, not confirmation of both maternity and paternity; therefore the defining PS2 requirement is not satisfied.

Rule

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

Evidence considered

• PMID:30692697 reports FGFR3 c.746C>G (p.S249C) in subject P18 with skeletal dysplasia and thanatophoric dysplasia I; the study describes the FGFR3 variants in subjects P9-P18 as de novo and states that P18 was confirmed by secondary assay and postnatal testing, with neonatal demise.
• The P18 table entry provides fetal variant fraction 11.6%, phenotype skeletal dysplasia, and the result FGFR3 c.746C>G (p.S249C), but does not report maternity and paternity confirmation.

Still missing

• Documented confirmation of both maternity and paternity for subject P18 is unavailable.

Sources

Reasoning

The governing rule is: "Reputable source recently reports variant as pathogenic, but the evidence is not available to the laboratory to perform an independent evaluation." ClinVar contains Pathogenic and Likely pathogenic submissions for the exact variant, but the audit records zero expert-panel submissions. The supporting literature is available for independent evaluation, so ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions do not trigger 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: aggregate classification Pathogenic, but expert_panel_submissions = 0; submissions are non-expert laboratory, literature-only, curation, or other assertions.
• Exact-variant clinical and functional literature is available for independent evaluation.

Sources

Reasoning

BA1 is not met. The variant was absent from gnomAD v4.1 and had an allele frequency of 0 (0/238,624 alleles; 0 homozygotes) in gnomAD v2.1 all-comers exomes, far below the generic BA1 stand-alone threshold of >=0.05. The separately reported gnomAD v2.1 non-cancer exome subset also had AF 0 (0/227,558 alleles). The variant is a single-nucleotide substitution, 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 exomes: 0 alternate alleles among 238,624 total alleles, AF 0, and 0 homozygotes; highest reported subpopulation AF was also 0 (African/African American, 0/14,544 alleles).
• gnomAD v4.1: variant absent.
• The separately reported gnomAD v2.1 non-cancer exome subset showed AF 0 (0/227,558 alleles; 0 homozygotes); no governing VCEP or source-specific non-cancer requirement applies here, so all-comers totals remain the default.
• Generic BA1 calibration supplied for this case: >=0.05, ACMG/AMP 2015 stand-alone benign allele-frequency threshold (PMID:25741868).

Sources

Reasoning

BS1 is not met. The variant was absent from gnomAD v4.1 and had AF 0 in gnomAD v2.1 all-comers exomes, so there is no elevated population frequency to compare with the disorder's expected frequency. The generic default BS1 threshold of >=0.01 is not reached; no maximum-credible-frequency calculation is needed for an observed AF of zero.

Rule

Allele frequency is greater than expected for the disorder.

Evidence considered

• gnomAD v2.1 all-comers exomes: 0 alternate alleles among 238,624 total alleles, AF 0, and 0 homozygotes; the highest reported subpopulation AF was 0.
• gnomAD v4.1: variant absent.
• Generic BS1 default supplied for this case: >=0.01; a disorder-specific maximum-credible-frequency derivation is unnecessary because the observed AF is zero.

Sources

Reasoning

BS3 is not met. The reviewed variant-specific functional studies do not show no damaging effect on protein function or splicing. Instead, the human-cell study found stable mutant-receptor dimerization, increased phosphorylation, mutant-dependent growth phenotypes, and rescue after mutant re-expression, while the quantitative assay found increased ligand-independent dimerization and altered dimer structure. These findings are directionally inconsistent with a benign functional result and provide no basis for applying BS3.

Rule

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

Evidence considered

• Tomlinson et al. 2007 (PMID:17384684), Results: S249C FGFR3 formed stable dimers, showed higher phosphorylation than wild type, and supported cellular growth phenotypes that were reduced by knockdown and rescued by mutant re-expression.
• Del Piccolo et al. 2015 (PMID:25606676), Results: S249C increased ligand-independent dimerization relative to wild type and altered the measured dimer structure; the study did not establish normal protein function or normal splicing.

Still missing

• No well-established benign functional assay demonstrating normal protein function or splicing was identified.

Sources

Reasoning

BP1 applies to a missense variant in a gene for which disease is primarily caused by truncating variants. The reviewed FGFR3 evidence instead documents recurrent disease-associated missense variants, including p.Ser249Cys, and functional studies demonstrate activating effects of this missense change. This evidence is inconsistent with FGFR3 being a primarily truncating-variant disease gene for the relevant disorder context, so BP1 is not met.

Rule

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

Evidence considered

• A cohort of 173 thanatophoric dysplasia type I cases reports p.Ser249Cys as the third most common mutation, with 11 cases carrying this missense change.
• FGFR3 S249C functional studies demonstrate stable receptor dimers, higher phosphorylation, and mutant-dependent cellular phenotypes, supporting a disease-relevant activating missense mechanism.
• The reviewed evidence does not show that the relevant FGFR3 disease is primarily caused by truncating variants.

Sources

Reasoning

The variant is a missense change, so the governing generic rule selects the REVEL path only; the SpliceAI result cannot substitute for a gray-zone REVEL result. REVEL is 0.902, which is above the supplied BP4 supporting cutoff of <=0.29 and therefore does not support BP4. 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). ... BP4 (supporting): <= 0.29 — ClinGen SVI REVEL calibration (Pejaver et al. 2022, Am J Hum Genet) (PMID:36413997)

Evidence considered

• NM_000142.5:c.746C>G is a missense variant producing NP_000133.1:p.(Ser249Cys); therefore REVEL is the applicable computational path and SpliceAI is excluded.
• Local REVEL lookup reports score 0.902 for this variant.
• ClinGen SVI REVEL calibration (Pejaver et al. 2022, Am J Hum Genet; PMID:36413997) supplies the generic BP4 supporting threshold <=0.29, moderate threshold <=0.183, and strong threshold <=0.016; 0.902 meets none of them.
• BayesDel score 0.356866 was not used because no generic ACMG-strength calibration is established.

Sources

Reasoning

The governing rule is: "Reputable source recently reports variant as benign, but the evidence is not available to the laboratory to perform an independent evaluation." The exact-variant ClinVar record contains zero expert-panel submissions and no expert-panel Benign or Likely benign classification. No qualifying reputable inaccessible benign report is present, so BP6 is 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 exact-variant audit: expert_panel_submissions = 0 and no expert-panel Benign or Likely benign classification is recorded.
• Ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions were not used to trigger BP6.

Sources

Reasoning

FGFR3 p.Ser249Cys is reported in a disease-associated extracellular region and the case evidence reports that the variant lies in a statistically significant hotspot. Generic PM1 requires a mutational hotspot or a critical, well-established functional domain without benign variation. Because no FGFR3 VCEP governs this case, the somatic hotspot signal cannot be used as a VCEP-defined PM1 route, and the available sources do not establish a qualifying critical domain together with absence of benign variation at the required standard. The criterion remains not assessed.

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 reviewed FGFR3 cohort paper reports p.Ser249Cys among thanatophoric dysplasia type I mutations and describes TD I missense mutations in the extracellular domain, but does not establish a complete critical-domain boundary or absence of benign variation.
• The evidence record states that the variant lies in a statistically significant hotspot; generic PM1 assessment here does not treat that somatic hotspot signal as sufficient without a governing VCEP-defined hotspot route.
• The case's vcep_materials.json has no FGFR3 VCEP materials and an empty domain_tables object.

Still missing

• A source establishing a qualifying FGFR3 critical functional domain at residue 249 and absence of benign variation is needed for generic PM1.

Sources

Reasoning

The target is FGFR3 p.Ser249Cys. The reviewed evidence establishes this exact variant, but it does not provide a different missense substitution at residue 249 with a validated pathogenic determination. The dedicated PM5 candidate artifact contains zero same-residue candidates and indicates that classic PM5 semantics could not be confirmed safely; therefore the criterion cannot be met from the available evidence.

Rule

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

Evidence considered

• pm5_candidates.json reports same_residue_candidates = 0 and no collected comparator variants.
• The reviewed clinical paper reports the exact p.Ser249Cys substitution, not a different missense change at residue 249.
• ClinVar variation 16339 concerns the exact c.746C>G (p.Ser249Cys) allele and does not supply a different pathogenic missense change at residue 249.

Still missing

• A validated pathogenic FGFR3 missense comparator at residue 249, different from p.Ser249Cys, is needed.

Sources

Reasoning

The available evidence describes a single affected fetus, subject P18, and provides no affected relatives, informative meioses, or family-based transmission data for the exact variant. The PP1 requirement cannot therefore be assessed.

Rule

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

Still missing

• Family pedigree and genotype-phenotype results for affected relatives are unavailable.
• No segregation count or informative meiosis data are available.

Reasoning

FGFR3 disease-associated missense variants and activating missense biology are documented in the reviewed literature, but the case evidence does not provide a gene-level assessment demonstrating both a low rate of benign missense variation and that missense variants are a common disease mechanism under the generic PP2 rule. The absence of this paired gene-level evidence prevents assessment.

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

• FGFR3 p.Ser249Cys is reported repeatedly in a thanatophoric dysplasia type I cohort, supporting that missense variants can cause FGFR3 disease.
• Functional studies show Ser249Cys receptor dimerization, phosphorylation, and cellular effects consistent with activating missense biology.
• No gene-level benign-missense-rate metric or curated benign missense spectrum sufficient for the PP2 requirement is provided in the reviewed evidence.

Still missing

• A gene-level analysis establishing low benign missense variation and common missense disease mechanism for FGFR3 is needed.

Sources

Reasoning

The governing rule is: "Patient’s phenotype or family history is highly specific for a disease with a single genetic etiology." The case record does not provide the evaluated patient's individual phenotype or family history. Published TD1 reports cannot substitute for the present patient's phenotype, so PP4 cannot be assessed.

Rule

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

Evidence considered

• No patient-level phenotype or family-history evidence is available in the supplied case materials.
• PMID:30692697 reports the exact variant in a fetal TD1 case, but that published case is not the evaluated patient's phenotype.

Still missing

• Patient-level phenotype and family-history details are required.

Sources

Reasoning

BS2 cannot be assessed from the available population data. gnomAD reports 0 homozygotes, but BS2 requires observation of the variant in a healthy adult individual in the relevant zygosity state; aggregate allele counts do not establish an adult's health status, age, or heterozygous genotype. No curated healthy-adult observation was provided.

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 all-comers exomes: 0 alternate alleles among 238,624 total alleles and 0 homozygotes; this aggregate result does not identify healthy adult individuals or establish their clinical status.
• gnomAD v4.1: variant absent; no individual-level healthy-adult observation was available from this source.

Still missing

• Individual-level evidence showing whether the variant was observed in a healthy adult, including genotype, age, phenotype, and relevant inheritance state.

Sources

Reasoning

No affected family members carrying or lacking the variant are described, so there is no family-based non-segregation observation to evaluate. The absence of family segregation data is insufficient to apply BS4.

Rule

Lack of segregation in affected members of a family.

Still missing

• No family genotype data are available to demonstrate lack of segregation in affected members.
• Potential phenocopy and multiple-variant explanations cannot be evaluated without a family history and pedigree.

Reasoning

BP2 is applicable to a fully penetrant dominant disorder when the evaluated variant is observed in trans with a pathogenic variant, or to any inheritance pattern when observed in cis with a pathogenic variant. The available evidence reports FGFR3 c.746C>G/p.S249C in a heterozygous bladder cancer cell line, but does not identify a second pathogenic variant or establish cis/trans phase in an affected proband. Therefore BP2 cannot be assessed.

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

• The study reports a heterozygous S249C FGFR3 mutation in the 97-7 bladder cancer cell line, but does not report a pathogenic partner variant or cis/trans phase in an affected proband.

Still missing

• Affected-proband testing identifying a second pathogenic variant and parental or molecular phasing is unavailable.

Sources

Reasoning

The governing rule is: "Variant found in a case with an alternate molecular basis for disease." No patient-level evidence establishes that the evaluated case has an alternate molecular diagnosis explaining its disease. PMID:39243045 describes a different dual-diagnosis patient but does not identify the FGFR3 variant as c.746C>G/p.Ser249Cys. BP5 therefore cannot be assessed for this exact case.

Rule

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

Evidence considered

• PMID:39243045 describes dual achondroplasia and mandibulofacial dysostosis with microcephaly, but does not identify the FGFR3 variant as c.746C>G/p.Ser249Cys.
• No exact-variant alternate-molecular-basis case is documented in the available case evidence.

Still missing

• Patient-level molecular testing and clinical data demonstrating an alternate molecular basis are required.