ESR1 c.1613A>G, p.Asp538Gly

Reasoning

PS3 is met at strong strength. The exact ESR1 p.Asp538Gly (D538G), corresponding to NM_000125.4:c.1613A>G, has been evaluated in multiple independent functional systems with appropriate wild-type comparisons and orthogonal readouts. In HEK293T, SKBR3, MCF7, and other breast-cancer-cell assays, D538G produced ligand-independent estrogen-receptor transcriptional activity, increased ER target-gene signaling, coactivator recruitment, proliferation or tumor growth after estrogen withdrawal, and reduced sensitivity to tamoxifen and fulvestrant. The strongest quantitative support includes SRC3 coactivator binding without estradiol in purified-protein assays (Kd 151 +/- 20 nM for D538G versus no detectable wild-type binding), altered ligand-binding affinities, and ERE-luciferase and xenograft results reproduced across several papers. These are biologically relevant assays for ESR1 activation and endocrine resistance and consistently support an altered, activating functional effect. The evidence is not described as validated and reproducible in a clinical diagnostic laboratory, so the most stringent validation note is not independently satisfied; nevertheless, the generic PS3 definition is satisfied by the convergent well-established in vitro and in vivo functional evidence.

Rule

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

Evidence considered

• PMID:24185510: In an estrogen-response-element luciferase assay in HEK293T cells, exact p.Asp538Gly showed strong constitutive activity without estrogen, with 4-hydroxytamoxifen and fulvestrant IC50 values two- to fourfold higher than wild type.
• PMID:24185512: Exact Asp538Gly showed high estrogen-independent ER transcriptional activity, increased ER target-gene expression and Ser118 phosphorylation, increased AIB1 association, and increased tumor growth in an MCF7 xenograft model after estrogen withdrawal; higher tamoxifen and fulvestrant concentrations were required for near-complete inhibition.
• PMID:24217577: Exact A1613G/D538G showed ligand-independent transcriptional activity, SRC-1 interaction, increased ER-regulated gene expression, proliferation and migration, and relative resistance to 4-hydroxytamoxifen and fulvestrant.
• PMID:24398047: Exact c.1613A>G/p.D538G was tested in ERE-luciferase assays and showed constitutive estrogen-receptor transcriptional activity, reduced antiestrogen response, and relative resistance to estrogen deprivation.
• PMID:26836308: In purified ERalpha ligand-binding-domain assays, D538G bound SRC3 without estradiol at Kd 151 +/- 20 nM while wild-type ERalpha showed no detectable binding; D538G also showed altered estradiol and 4-hydroxytamoxifen binding affinities and conformational changes by orthogonal biophysical and structural methods.

Still missing

• The reviewed studies are research assays rather than validated, reproducible clinical diagnostic-laboratory assays.

Sources

Reasoning

PM2 is met at supporting strength. NM_000125.4:c.1613A>G (ESR1 p.Asp538Gly) is absent from gnomAD v2.1 and gnomAD v4.1, and the separately reported gnomAD v2.1 non-cancer and v3.1 non-cancer entries also report absence. Thus the variant has an allele frequency of 0 in the available population datasets, which is below the generic supporting PM2 threshold of <=0.0001. No VCEP-specific population source requirement applies because no VCEP covers ESR1.

Rule

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

Evidence considered

• gnomAD v2.1 reports the variant absent.
• gnomAD v4.1 reports the variant absent.
• The separately reported gnomAD v2.1 non-cancer and v3.1 non-cancer entries also report absence; v2.1 non-cancer is reported from exomes and v3.1 non-cancer from genomes.
• No VCEP covers ESR1, so the generic ACMG/AMP framework and generic ClinGen SVI threshold supplied for this case govern.

Sources

Reasoning

NM_000125.4:c.1613A>G is a missense variant producing NP_000116.2:p.(Asp538Gly), so the generic PP3 rule selects the missense REVEL path only. REVEL is 0.914, meeting the supplied generic PP3 supporting threshold of >=0.644, but not the moderate threshold of >=0.773. SpliceAI is not used for a missense PP3 assessment, and BayesDel is not used because no generic ACMG-strength calibration is established. PP3 is met at supporting strength.

Rule

Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.).

Evidence considered

• NM_000125.4:c.1613A>G is a missense variant producing NP_000116.2:p.(Asp538Gly), so the generic missense REVEL path applies.
• The local REVEL lookup reports a score of 0.914 for this variant; source_registry key: revel.
• The supplied generic ClinGen SVI REVEL calibration sets PP3 supporting at >=0.644 and PP3 moderate at >=0.773. REVEL 0.914 meets supporting but not moderate strength (Pejaver et al. 2022, Am J Hum Genet, PMID:36413997).
• SpliceAI is excluded because the generic guidance assigns missense variants to REVEL only. BayesDel is not used because no generic ACMG-strength calibration is established.

Sources

Reasoning

PS4 requires a significantly increased prevalence in affected individuals compared with controls, with an odds ratio or relative risk >5.0 and a confidence interval that excludes 1.0. Jeselsohn et al. directly reported the exact D538G variant in 3 of 76 ER-positive metastatic breast tumors and in 0 of 115 ER-negative tumor controls. Treating these as a case-control comparison and applying a Haldane-Anscombe 0.5 correction for the zero control cell gives OR = 11.00, 95% CI 0.56-216.06. The point estimate exceeds 5.0, but the confidence interval includes 1.0; therefore the generic PS4 requirement is not met. The control group was tumor-based rather than unaffected/population controls, which further limits interpretation, but the reported counts do constitute a comparison and were not treated as absent evidence.

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

• Jeselsohn et al. 2014, Results and Table 2: exact ESR1 1613A>G/D538G was present in 3 ER-positive metastatic specimens out of 76 and absent from 115 ER-negative tumors.
• Computed the case-control odds ratio from the reported counts using a Haldane-Anscombe 0.5 correction for the zero control cell: OR = (3.5 x 115.5)/(73.5 x 0.5) = 11.00; 95% Wald CI = 0.56-216.06.
• The point estimate is above the ACMG/AMP >5.0 threshold, but the 95% confidence interval includes 1.0, so PS4 is not met.

Sources

Reasoning

The available clinical observations concern ER-positive metastatic breast cancer and acquired endocrine resistance, including somatic ESR1 D538G in treated metastatic tumors. This phenotype is not highly specific for a disease with a single genetic etiology: metastatic breast cancer and endocrine resistance have multiple molecular and clinical causes, and the evidence is somatic tumor evidence rather than a distinctive constitutional phenotype or family history. Therefore PP4 is not met.

Rule

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

Evidence considered

• PMID:24217577 reports D538G in liver metastases from 5 of 13 ER-positive metastatic breast cancer patients after prolonged endocrine treatment.
• PMID:24398047 describes D538G in ER-positive metastatic tumors and frames the finding as an acquired endocrine-resistance mechanism, not a uniquely diagnostic phenotype or family history.

Sources

Reasoning

The exact variant has a ClinVar record for NM_000125.4:c.1613A>G (p.Asp538Gly), but the record is classified Oncogenic with criteria provided by a single submitter and one-star review status. No ClinVar expert-panel classification of Pathogenic or Likely pathogenic is present. Under the specified generic rule, this does not qualify for 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 variation 3764845, exact NM_000125.4:c.1613A>G (p.Asp538Gly): classification Oncogenic, review status criteria provided/single submitter, one star, one submission, and no expert-panel entry.
• Because the record is not an expert-panel Pathogenic or Likely pathogenic classification, it cannot trigger PP5 under the specified generic rule.

Sources

Reasoning

BA1 is not met. NM_000125.4:c.1613A>G (ESR1 p.Asp538Gly) is absent from gnomAD v2.1 and gnomAD v4.1; therefore no filtering allele frequency or grpmax FAF reaches the generic stand-alone threshold of >=0.05. The gnomAD non-cancer entries are also reported as absent, and no population-specific frequency supports BA1. The somatic oncogenic ClinVar record does not establish a benign population frequency and cannot substitute for the required frequency evidence.

Rule

BA1 (stand-alone): >= 0.05 — ACMG/AMP 2015 stand-alone benign allele frequency (PMID:25741868)

Evidence considered

• gnomAD v2.1 reports the variant absent; no allele frequency, grpmax FAF, popmax estimate, or observed-allele count is available.
• gnomAD v4.1 reports the variant absent; no allele frequency, grpmax FAF, popmax estimate, or observed-allele count is available.
• The separately reported gnomAD v2.1 non-cancer and v3.1 non-cancer entries also report absence; neither provides a frequency exceeding 0.05.
• ClinVar lists an Oncogenic classification for a single somatic submission with one-star criteria-provided review status, so it is not evidence for a benign population classification.

Sources

Reasoning

BS1 is not met on the available population data. The variant is absent from gnomAD v2.1 and v4.1, including the separately reported non-cancer entries, so no filtering allele frequency or grpmax FAF is greater than the generic 0.01 default. A disorder-specific maximum credible allele-frequency calculation cannot be made because no established germline ESR1 disorder prevalence, penetrance, or allelic heterogeneity inputs are provided; this limitation does not create positive BS1 evidence. The available literature describes D538G as a recurrent somatic metastatic-breast-cancer mutation, not as a population-observed germline allele.

Rule

BS1 (strong): >= 0.01 — Generic SVI default; per-gene derivation preferred (Whiffin et al. 2017) (PMID:28518168)

Evidence considered

• gnomAD v2.1 and v4.1 report the variant absent, with no filtering allele frequency, grpmax FAF, popmax point estimate, or observed-allele count.
• The separately reported gnomAD v2.1 non-cancer and v3.1 non-cancer entries also report absence.
• The reviewed variant-specific studies describe ESR1 p.Asp538Gly as a somatic mutation in metastatic breast-cancer specimens and do not provide germline population-frequency, disease-prevalence, or penetrance estimates.
• The generic SVI default of 0.01 is used only as a comparison threshold; because the observed frequency is absent, the result does not depend on assuming that 1% is appropriate for ESR1 disease biology.

Still missing

• No established germline ESR1 disorder prevalence, penetrance, or allelic heterogeneity parameters were available for a disorder-specific maximum credible allele-frequency calculation.

Sources

Reasoning

BS3 is not met. No reviewed study demonstrates that the exact ESR1 p.Asp538Gly variant has no damaging effect on protein function or splicing. Instead, the available functional studies consistently show constitutive ESR1 activation, ligand-independent coactivator recruitment, altered ligand binding, endocrine resistance, and increased growth-related phenotypes. The available SpliceAI result is computational rather than an in vitro or in vivo functional study and therefore does not establish BS3.

Rule

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

Evidence considered

• The reviewed exact-variant studies consistently reported altered or activating ESR1 function rather than normal protein or splicing function.
• SpliceAI max delta score was 0.011, but this is an in-silico prediction and cannot satisfy the BS3 requirement for well-established in vitro or in vivo functional studies.

Still missing

• No well-established benign functional or splicing assay for the exact variant was identified.

Sources

Reasoning

The variant is missense, but the available ESR1 evidence describes recurrent activating missense mutations as the relevant mechanism in hormone-resistant metastatic breast cancer, including D538G, and does not show that disease-causing ESR1 variation is primarily truncating. The criterion's gene-level premise is not supported and is contradicted by the documented missense mechanism.

Rule

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

Evidence considered

• OncoKB identifies recurrent ESR1 alterations in hormone-resistant metastatic breast cancer and reports a gain-of-function context for D538G.
• PMID:24185510, PMID:24185512, PMID:24217577, PMID:24398047, and PMID:26836308 document recurrent ESR1 activating missense substitutions and functional consequences.
• No reviewed source establishes that ESR1 disease-causing variants are primarily truncating.

Sources

Reasoning

NM_000125.4:c.1613A>G is a missense variant, so the generic BP4 rule selects the missense REVEL path only. REVEL is 0.914, which does not meet the supplied generic BP4 supporting threshold of <=0.29. The SpliceAI max delta of 0.011 cannot provide BP4 support for a missense variant, and BayesDel has no generic ACMG-strength calibration. BP4 is not met.

Rule

Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.).

Evidence considered

• NM_000125.4:c.1613A>G is a missense variant producing NP_000116.2:p.(Asp538Gly), so the generic missense REVEL path applies.
• The local REVEL lookup reports a score of 0.914 for this variant; source_registry key: revel.
• The supplied generic ClinGen SVI REVEL calibration sets BP4 supporting at <=0.29. REVEL 0.914 does not meet that threshold (Pejaver et al. 2022, Am J Hum Genet, PMID:36413997).
• The SpliceAI max delta score of 0.011 is excluded because the generic guidance assigns missense variants to REVEL only. BayesDel is not used because no generic ACMG-strength calibration is established.

Sources

Reasoning

The exact variant has a ClinVar record classified Oncogenic by a single submitter, with one-star criteria-provided review status. There is no expert-panel Benign or Likely benign classification. The record therefore cannot trigger BP6 under the specified generic rule.

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 3764845, exact NM_000125.4:c.1613A>G (p.Asp538Gly): Oncogenic, criteria provided/single submitter, one-star review status, and no expert-panel entry.
• Because there is no exact-variant ClinVar expert-panel Benign or Likely benign classification, BP6 is not met.

Sources

Reasoning

The variant is NP_000116.2:p.(Asp538Gly), and multiple publications directly report the same ESR1 D538 substitution with constitutive receptor activity and endocrine resistance. However, the available ClinVar record is a single-submitter somatic Oncogenic classification, and the reviewed papers do not provide a prior explicitly established pathogenic classification for the same amino-acid change. The evidence therefore establishes prior observation and functional activity but does not safely satisfy the required pathogenic-classification premise for PS1 under generic ACMG/AMP.

Rule

Same amino acid change as a previously established pathogenic variant regardless of nucleotide change. ... Caveat: Beware of changes that impact splicing rather than at the amino acid/protein level.

Evidence considered

• PMID:24185510 reports p.Asp538Gly in metastatic ER-positive breast cancer and constitutive activity in a luciferase assay.
• PMID:24185512 reports recurrent p.Asp538Gly, estrogen-independent ER signaling, xenograft growth after estrogen withdrawal, and reduced endocrine-drug sensitivity.
• PMID:24217577, PMID:24398047, and PMID:26836308 directly report or characterize D538G but do not provide an authoritative pathogenic classification.
• ClinVar variation 3764845 is a single-submitter somatic Oncogenic record with criteria provided.

Still missing

• A prior authoritative pathogenic classification for ESR1 D538G is not available in the reviewed evidence.

Sources

Reasoning

The available evidence concerns somatic ESR1 p.D538G in metastatic breast tumors and does not provide a germline proband, confirmed maternity and paternity, or documented absence of family history. Therefore, the PS2 de novo requirement cannot be evaluated.

Rule

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

Evidence considered

• PMID:24398047 reports ESR1 c.1613A>G/p.D538G in metastatic tumor specimens and explicitly reports no segregation data; it does not establish a germline de novo event or parental confirmation.

Still missing

• Germline testing of the proband is not documented.
• Maternity and paternity confirmation is not documented.
• Family history and parental genotype results are not documented.

Sources

Reasoning

D538 lies in the ESR1 ligand-binding domain, which is functionally critical and contains recurrent activating mutations, and the variant is reported in a statistically significant cancer hotspot. Generic ACMG/AMP provides no VCEP-authorized domain table or hotspot route here, and the reviewed evidence does not establish absence of benign variation throughout the relevant domain. The somatic hotspot cannot independently establish PM1 under the supplied rule, while the critical-domain route remains incompletely documented.

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

• PMID:24185510 and PMID:24185512 place recurrent activating ESR1 mutations, including D538G, in the ligand-binding domain.
• PMID:26836308 provides direct structural and biochemical evidence for the critical ligand-binding domain.
• Cancer Hotspots reports ESR1 D538 as a statistically significant hotspot, but no VCEP-specific hotspot rule applies.
• The target variant is absent from gnomAD v2.1 and v4.1; this does not establish absence of benign variation across the entire domain.

Still missing

• A complete authoritative assessment showing no benign variation in the relevant ESR1 functional domain is unavailable.

Sources

Reasoning

The reviewed evidence documents D538G but does not document a different missense substitution at ESR1 residue 538 with an authoritative pathogenic determination. The PM5 candidate artifact contains no same-residue candidates and explicitly does not establish classic PM5 semantics. PM5 cannot therefore be assessed 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. ... Caveat: Beware of changes that impact splicing rather than at the amino acid/protein level.

Evidence considered

• The reviewed papers report D538G itself and other ESR1 residues such as Y537, but do not provide a different pathogenic missense substitution at residue 538.
• pm5_candidates.json contains zero same-residue candidates and states that classic PM5 semantics could not be safely confirmed.

Still missing

• A different missense variant at residue 538 with an authoritative pathogenic classification is not available.

Sources

Reasoning

The available evidence does not identify a germline proband with an assumed de novo variant and does not document parental testing. Somatic tumor observations cannot establish assumed de novo inheritance, so PM6 cannot be evaluated.

Rule

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

Evidence considered

• PMID:24398047 reports somatic metastatic tumor findings and no segregation data; it does not provide parental testing or support assumed de novo germline inheritance.

Still missing

• A germline proband-level variant result is not documented.
• Parental testing or evidence supporting assumed de novo status is not documented.

Sources

Reasoning

No affected family members, informative meioses, or genotype-phenotype segregation data are reported for this variant. The available publication describes somatic tumor specimens rather than familial segregation, so PP1 cannot be evaluated.

Rule

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

Evidence considered

• PMID:24398047 reports D538G in metastatic tumor specimens and states that no segregation data were reported; it provides no familial cosegregation evidence.

Still missing

• Affected relatives and their phenotypes are not documented.
• Genotypes of affected and unaffected relatives are not documented.
• Informative meioses or a segregation count are not documented.

Sources

Reasoning

The available evidence supports recurrent activating ESR1 missense mutations in metastatic cancer, but it does not establish the gene-level benign missense rate or demonstrate that missense variants are a common mechanism of the disease context required by PP2. gnomAD absence for this variant alone cannot establish a low benign missense rate for ESR1.

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

• OncoKB and the reviewed cancer studies document recurrent activating ESR1 missense mutations in hormone-resistant metastatic breast cancer.
• The case evidence reports the target variant absent from gnomAD v2.1 and v4.1, but provides no gene-wide benign missense rate or missense-enrichment analysis.

Still missing

• A gene-level analysis demonstrating a low rate of benign ESR1 missense variation and a qualifying disease mechanism is unavailable.

Sources

Reasoning

BS2 cannot be assessed from the available evidence. The criterion requires observation in a healthy adult in the relevant genotype for a fully penetrant early-onset disorder. The gnomAD records provide no variant observation because the variant is absent, and the paper reporting a clonal homozygous D538G finding concerns a metastatic tumor sample rather than a healthy adult. No healthy-adult heterozygous, homozygous, or hemizygous observation and no established ESR1 germline disorder with the required penetrance context were identified.

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 and v4.1 report the variant absent, so they provide no healthy-adult carrier or homozygote observation.
• PMID:24398047 reports a clonal homozygous D538G metastatic tumor specimen; this is a somatic tumor observation, not observation in a healthy adult, and therefore does not satisfy BS2.
• No healthy adult carrying this variant and no penetrance-qualified germline ESR1 disease context were documented in the reviewed evidence.

Still missing

• No healthy-adult carrier observation or established fully penetrant early-onset ESR1 disorder context was available.

Sources

Reasoning

No affected family members carrying or lacking the variant are described, and no family-level non-segregation analysis is available. The somatic tumor literature cannot establish lack of segregation in an affected family, so BS4 cannot be evaluated.

Rule

Lack of segregation in affected members of a family.

Evidence considered

• PMID:24398047 reports no segregation data and describes somatic metastatic tumors, not affected relatives tested for the germline variant; therefore it cannot establish non-segregation.

Still missing

• An affected family cohort is not documented.
• Variant genotypes for affected relatives are not documented.
• Phenocopy and multiple-variant explanations cannot be assessed without family data.

Sources

Reasoning

BP2 cannot be assessed from the available evidence. No pathogenic variant in trans or cis with NM_000125.4:c.1613A>G was documented, and no parent, offspring, tumor-normal, or other phasing result establishes the allelic relationship. The literature reports the queried ESR1 p.Asp538Gly variant in somatic metastatic breast tumors and does not provide the phase or inheritance evidence required for either BP2 branch.

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 available case evidence reports ESR1 p.Asp538Gly in somatic metastatic breast cancer and functional studies, but contains no pathogenic partner-variant observation with documented cis or trans phase.
• The available evidence does not establish a fully penetrant dominant ESR1 germline disorder for applying the trans branch of BP2.

Still missing

• No pathogenic partner variant was identified for comparison.
• No parent, offspring, tumor-normal, or other phasing data establish cis or trans status.
• No fully penetrant dominant ESR1 disorder context is established in the supplied evidence.

Reasoning

The available evidence documents ESR1 D538G in metastatic breast tumors and functional endocrine resistance, but does not establish that this variant was found in a case with an alternate molecular basis for the disease. The required case-level alternate molecular diagnosis is unavailable.

Rule

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

Still missing

• A case-level diagnosis and molecular findings demonstrating an alternate molecular basis are required to evaluate BP5.