ATM c.4306C>T, p.His1436Tyr

Reasoning

The variant is missense, and the governing VCEP BP4 missense path is REVEL <=.249. The REVEL score is 0.166, which satisfies that threshold, so BP4 is met at Supporting strength. The VCEP also lists a splice path, but the missense protein-impact path is sufficient and the same prediction is not counted again; the local gene note's missense-only approach reaches the same outcome.

VCEP rule

* Missense: REVEL score **≤.249** * Splicing: No predicted impact via splicing (SpliceAI **≤0.1**).

Evidence considered

• ClinGen ATM VCEP v1.6 BP4 rule: Missense REVEL score <=.249; the VCEP separately lists a SpliceAI <=0.1 splicing path.
• REVEL lookup for ATM NM_000051.4:c.4306C>T: score 0.166, satisfying the missense BP4 threshold.
• Sun et al. 2025 Supplementary Table S1, exact row ATM c.4306C>T / H1436Y, exon_29, Missense: REVEL 0.166; Classification Functional; Confidence Computational prediction; DeepATM_predicted Yes. This table value corroborates the primary REVEL lookup.

Sources

Reasoning

The ATM VCEP PS1 rule requires the same amino acid change as a previously established pathogenic variant and, for missense changes, that splicing be ruled out for both alterations. The required ATM_PS1_1.6.pdf file was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no entry was found. The exact-variant literature reports this variant as VUS or describes it without establishing pathogenicity, so no previously established pathogenic same-amino-acid comparator is available. SpliceAI predicts no significant splice impact for the assessed variant (maximum delta score 0.075), but that does not supply the missing pathogenic comparator.

VCEP rule

Same amino acid change as a previously established pathogenic variant regardless of nucleotide change. ... Missense: This rule may be applied only when a splice defect is ruled out for both alterations either by RNA analysis and/or in silico splice predictions

Evidence considered

• ATM PS1 VCEP guide: the missense PS1 route requires the same amino-acid change as a previously established pathogenic variant and requires splice defect exclusion for both alterations.
• ATM_PS1_1.6.pdf was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no matching entry was found.
• PMID:32066632, Table 2, P18: ATM c.4306C>T is reported as a missense VUS in 1 FIGC proband and 1 TSI control.
• PMID:32325837, Table 2, CM (50): ATM c.4306C>T:p.(His1436Tyr) is reported as a missense VUS.
• PMID:36555667, Results section 2.1: p.His 1436Tyr is mentioned as previously described in breast and prostate cancer, without a pathogenic classification or variant-specific functional result in the extracted text.
• Case splice evidence: SpliceAI maximum delta score is 0.075, below the VCEP's splice-impact threshold, but this does not establish the required pathogenic comparator.

Still missing

• ATM_PS1_1.6.pdf was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no entry for this variant was found.
• No previously established pathogenic ATM variant with the same amino-acid change was identified in the available governing VCEP materials or exact-variant literature.

Sources

Reasoning

The VCEP requires a case-control study with p-value ≤.05 and either an odds ratio, hazard ratio, or relative risk ≥2, or a lower 95% confidence interval ≥1.5. Carvalho et al. (PMID:32066632) reported this exact variant in 1/50 familial intestinal gastric cancer probands and 1/107 TSI controls, which is a case-control comparison. Using the reported counts, I calculated OR=(1×106)/(49×1)=2.163; the two-sided Fisher exact p-value is 0.536910 and the Wald 95% CI is 0.132551–35.305157. The p-value criterion is not met and the lower CI is below 1.5, so the VCEP PS4 rule is not met despite the point estimate being ≥2. The other reviewed case-control studies reported aggregate ATM variant classes or did not report this variant specifically and therefore cannot establish variant-specific PS4.

VCEP rule

Case-control studies; p-value ≤.05 **and** (Odds ratio, hazard ratio, or relative risk ≥2 **or** lower 95% CI ≥1.5).

Evidence considered

• ClinGen HBOP ATM VCEP v1.6 PS4 rule: case-control studies require p-value ≤.05 and (odds ratio, hazard ratio, or relative risk ≥2 or lower 95% CI ≥1.5).
• Carvalho et al. 2021, Table 2, P18: ATM c.4306C>T was present in 1 of 50 FIGC probands and 1 of 107 TSI controls.
• From those counts, I calculated OR=(1×106)/(49×1)=2.163265; Wald 95% CI=0.132551–35.305157; two-sided Fisher exact p=0.536910.
• Decker et al. 2017 analyzed aggregate rare ATM missense variants, not this variant: 489/13,087 cases versus 175/5,488 controls, OR 1.18, 95% CI 0.99–1.40, p=0.073.
• Dalmasso et al. 2021 reported aggregate ATM variant-class comparisons and did not report this variant specifically; its aggregate result cannot be used as variant-specific PS4 evidence.

Sources

Reasoning

The ATM VCEP applies PM2 Supporting when frequency is at most 0.001% in the gnomAD subpopulation with the highest frequency, with an exception for a frequency above 0.001% only when n=1 in a single subpopulation and absent from every other subpopulation. The highest reported subpopulation frequencies exceed the threshold: gnomAD v2.1 all-comers African/African American AF 0.01854% (3 alleles), gnomAD v4.1 all-comers African/African American AF 0.00400% (3 alleles), gnomAD v2.1 non-cancer exomes African/African American AF 0.02023% (3 alleles), and gnomAD v3.1 non-cancer genomes 1KG:TSI AF 0.48544% (1 allele). The single-allele v3.1 TSI observation does not qualify for the exception because the variant is also present in other subpopulations in that dataset.

VCEP rule

Frequency ≤.001% in gnomAD subpopulation with the highest frequency ... Exception: If >0.001% but n=1 in a single subpopulation (not present in any other subpopulation), that is sufficiently rare and PM2_supporting would apply.

Evidence considered

• ATM VCEP v1.6 CSPEC PM2 entry: "Frequency ≤.001% in gnomAD subpopulation with the highest frequency ... Exception: If >0.001% but n=1 in a single subpopulation (not present in any other subpopulation), that is sufficiently rare and PM2_supporting would apply."
• gnomAD v2.1 all-comers: highest reported population AF 0.000185414 (0.01854%), African/African American, 3/16180 alleles; gnomAD v4.1 all-comers: highest reported population AF 4.00352e-05 (0.00400%), African/African American, 3/74934 alleles.
• gnomAD v2.1 non-cancer exomes: highest reported population AF 0.000202265 (0.02023%), African/African American, 3/14832 alleles; gnomAD v3.1 non-cancer genomes: highest reported population AF 0.00485437 (0.48544%), 1KG:TSI, 1/206 alleles. The v3.1 non-cancer record also reports the variant in other subpopulations, so the n=1 exception is not satisfied.

Sources

Reasoning

The variant is missense, so the VCEP PP3 rule permits either REVEL >.7333 or predicted splice impact by SpliceAI >=0.2 for missense variants. REVEL is 0.166 and SpliceAI max delta is 0.075, so neither governing path reaches its threshold. The exact VCEP Table S1 row for c.4306C>T lists REVEL 0.166 and DeepATM_predicted=Yes with Confidence=Computational prediction; this model prediction is corroborative only and does not meet either PP3 rule path. No other computational predictor was used because it is not named by the governing PP3 rule.

VCEP rule

* Missense: REVEL **\>.7333** * Splicing: Predicted impact via splicing (SpliceAI **≥0.2**) for silent, missense/in-frame and for intronic variants outside of donor and acceptor 1,2 sites.

Evidence considered

• ClinGen ATM VCEP v1.6 PP3 rule: missense REVEL >.7333, or predicted splice impact via SpliceAI >=0.2 for silent, missense/in-frame, and applicable intronic variants.
• REVEL lookup for ATM NM_000051.4:c.4306C>T: score 0.166.
• SpliceAI Lookup for NM_000051.4:c.4306C>T: DS_AG 0.005, DS_AL 0.002, DS_DG 0.075, DS_DL 0.002; maximum delta score 0.075.
• Sun et al. 2025 Supplementary Table S1, exact row ATM c.4306C>T / H1436Y, exon_29, Missense: REVEL 0.166; Classification Functional; Confidence Computational prediction; DeepATM_predicted Yes. The computational prediction is secondary corroboration only, not an assay result or an additional PP3 path.

Sources

Reasoning

The ATM VCEP applies BA1 when the gnomAD group maximum filtering allele frequency is greater than 0.5%. The available all-comers data show gnomAD v2.1 grpmax FAF 5.0e-05 (0.005%) and gnomAD v4.1 joint grpmax FAF 2.475e-05 (0.002475%); the VCEP-calibrated threshold is not reached. The VCEP's non-cancer datasets are also below the threshold: gnomAD v2.1 non-cancer exomes grpmax FAF 5.443e-05 (0.005443%), and gnomAD v3.1 non-cancer genomes grpmax FAF 1.171e-05 (0.001171%).

VCEP rule

Grpmax Filtering AF >.5% in gnomAD

Evidence considered

• ATM VCEP v1.6 CSPEC BA1 entry: "Grpmax Filtering AF >.5% in gnomAD".
• gnomAD v2.1 all-comers: grpmax FAF 5.0e-05 (0.005%), based on 8/250480 alleles; gnomAD v4.1 all-comers: joint grpmax FAF 2.475e-05 (0.002475%), based on 44/1612928 alleles.
• gnomAD v2.1 non-cancer exomes: grpmax FAF 5.443e-05 (0.005443%); gnomAD v3.1 non-cancer genomes: grpmax FAF 1.171e-05 (0.001171%).

Sources

Reasoning

The ATM VCEP applies BS1 when the gnomAD group maximum filtering allele frequency is greater than 0.05%. The available all-comers data show gnomAD v2.1 grpmax FAF 5.0e-05 (0.005%) and gnomAD v4.1 joint grpmax FAF 2.475e-05 (0.002475%); both are below 0.05%. The non-cancer datasets are also below the threshold: gnomAD v2.1 non-cancer exomes grpmax FAF 5.443e-05 (0.005443%), and gnomAD v3.1 non-cancer genomes grpmax FAF 1.171e-05 (0.001171%).

VCEP rule

Grpmax Filtering AF >.05% in gnomAD

Evidence considered

• ATM VCEP v1.6 CSPEC BS1 entry: "Grpmax Filtering AF >.05% in gnomAD".
• gnomAD v2.1 all-comers: grpmax FAF 5.0e-05 (0.005%); gnomAD v4.1 all-comers: joint grpmax FAF 2.475e-05 (0.002475%).
• gnomAD v2.1 non-cancer exomes: grpmax FAF 5.443e-05 (0.005443%); gnomAD v3.1 non-cancer genomes: grpmax FAF 1.171e-05 (0.001171%).

Sources

Reasoning

The ClinGen HBOP ATM supplementary assay table was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no variant-specific entry was found, and the table itself states that it is an assay-methodology and strength-calibration table rather than a per-variant results table. Sun et al. 2025 Supplementary Table S1 contains the exact row for c.4306C>T / H1436Y: Classification=Functional, Confidence=Computational prediction, and DeepATM_predicted=Yes. Under the VCEP file-specific rule, this is an unassayed computational prediction and cannot support PS3. No qualifying variant-specific failure of an ATM-specific functional feature or radiosensitivity assay was identified.

VCEP rule

Use when a variant fails to rescue an ATM specifc feature, only (e.g. phosphorylation of ATM-specific targets). Do not use for radiosensitivity-only as that is not a feature specific to ATM deficiency

Evidence considered

• Sun et al. 2025 Supplementary Table S1, exact row c.4306C>T / H1436Y: Classification=Functional; Confidence=Computational prediction; DeepATM_predicted=Yes.
• ClinGen HBOP ATM Supplementary Tables 1 and 2, ATM kinase activity and ATM radiosensitivity sheets: the file describes approved assay methodologies and strength calibration but contains no row for c.4306C>T, p.His1436Tyr, or p.H1436Y.
• The VCEP-specific Table S1 policy states that when DeepATM_predicted=Yes, the variant was not assayed and a prediction cannot be the only basis for PS3/BS3.

Still missing

• No entry for c.4306C>T, p.His1436Tyr, or p.H1436Y was found in clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx; variant-specific approved-assay results remain unavailable.
• No direct ATM-specific functional assay result demonstrating failure to rescue an ATM-specific feature and/or radiosensitivity was available for this variant.

Sources

Reasoning

The governing ATM PM3/BP2 specification requires recessive-disease evidence with the variant detected in trans with a pathogenic variant and assigns PM3 points according to affected A-T proband phenotype and phase. The case evidence reports ATM c.4306C>T/p.His1436Tyr in a familial intestinal gastric cancer proband (PMID:32066632), a melanoma patient (PMID:32325837), and a melanoma cohort paper (PMID:36555667), but provides no A-T diagnosis, affected A-T phenotype, second pathogenic ATM variant, or phase determination for any observation. The governing file ATM_PM3_BP2_1.6.pdf.txt was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no variant-specific entry was found. Therefore no VCEP PM3 points or strength can be assigned.

VCEP rule

PM3: For recessive disorders, detected in trans with a pathogenic variant; use the ATM PM3/BP2 points table for affected A-T probands and assign PM3 strength from the accumulated points.

Evidence considered

• ATM PM3/BP2 specification, page 1: PM3 is for recessive disorders when the variant is detected in trans with a pathogenic variant.
• ATM PM3/BP2 specification, page 2: affected A-T probands receive 4.0 points for a confident phenotype with confirmed in-trans phase, 2.0 with phase unknown, 1.0 when the second variant is unidentified or VUS, and 2.0 when homozygous; consistent-phenotype values are 2.0, 1.0, 0.5, and 1.0 respectively.
• ATM PM3/BP2 specification, page 3, footnote 2: in-trans phase can be established by genotyping at least one parent or assumed when the variant under assessment is seen with at least two different pathogenic or likely pathogenic variants.
• Carvalho et al. 2021, PMID:32066632, Table 2 row P18: ATM c.4306C>T was reported in one familial intestinal gastric cancer proband and one control; the report does not establish A-T phenotype, a second pathogenic ATM allele, or phase.
• Pastorino et al. 2020, PMID:32325837, Table 2 ATM row: ATM c.4306C>T:p.(His1436Tyr) was reported in one melanoma patient as a VUS; no variant-specific co-occurring pathogenic ATM variant or phase was reported.
• Pastorino et al. 2022, PMID:36555667, Results section 2.1: p.His1436Tyr was mentioned as previously described in breast and prostate cancer; no A-T proband, phase, or second pathogenic ATM allele was reported in the extracted finding.
• The governing ATM_PM3_BP2_1.6.pdf.txt file was searched using c.4306C>T, p.His1436Tyr, and p.H1436Y; no matching variant-specific row or pre-assigned PM3 code was found.

Still missing

• ATM_PM3_BP2_1.6.pdf.txt was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no variant-specific entry was found.
• No affected A-T proband phenotype, second pathogenic or likely pathogenic ATM variant, or confirmed phase in trans was documented for the reported cancer-cohort observations.

Sources

Reasoning

The ATM VCEP v1.6 specifies PP1 for an autosomal-recessive condition as segregation in at least 1 affected relative for Supporting, 2 affected relatives for Moderate, or at least 3 affected relatives for Strong. The available variant-specific reports identify the variant in individual probands but provide no affected-relative segregation, parental testing, meiosis count, or co-segregation analysis; one report explicitly states that co-segregation analysis was not performed. The PP1 rule therefore cannot be applied.

VCEP rule

AR Condition: Segregation in ≥3 affected relatives ... AR Condition: Segregation in 2 affected relatives ... AR Condition: Segregation in 1 affected relative

Evidence considered

• ClinGen HBOP ATM VCEP v1.6 PP1 rule: "AR Condition: Segregation in ≥3 affected relatives" = Strong, "AR Condition: Segregation in 2 affected relatives" = Moderate, and "AR Condition: Segregation in 1 affected relative" = Supporting.
• Carvalho et al. (PMID:32066632), Table 2 row P18: ATM c.4306C>T was reported in one FIGC proband; the extracted report provides no segregation analysis or causative claim.
• Pastorino et al. (PMID:32325837), ATM variant row for the melanoma case: the variant was reported in one proband and co-segregation analysis was not performed.

Still missing

• Affected-relative segregation data
• Parental testing and phase information
• Number of informative meioses

Sources

Reasoning

The ClinGen HBOP ATM supplementary assay table was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no variant-specific entry was found, and the table itself states that it is an assay-methodology and strength-calibration table rather than a per-variant results table. Sun et al. 2025 Supplementary Table S1 contains the exact row for c.4306C>T / H1436Y: Classification=Functional, Confidence=Computational prediction, and DeepATM_predicted=Yes. Under the VCEP file-specific rule, this is an unassayed computational prediction and cannot support BS3. No qualifying variant-specific rescue of an ATM-specific feature or radiosensitivity assay was identified.

VCEP rule

Use when a variant rescues EITHER an ATM specifc feature OR rescues radiosensitivity.

Evidence considered

• Sun et al. 2025 Supplementary Table S1, exact row c.4306C>T / H1436Y: Classification=Functional; Confidence=Computational prediction; DeepATM_predicted=Yes.
• ClinGen HBOP ATM Supplementary Tables 1 and 2, ATM kinase activity and ATM radiosensitivity sheets: the file describes approved assay methodologies and strength calibration but contains no row for c.4306C>T, p.His1436Tyr, or p.H1436Y.
• The VCEP-specific Table S1 policy states that when DeepATM_predicted=Yes, the variant was not assayed and a prediction cannot be the only basis for PS3/BS3.

Still missing

• No entry for c.4306C>T, p.His1436Tyr, or p.H1436Y was found in clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx; variant-specific approved-assay results remain unavailable.
• No direct ATM-specific functional assay result demonstrating rescue of an ATM-specific feature and/or radiosensitivity was available for this variant.

Sources

Reasoning

The governing ATM BP2 specification applies to the variant observed in trans with a pathogenic variant in a fully penetrant dominant disorder or observed in cis with a pathogenic variant in any inheritance pattern, with points assigned for an unaffected non-A-T adult proband. The available reports of ATM c.4306C>T/p.His1436Tyr do not document an unaffected adult with confirmed in-trans phase, a pathogenic or likely pathogenic ATM variant in cis, or the laboratory/database setting required by the BP2 table. The governing file ATM_PM3_BP2_1.6.pdf.txt was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no variant-specific entry was found. Therefore no VCEP BP2 points or strength can be assigned.

VCEP rule

BP2: Observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder or observed in cis with a pathogenic variant in any inheritance pattern; use the ATM PM3/BP2 table for BP2 points.

Evidence considered

• ATM PM3/BP2 specification, page 2: BP2 is 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.
• ATM PM3/BP2 specification, page 2: for an unaffected non-A-T adult proband, confirmed in trans with a pathogenic or likely pathogenic variant is -4.0 points, phase unknown is -2.0 points, and homozygous observations are scored in laboratory or database settings as -2.0 or -1.0 respectively.
• ATM PM3/BP2 specification, page 2, footnote 3: the unaffected individual should be older than 18 years with no evidence of A-T, and the VUA should not be predicted or experimentally proven to have a clinically important splicing effect.
• Carvalho et al. 2021, PMID:32066632, Table 2 row P18: the report includes one cancer proband and one control observation but does not provide the qualifying unaffected-adult BP2 phase and allele details.
• Pastorino et al. 2020, PMID:32325837, Table 2 ATM row: the variant was reported in a melanoma patient, with no documented cis/trans relationship to a pathogenic ATM allele.
• Pastorino et al. 2022, PMID:36555667, Results section 2.1: the variant was referenced in cancer reports, without an unaffected adult BP2 observation or phase information.
• The governing ATM_PM3_BP2_1.6.pdf.txt file was searched using c.4306C>T, p.His1436Tyr, and p.H1436Y; no matching variant-specific row or pre-assigned BP2 code was found.

Still missing

• ATM_PM3_BP2_1.6.pdf.txt was searched for c.4306C>T, p.His1436Tyr, and p.H1436Y; no variant-specific entry was found.
• No qualifying unaffected non-A-T adult proband, pathogenic or likely pathogenic ATM allele in cis or trans, confirmed phase, or BP2 point-setting context was documented.

Sources