RET c.2944C>T, p.(Arg982Cys)

Reasoning

Generic ACMG/AMP is applicable because no RET-specific VCEP framework was identified. BS1 is met at strong strength: the variant has gnomAD v2.1 total AF 0.0185349, gnomAD v2.1 non-cancer exome AF 0.0194274, gnomAD v4.1 total AF 0.0163633, and gnomAD v3.1 non-cancer genome AF 0.0164404. These frequencies exceed the supplied generic BS1 threshold of 0.01, and the highest subgroup frequencies are 0.0431147 in gnomAD v2.1 and 0.0423366 in gnomAD v4.1. The frequency is therefore too high for a highly penetrant rare disease allele, although disease prevalence and reduced penetrance can limit interpretation of BS1 for susceptibility phenotypes.

Rule

{'source': 'generic_svi_default', 'threshold': '>=0.01', 'threshold_publication': 'PMID:28518168'}

Evidence considered

• gnomAD v2.1 reports total AF 0.0185349 and non-cancer exome AF 0.0194274; its highest subgroup AF is 0.0431147.
• gnomAD v4.1 reports total AF 0.0163633; gnomAD v3.1 non-cancer genomes report AF 0.0164404; the gnomAD v4.1 highest subgroup AF is 0.0423366.
• The generic SVI BS1 threshold is allele frequency >=0.01; per-gene derivation is preferred (Whiffin et al., PMID:28518168).
• The observed frequencies are compatible with a common or reduced-penetrance RET susceptibility polymorphism but are too high for a rare, highly penetrant disease allele.

Sources

Reasoning

BS2 is met at supporting strength because the variant is observed in numerous homozygotes in population datasets: 82 homozygotes in gnomAD v2.1, including 78 in exomes, and 372 homozygotes in gnomAD v4.1, including 342 in exomes; the gnomAD v3.1 non-cancer genome subset contains 30 homozygotes. This is incompatible with a fully penetrant dominant RET disorder, but gnomAD is not a phenotype-confirmed healthy cohort and RET-associated susceptibility phenotypes may have incomplete penetrance, so the strength and application should receive human review.

Rule

{'source': 'generic_acmg_amp', 'application': 'observed in homozygous state for a dominant disorder', 'strength': 'supporting'}

Evidence considered

• gnomAD v2.1 reports 82 total homozygotes, including 78 exome homozygotes; its non-cancer exome subset also reports 78 homozygotes.
• gnomAD v4.1 reports 372 total homozygotes, including 342 exome homozygotes and 30 genome homozygotes.
• gnomAD v3.1 non-cancer genomes report 30 homozygotes.
• The homozygote observation supports benignity for a fully penetrant dominant disorder, but incomplete penetrance and the absence of individual phenotype data limit certainty.

Still missing

• Population datasets do not provide phenotype-confirmed unaffected status or penetrance estimates for every homozygous carrier.

Sources

Reasoning

Supporting benign functional evidence is present. PMID:22729463 tested RET R982C alone in a GDNF-dependent MAPK phosphorylation assay using transfected HEK293 cells and found no alteration of activity, a biologically relevant negative result with a defined assay context. However, the same study found near-complete loss for the paired G691S/R982C construct, and the available evidence is not sufficiently independent or comprehensive for a stronger BS3 level.

Rule

Generic ACMG/AMP BS3 supports benign interpretation when a well-established functional study shows no damaging effect; the single relevant negative assay supports supporting strength because the compound-variant result and limited independent replication constrain calibration.

Evidence considered

• PMID:22729463 reports that RET R982C alone did not alter GDNF-dependent MAPK activity in transfected HEK293 cells; the assay used site-directed mutagenesis and GDNF-stimulated MAPK phosphorylation.
• PMID:21655256 states that prior in vitro expression experiments found normal biological and biochemical behavior for p.R982C, although those experiments were not performed in the reported study.

Still missing

• Additional independent assays with validated positive and negative controls would be needed to justify stronger benign functional evidence.

Sources

Reasoning

No applicable RET-specific VCEP specification was identified, so generic ACMG/AMP is used. The queried RET c.2944C>T (p.R982C) variant was documented in a four-generation family on the paternal haplotype, while the established pathogenic p.C634Y variant was of maternal origin; the daughter therefore inherited p.R982C in trans with p.C634Y. This is direct phase evidence that p.R982C occurs in trans with a pathogenic RET variant in a dominant RET disease family and supports BP2 at supporting strength. The same report describes p.R982C as a polymorphism and notes prior normal biological and biochemical behavior, but those observations are not used to add criteria here.

Rule

Generic ACMG/AMP BP2: observed in trans with a pathogenic variant for a fully penetrant dominant disorder; apply at supporting strength.

Evidence considered

• In the reported Chinese MEN2A/FMTC family, haplotype analysis confirmed p.C634Y was maternal and p.V292M/p.R67H/p.R982C were paternal and on a common allele.
• The daughter inherited the paternal p.R982C-containing haplotype and the maternal p.C634Y allele, directly establishing p.R982C in trans with pathogenic p.C634Y.
• This phase observation satisfies generic BP2 for a dominant RET disease context at supporting strength.

Sources

Reasoning

The variant is a missense substitution, NP_066124.1:p.(Arg982Cys), so BP4 is evaluated using REVEL only. REVEL is 0.282, which is at or below the generic ClinGen SVI BP4 supporting threshold of <=0.29 from the REVEL calibration of Pejaver et al. 2022 (PMID:36413997); therefore BP4 is met at supporting strength. SpliceAI is not used because this variant is missense, and the same REVEL prediction is not double-counted for PP3.

Rule

For a missense variant, apply BP4 at supporting strength when REVEL is <=0.29 under the published ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997).

Evidence considered

• The variant is classified as missense, NP_066124.1:p.(Arg982Cys), so the missense REVEL path governs PP3/BP4.
• REVEL score is 0.282, meeting the BP4 supporting threshold of <=0.29 specified by the ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997).
• SpliceAI was not used for BP4 because a missense variant must be evaluated through REVEL rather than the splice-impact path.

Sources

Reasoning

Generic BP5 is supporting evidence when a variant is found in an affected individual with an alternate molecular basis for the disease. PMID:21655256 reports R982C in cis with p.V292M and p.R67H in a family whose daughter also inherited the established MEN2A-associated p.C634Y variant; PMID:22729463 reports R982C with RET-G691S in a CAKUT patient and found no functional effect for R982C alone, while the combined construct altered signaling. These observations support an alternate molecular basis, but the complex multilocus backgrounds and possible modifying effects warrant human review.

Rule

Generic ACMG/AMP BP5: variant found in a patient with an alternate molecular basis for disease; applied at supporting strength because the alternate-basis observations are present but occur in complex RET genotypes.

Evidence considered

• PMID:21655256 reports a daughter with the pathogenic p.C634Y variant plus paternal p.V292M/p.R67H/R982C, supporting an alternate molecular basis for the MEN2A phenotype; the study describes R982C as a possible modifier rather than an independently established cause.
• PMID:22729463 reports a CAKUT patient with RET-G691S and R982C; R982C alone did not alter GDNF-dependent MAPK activity, whereas the double construct showed near-complete loss of signaling, supporting the presence of another molecular basis while leaving possible interaction effects.

Sources

Reasoning

PS1 requires a different nucleotide change producing the same amino-acid substitution that is established pathogenic. The available literature reports the queried R982C variant itself, not a distinct nucleotide change producing R982C, and describes it as debated, a polymorphism, or of uncertain significance; no established pathogenic same-amino-acid comparator is documented.

Evidence considered

• PMID:14633923 reports RET R982C itself in three Hirschsprung disease patients, notes prior observation in control chromosomes, and does not establish pathogenicity.
• PMID:15741265 reports R982C as a second mutation but states that its significance and functional consequences were uncertain.
• PMID:21655256 describes p.R982C as a rare polymorphism with normal biological and biochemical behavior reported from prior in-vitro experiments.
• PMID:22729463 evaluates the same R982C variant, finding no effect alone in a GDNF-dependent MAPK assay; the observed loss of activity required the combined G691S/R982C construct.

Sources

Reasoning

PS2 is not met. No report documents RET c.2944C>T (p.Arg982Cys) as a confirmed de novo variant with maternity and paternity excluded. Instead, the variant was inherited from the mother in one patient and was inherited from the father in another reported family; these observations do not provide de novo evidence for this variant.

Rule

Generic ACMG/AMP PS2: confirmed de novo occurrence in a patient with the relevant phenotype, with maternity and paternity confirmed or otherwise appropriately excluded; no RET-specific specification was available.

Evidence considered

• PMID:14633923 reports that patient 55 inherited R982C from her mother; the paper does not report R982C as de novo.
• PMID:21655256 reports p.R982C on the paternal allele in the proband's husband and paternal relatives, with inheritance by their daughter.
• PMID:22729463 reports that the R982C-containing RET haplotype was inherited from the father in the affected family.
• Under generic ACMG/AMP PS2, confirmed de novo occurrence with parental relationships established is required; no such observation is documented for this variant.

Still missing

• No case-specific trio testing or validated maternity/paternity documentation is available showing this variant arose de novo.

Sources

Reasoning

No assay demonstrates that RET p.Arg982Cys independently causes loss of function. In the relevant HEK293 GDNF-stimulated MAPK phosphorylation assay, the R982C construct alone did not alter activity; near-complete loss was observed only for the combined G691S/R982C construct, so the result cannot be attributed specifically to R982C. Other reviewed reports describe R982C as having unknown or normal functional consequences, without a validated variant-specific damaging assay.

Rule

Generic ACMG/AMP PS3 requires well-established functional studies showing a damaging effect; a damaging result attributable only to a compound construct does not establish PS3 for R982C alone.

Evidence considered

• PMID:22729463 reports site-directed RET mutagenesis and GDNF-stimulated MAPK phosphorylation in transfected HEK293 cells: R982C alone did not alter activity, whereas the combined G691S/R982C construct showed near-complete loss of GDNF-dependent MAPK phosphorylation.
• PMID:21655256 describes prior in vitro expression work as showing normal biological and biochemical behavior for p.R982C, but the study itself did not independently perform that functional experiment.
• PMID:15741265 explicitly states that the functional consequences of R982C were unknown.

Still missing

• An independent, well-controlled assay testing R982C alone with appropriate RET signaling controls is needed to establish a damaging effect.

Sources

Reasoning

Generic PS4 requires a convincing disease-specific case-control enrichment. One Chinese Hirschsprung disease study observed R982C in 3 of 84 affected individuals and 0 of 96 matched controls, but the authors explicitly note that R982C had also been observed in control chromosomes and that two additional de novo changes in one patient could contribute to disease. The available evidence therefore does not establish sufficiently specific or independent enrichment for this variant, particularly in view of its substantial population frequency in gnomAD v4.1 (AF 0.01636).

Rule

Generic ACMG/AMP PS4: statistically significant increased prevalence of the variant in affected individuals compared with controls; no disease-specific VCEP threshold was available.

Evidence considered

• PMID:14633923 reports R982C in three sporadic Hirschsprung disease patients and none of 96 ethnically matched controls, but also describes prior detection in control chromosomes and possible contribution from two additional de novo silent variants.
• gnomAD v4.1 reports R982C at allele frequency 0.0163633, including 372 homozygotes, which is inconsistent with a highly penetrant rare disease allele and weakens interpretation of the isolated case-control observation.

Sources

Reasoning

No governing VCEP or RET-specific domain table is available. R982 lies within the RET kinase-domain region discussed in the literature, but PM1 also requires a critical functional domain or established hotspot with little benign variation. This residue is not a statistically significant hotspot, and the variant has substantial population frequency (gnomAD v4.1 AF 0.0163633; grpmax FAF 0.04122125), which is inconsistent with applying PM1 to this variant.

Evidence considered

• PMID:22837065 concerns RET kinase-domain sequence variants and supports the general functional importance of the kinase domain, but does not establish R982 as a critical PM1 hotspot or domain subregion.
• The case evidence reports no statistically significant hotspot at RET R982.
• gnomAD v4.1 reports the queried variant at AF 0.0163633 with maximum population frequency 0.04122125, providing substantial benign-population variation at the queried site.

Sources

Reasoning

Generic ACMG/AMP is applicable because no RET-specific VCEP framework was identified. PM2 is not met: the variant is common in every relevant population dataset, with gnomAD v2.1 non-cancer exome AF 0.0194274, gnomAD v3.1 non-cancer genome AF 0.0164404, gnomAD v2.1 total AF 0.0185349, and gnomAD v4.1 total AF 0.0163633. These values are far above the supplied PM2 supporting threshold of <=0.0001, and the variant has 82 gnomAD v2.1 and 372 gnomAD v4.1 homozygotes.

Rule

{'source': 'generic_clingen_svi', 'threshold': '<=0.0001', 'threshold_publication': 'PMID:25741868'}

Evidence considered

• gnomAD v2.1 non-cancer exomes report AF 0.0194274 and total AF 0.0185349.
• gnomAD v3.1 non-cancer genomes report AF 0.0164404.
• gnomAD v4.1 total AF is 0.0163633 with 372 homozygotes.
• The supplied ClinGen SVI PM2 recommendation uses AF <=0.0001 for supporting evidence (PMID:25741868).

Sources

Reasoning

PM5 requires an established pathogenic missense change at the same residue. The reviewed evidence identifies R982C but no alternate missense substitution at residue 982 with an established pathogenic classification; the comparator search contains zero same-residue candidates. Therefore PM5 is not met.

Evidence considered

• The same-residue comparator collection reports zero candidates and cannot support a classic PM5 application.
• The reviewed papers report the queried R982C substitution itself, not a distinct pathogenic alternate substitution at residue 982.

Sources

Reasoning

PM6 is not met. The available variant-specific reports do not describe an assumed de novo occurrence without parental testing. The documented observations instead describe maternal or paternal inheritance, including a paternal haplotype transmitted through a family, so they cannot support PM6.

Rule

Generic ACMG/AMP PM6: assumed de novo occurrence in a patient with the relevant phenotype when parental testing is unavailable or incomplete; no RET-specific specification was available.

Evidence considered

• PMID:14633923 explicitly states that patient 55 inherited R982C from her mother; the de novo changes in that patient were different silent RET variants.
• PMID:21655256 documents p.R982C in a paternal allele and transmission of that allele to the daughter.
• PMID:22729463 describes paternal inheritance of the R982C-containing RET haplotype.
• Under generic ACMG/AMP PM6, an assumed de novo event may support the criterion only when parental testing is unavailable or incomplete; no such R982C event is reported.

Still missing

• No variant-specific clinical report documents a plausible de novo event when parental testing was incomplete or unavailable.

Sources

Reasoning

PP1 is not met. Although variant transmission is described, the reports do not establish cosegregation of R982C with the relevant disease phenotype across informative affected relatives. One report describes maternal inheritance in a patient while also identifying other de novo changes, and another describes a paternal multivariant haplotype in a family without sufficient phenotype-by-carrier segregation data; the evidence therefore cannot be counted as disease-specific segregation.

Rule

Generic ACMG/AMP PP1: cosegregation with disease in multiple affected family members, with strength dependent on the number and informativeness of meioses; no RET-specific specification was available.

Evidence considered

• PMID:14633923 reports R982C in three Hirschsprung disease patients, but patient 55 inherited it from her mother and also carried two de novo silent RET changes; the authors do not establish R982C-specific cosegregation or causality.
• PMID:21655256 reports p.R982C in cis with p.V292M and p.R67H on a paternal allele and transmission of a multivariant haplotype, so segregation cannot be attributed to R982C alone.
• PMID:22729463 reports R982C together with RET-G691S in a patient whose variants were inherited from the father; the paper does not provide sufficient R982C-only phenotype segregation across affected relatives.
• Generic ACMG/AMP PP1 requires cosegregation with disease in affected family members; the available observations do not provide enough informative R982C-specific meioses.

Still missing

• No pedigree-level table identifies affected and unaffected relatives, carrier status, phenotype status, and informative meioses for R982C alone.
• Reported families contain additional RET variants or other potential contributors, preventing attribution of phenotype to R982C alone.

Sources

Reasoning

PP2 requires a gene in which missense variation is a common disease mechanism and benign missense variation is uncommon. RET has documented benign or polymorphic missense variation, including R982C, which is present in gnomAD v4.1 at AF 0.0163633 and grpmax FAF 0.04122125; the available evidence therefore does not support PP2.

Evidence considered

• gnomAD v4.1 reports R982C at AF 0.0163633 and grpmax FAF 0.04122125, demonstrating that this missense variant is not rare in population data.
• PMID:14633923 describes R982C as debated and previously observed in control chromosomes.
• PMID:21655256 calls R982C a rare polymorphism and cites normal biological and biochemical behavior in prior in-vitro expression experiments.

Sources

Reasoning

The variant is a missense substitution, NP_066124.1:p.(Arg982Cys), so PP3 is evaluated using REVEL only. REVEL is 0.282, below the generic ClinGen SVI PP3 supporting threshold of >=0.644 from the REVEL calibration of Pejaver et al. 2022 (PMID:36413997); therefore PP3 is not met. The splice predictor is not used because this variant is missense, and BayesDel has no generic ACMG strength calibration.

Rule

For a missense variant, apply PP3 only when REVEL meets a published calibrated threshold; generic supporting threshold is >=0.644 (ClinGen SVI REVEL calibration, Pejaver et al. 2022, PMID:36413997).

Evidence considered

• The variant is classified as missense, NP_066124.1:p.(Arg982Cys), so the missense REVEL path governs PP3/BP4.
• REVEL score is 0.282, below the PP3 supporting threshold of >=0.644 specified by the ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997).
• SpliceAI and BayesDel were not used for PP3: SpliceAI is outside the selected missense path, and no generic ACMG strength calibration is established for BayesDel.

Sources

Reasoning

PP4 requires a highly specific phenotype or phenotype pattern for a disorder with a single-gene etiology. The extracted reports describe heterogeneous Hirschsprung disease, CAKUT, and MEN-related contexts, often with additional RET variants, and do not establish a distinctive phenotype uniquely attributable to R982C. Generic PP4 is therefore not met.

Rule

Generic ACMG/AMP PP4: patient phenotype or family history is highly specific for a disease with a single genetic etiology; no disease-specific VCEP rule was available.

Evidence considered

• PMID:14633923 describes R982C in sporadic Hirschsprung disease patients but notes possible contribution from additional silent variants.
• PMID:21655256 describes R982C in a multivariant RET haplotype with p.C634Y in a MEN2A/familial medullary thyroid carcinoma family and discusses a possible modifying effect rather than a specific R982C phenotype.
• PMID:22729463 describes R982C in a CAKUT patient carrying another RET variant, G691S, with complex renal and extra-renal abnormalities.

Sources

Reasoning

PP5 is restricted here to an exact-variant ClinVar expert-panel classification of Pathogenic or Likely pathogenic. ClinVar variation 13938 has no expert-panel submissions; its displayed benign/likely benign classifications are laboratory submissions and the aggregate record is not an expert-panel assertion. PP5 is therefore not met.

Rule

PP5 applies only to an exact-variant ClinVar expert-panel classification of Pathogenic or Likely pathogenic; ordinary laboratory submissions, aggregate labels, and non-expert assertions are excluded.

Evidence considered

• ClinVar exact variant NM_020975.6:c.2944C>T (p.Arg982Cys), variation 13938, reports zero expert-panel submissions and only laboratory submissions, with the aggregate record carrying one-star/conflicting review information.

Sources

Reasoning

Generic ACMG/AMP is applicable because no RET-specific VCEP framework was identified. BA1 is not met: the highest reported population frequency is 0.0431147 in the gnomAD v2.1 South Asian subgroup (grpmax FAF 0.0411806), and gnomAD v4.1 reports a highest subgroup frequency of 0.0423366 (grpmax FAF 0.0412213); all are below the generic stand-alone threshold of 0.05 from ACMG/AMP 2015 (PMID:25741868). The specified non-cancer subsets do not change this conclusion: gnomAD v2.1 non-cancer exomes have AF 0.0194274 and gnomAD v3.1 non-cancer genomes have AF 0.0164404.

Rule

{'source': 'generic_acmg_amp', 'threshold': '>=0.05', 'threshold_publication': 'PMID:25741868'}

Evidence considered

• gnomAD v2.1 reports total AF 0.0185349, South Asian AF 0.0431147, and grpmax FAF 0.0411806.
• gnomAD v4.1 reports total AF 0.0163633, South Asian AF 0.0423366, and grpmax FAF 0.0412213.
• The generic ACMG/AMP 2015 stand-alone BA1 threshold is allele frequency >=0.05 (PMID:25741868).
• The non-cancer subset frequencies are also below 0.05: gnomAD v2.1 non-cancer exomes AF 0.0194274 and gnomAD v3.1 non-cancer genomes AF 0.0164404.

Sources

Reasoning

BP1 applies when a gene's disease mechanism is predominantly established by truncating variants, making a missense variant relatively unlikely to be causal. RET disease biology includes established missense mechanisms, including activating and loss-of-function effects, so the gene-level premise for BP1 is not satisfied.

Evidence considered

• RET is a receptor tyrosine kinase with disease-associated missense variation and kinase-domain functional effects documented in the reviewed literature.
• PMID:22837065 specifically studies functional effects of RET kinase-domain sequence variants, demonstrating that missense variation is a relevant RET disease mechanism rather than an exception to a truncating-only mechanism.
• The RET literature also includes activating missense disease alleles in endocrine neoplasia, so BP1's predominantly-truncating-gene assumption does not apply.

Sources

Reasoning

BP6 is restricted here to an exact-variant ClinVar expert-panel classification of Benign or Likely benign. ClinVar variation 13938 has zero expert-panel submissions; the benign and likely benign records are ordinary laboratory submissions, and their aggregate presence cannot activate BP6.

Rule

BP6 applies only to an exact-variant ClinVar expert-panel classification of Benign or Likely benign; ordinary laboratory submissions, aggregate labels, and non-expert assertions are excluded.

Evidence considered

• ClinVar exact variant NM_020975.6:c.2944C>T (p.Arg982Cys), variation 13938, reports zero expert-panel submissions; benign and likely benign classifications come from laboratory submissions and are not eligible for BP6 under the specified rule.

Sources

Reasoning

BS4 is not assessed. The literature documents R982C in population controls and in relatives, but it does not reliably identify clinically unaffected carriers within a disease pedigree or provide phenotype evaluation sufficient to demonstrate non-segregation. Presence in unrelated controls is population evidence, not BS4 non-segregation evidence.

Rule

Generic ACMG/AMP BS4: lack of segregation in affected family members or presence in unaffected family members when the disorder is expected to be fully penetrant; no RET-specific specification was available.

Evidence considered

• PMID:14633923 reports R982C in three patients and not in 96 ethnically matched controls, but does not establish BS4 from unaffected familial carriers.
• PMID:15741265 reports R982C in one patient and discusses its occurrence in prior families, but does not provide adequately phenotyped unaffected carriers demonstrating non-segregation.
• PMID:21655256 reports the variant in relatives and states that it was absent from 100 unrelated healthy controls; unrelated controls are not sufficient evidence for BS4.
• Generic ACMG/AMP BS4 requires lack of segregation in affected family members or presence in unaffected family members when the disorder is expected to be fully penetrant; the required familial phenotype data are unavailable.

Still missing

• No pedigree documents carrier status and clinical phenotype for unaffected relatives with sufficient evaluation for the relevant disease.
• Control-population observations cannot substitute for documented non-segregation within an informative family.

Sources