ATM c.1440A>C, p.(Leu480Phe)

Reasoning

ATM c.1440A>C is a missense substitution, p.(Leu480Phe), rather than a null variant. The ATM VCEP PVS1 decision guide limits PVS1 to nonsense, frameshift, canonical ±1/2 splice, initiation-codon, or single/multi-exon deletion variants. SpliceAI predicts no significant splice impact (maximum delta 0.053); therefore, there is no evidence that this exonic missense change creates a loss-of-function transcript subject to NMD.

Rule

ATM VCEP v1.5 PVS1 decision tree: null variants only (nonsense, frameshift, canonical ±1/2 splice, initiation codon, or single/multi-exon deletion).

Evidence considered

• The case normalization identifies NM_000051.4:c.1440A>C as NP_000042.3:p.(Leu480Phe), a missense substitution.
• The ATM VCEP PVS1 guide specifies PVS1 for null-variant classes and cautions on NMD and transcript effects; this variant is not in a PVS1 null-variant class.
• SpliceAI maximum delta score is 0.053, below the ATM VCEP PP3 splice-impact threshold of 0.2 and consistent with no predicted significant splice impact.
• PMID:19638463 explicitly reports the exact protein change L480F as a coding variant.

Sources

Reasoning

The governing ATM VCEP v1.5 PS1 specification permits PS1 for a missense change only when it produces the same amino-acid change as an established pathogenic or likely pathogenic variant and splicing is ruled out for both alterations. The case variant is ATM p.Leu480Phe, and PMID:19638463 reports L480F in one prostate brachytherapy patient, but does not establish that it is pathogenic or likely pathogenic; no qualifying P/LP same-amino-acid comparator is present in the case evidence bundle. SpliceAI predicts no significant splice impact (maximum delta 0.053), but this does not satisfy the missing P/LP comparator requirement.

Rule

Use for missense changes as long as splicing is ruled out for both alterations; standard PS1 requires the same amino-acid change as an established P/LP variant.

Evidence considered

• ATM VCEP v1.5 PS1 guidance requires the same amino-acid change as an established P/LP variant for the standard missense PS1 pathway, with splicing ruled out for both alterations.
• PMID:19638463 reports ATM L480F as a novel coding variant observed heterozygously in 1 of 21 high-radiation-toxicity patients and in none of 20 low-toxicity patients; the paper does not establish a P/LP classification or a statistically significant association.
• SpliceAI maximum delta score is 0.053, reported as no significant splice impact for this variant.

Still missing

• No established pathogenic or likely pathogenic ATM p.Leu480Phe comparator was identified in the supplied evidence.

Sources

Reasoning

The applicable ClinGen HBOP ATM VCEP v1.5 PM2 rule applies PM2 Supporting when the variant frequency is <=0.001% in gnomAD v4, including when n=1 in a single subpopulation. gnomAD v4.1 shows AF 5.88571e-05 (0.00589%) overall and grpmax FAF 6.641e-05 (0.006641%), both exceeding 0.001%. Therefore the VCEP PM2 threshold is not satisfied.

Rule

ClinGen HBOP ATM VCEP v1.5: PM2 Supporting when frequency <=0.001% in gnomAD v4.

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 specifies PM2 Supporting for frequency <=0.001% in gnomAD v4.
• gnomAD v4.1 reports 95/1,614,080 alleles, AF 5.88571e-05, equivalent to 0.00589%, and grpmax FAF 6.641e-05, equivalent to 0.006641%; both exceed the VCEP PM2 threshold. The variant is absent from gnomAD-Canada v1.0, but that does not override the VCEP's gnomAD v4 threshold assessment.

Sources

Reasoning

ATM c.1440A>C causes the single amino-acid substitution p.(Leu480Phe), with no change in protein length. The ATM VCEP specification restricts PM4 to stop-loss variants; this variant is not a stop-loss variant.

Rule

ATM VCEP v1.5 PM4: use for stop-loss variants.

Evidence considered

• The case normalization identifies p.(Leu480Phe), indicating an amino-acid substitution rather than an in-frame length change or stop-loss.
• The ATM VCEP PM4 rule states: Use for stop-loss variants.
• PMID:19638463 explicitly reports the exact protein change L480F as a coding variant.

Sources

Reasoning

ATM HBOP VCEP v1.5 permits PP3_Supporting for a missense variant with REVEL >0.7333 or for a predicted splice impact with SpliceAI >=0.2. This missense variant has REVEL 0.312 and maximum SpliceAI delta 0.053; neither calibrated threshold is met.

Rule

ATM HBOP VCEP v1.5 PP3_Supporting: missense REVEL >0.7333; splicing predicted impact by SpliceAI >=0.2.

Evidence considered

• The governing ATM HBOP VCEP v1.5 PP3 rule is REVEL >0.7333 for missense variants or SpliceAI >=0.2 for predicted splice impact.
• The local REVEL lookup reports 0.312 for ATM c.1440A>C (p.Leu480Phe), below the VCEP PP3 threshold.
• SpliceAI reports a maximum delta score of 0.053, below the VCEP PP3 splice-impact threshold.
• PMID:19638463 reports a BLOSUM62 conservation score of 0 for p.L480F, but supplies no published, calibrated PP3 threshold; it is therefore not used to assign PP3.

Sources

Reasoning

The applicable ClinGen HBOP ATM VCEP v1.5 BA1 rule requires gnomAD v4 grpmax filtering AF >0.5%. The observed gnomAD v4.1 grpmax FAF is 0.00006641 (0.006641%), far below 0.5%; therefore BA1 is not met.

Rule

ClinGen HBOP ATM VCEP v1.5: BA1 Stand Alone when grpmax filtering AF >0.5% in gnomAD v4.

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 specifies BA1 for grpmax filtering AF >0.5% in gnomAD v4.
• gnomAD v4.1 reports total AF 5.88571e-05, European non-Finnish AF 7.96593e-05, and grpmax FAF 6.641e-05 (0.006641%), with 95 of 1,614,080 alleles and no homozygotes.

Sources

Reasoning

The applicable ClinGen HBOP ATM VCEP v1.5 BS1 rule requires gnomAD v4 grpmax filtering AF >0.05%. The observed gnomAD v4.1 grpmax FAF is 0.00006641 (0.006641%), below 0.05%; therefore BS1 is not met.

Rule

ClinGen HBOP ATM VCEP v1.5: BS1 Strong when grpmax filtering AF >0.05% in gnomAD v4.

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 specifies BS1 Strong for grpmax filtering AF >0.05% in gnomAD v4.
• gnomAD v4.1 reports grpmax FAF 6.641e-05 (0.006641%), the maximum population filtering allele frequency observed for this variant; this is below the VCEP BS1 threshold.

Sources

Reasoning

The ATM laboratory BP4 missense/splicing override requires use of the missense sub-path only for this missense variant. REVEL is 0.312, which is above the BP4 threshold of <=0.249 and in the specified 0.249-0.7333 gray zone. Although SpliceAI is 0.053 (<=0.1), the override changes the literal VCEP splicing-path outcome: clean SpliceAI cannot establish BP4 for a missense variant when REVEL does not meet the benign threshold.

Rule

ATM HBOP VCEP v1.5 BP4_Supporting: missense REVEL <=0.249 or SpliceAI <=0.1; local ATM override: for missense variants, assess BP4 only with REVEL <=0.249.

Evidence considered

• The governing ATM HBOP VCEP v1.5 BP4 rule is REVEL <=0.249 for missense variants or SpliceAI <=0.1 for no predicted splice impact.
• The local ATM laboratory override directs that a missense variant be evaluated only through REVEL for BP4 and specifies not_met/not_assessed when REVEL is in the 0.249-0.7333 gray zone regardless of SpliceAI.
• The local REVEL lookup reports 0.312, so the missense BP4 threshold is not met.
• SpliceAI maximum delta is 0.053, but this result is not independently used for BP4 because the variant is missense and the override applies.

Sources

Reasoning

The ClinGen HBOP ATM VCEP v1.5 specification requires a variant-specific functional result showing failure to rescue an ATM-specific feature for PS3_Supporting, and failure to rescue both an ATM-specific feature and radiosensitivity for PS3_Moderate; PS3_Strong is not used. The approved-assay table identifies kinase-activity assays and a radiosensitivity assay, but explicitly states that the table is methodology/calibration and not a per-variant result table. The only exact variant-specific publication finding is observation of p.L480F in one patient in a radiosensitivity cohort, not an engineered variant assay or an ATM-specific functional readout. The Sun et al. Table S1 row for c.1440A>C is labeled DeepATM_predicted=Yes with computational-prediction confidence, so it is not direct functional evidence. Therefore no qualifying variant-specific assay result is available to assign PS3.

Rule

ATM VCEP v1.5: PS3 Supporting requires failure to rescue an ATM-specific feature only; PS3 Moderate requires failure to rescue both an ATM-specific feature and radiosensitivity; do not use PS3 Strong.

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 specifies PS3_Supporting for failure to rescue an ATM-specific feature alone, PS3_Moderate for failure to rescue both an ATM-specific feature and radiosensitivity, and says not to use PS3_Strong.
• The VCEP-approved functional-assay table describes kinase-activity and radiosensitivity methodologies and their calibration, but warns that it is not a per-variant results table and requires confirmation that the variant was actually tested in the cited paper.
• The exact ATM p.L480F variant was observed heterozygously in 1 of 21 high-radiation-toxicity patients and in none of 20 low-toxicity patients in PMID:19638463; this is a cohort observation, not a variant-specific rescue/failure assay, and the paper reported no statistically significant association for this variant.
• Sun et al. Table S1 (PMID 40580951 source file) lists c.1440A>C/p.L480F as Functional but DeepATM_predicted=Yes and Confidence=Computational prediction, indicating that the variant was not directly assayed and cannot establish PS3.
• The observed radiosensitivity cohort result cannot be treated as PS3 because it lacks an ATM-specific functional readout and does not demonstrate failure of rescue by the variant.

Still missing

• A variant-specific validated assay measuring an ATM-specific feature, such as phosphorylation of ATM substrates, is not documented.
• A variant-specific approved radiosensitivity result and evidence of concordance with an ATM-specific functional assay are not documented.
• The exact p.L480F construct/readout in one of the VCEP-approved assay papers has not been confirmed.

Sources

Reasoning

ATM HBOP VCEP v1.5 requires a case-control study with p<=0.05 and OR, HR, or RR >=2 or lower 95% CI >=1.5. The only exact-variant observation identified was p.Leu480Phe in 1/21 high-radiotoxicity versus 0/20 low-radiotoxicity prostate-brachytherapy patients; the paper reports no significant association for this site. No qualifying exact-variant cancer or disease case-control enrichment result is available.

Rule

ATM HBOP VCEP v1.5: case-control studies with p-value <=0.05 AND OR, HR, or RR >=2 OR lower 95% CI >=1.5.

Evidence considered

• The applicable ATM HBOP VCEP v1.5 PS4 rule requires p<=0.05 plus OR/HR/RR >=2 or lower 95% CI >=1.5.
• PMID:19638463 reports ATM p.L480F in 1 of 21 high-toxicity and 0 of 20 low-toxicity prostate brachytherapy patients, and reports P values >0.05 for all sites other than the specified MDC1 variant.

Still missing

• A well-powered, exact-variant case-control study reporting a qualifying association statistic and confidence interval is required.

Sources

Reasoning

The ClinGen HBOP ATM VCEP v1.5 PM3/BP2 points-based framework requires observations in unrelated individuals with an A-T phenotype, with points determined by phenotype confidence and whether the assessed variant is confirmed in trans with a VCEP-classified P/LP ATM variant, phase unknown, a second variant unidentified/VUS, or homozygous. The bundle contains no affected A-T proband observation, second pathogenic/likely pathogenic ATM variant, or phase information for NM_000051.4:c.1440A>C (p.Leu480Phe). The exact-variant observation in PMID:19638463 is a heterozygous finding in a prostate brachytherapy toxicity cohort, not an A-T affected-proband observation and not a documented trans/cis observation. Therefore no PM3 points can be assigned from the available evidence.

Rule

ClinGen HBOP ATM VCEP v1.5 PM3/BP2 points table; assign PM3 only from qualifying unrelated A-T proband observations and convert total PM3 points to Supporting (1), Moderate (2), Strong (4), or Very Strong (≥8).

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 specifies PM3 for recessive A-T using additive points per unrelated A-T proband: confident phenotype scores 4.0 for confirmed trans, 2.0 for phase unknown, 1.0 for a second variant unidentified/VUS, or 2.0 for homozygosity (maximum two individuals); consistent phenotype receives half these values. PM3 strength is Supporting at 1 point, Moderate at 2 points, Strong at 4 points, and Very Strong at 8 or more points.
• The consolidated case evidence reports no A-T affected-proband, trans, cis, homozygous, or second-ATM-variant observation for NM_000051.4:c.1440A>C.
• PMID:19638463 reports ATM p.L480F heterozygously in 1 of 21 prostate brachytherapy patients with high late toxicity and in none of 20 low-toxicity patients; it does not report A-T, a second ATM variant, or phase for this variant.

Still missing

• Affected A-T proband case data for the assessed variant.
• A second ATM variant classified as pathogenic or likely pathogenic under the ATM VCEP framework.
• Parental or other phase testing demonstrating the assessed variant in trans with the second P/LP ATM variant, or qualifying homozygosity/phase-unknown observations in an affected individual.
• Phenotype details sufficient to classify any affected individual as confident or consistent A-T under the ATM VCEP criteria.

Sources

Reasoning

PP1 is applicable for the ATM autosomal-recessive condition, but the case bundle contains no documented affected relatives, parental testing, phase information, meioses, or family-level segregation analysis for NM_000051.4:c.1440A>C (p.Leu480Phe). PMID:19638463 reports the variant in one prostate brachytherapy patient, but does not report segregation with an ATM-related phenotype. Therefore no PP1 strength can be assigned.

Rule

ATM VCEP v1.5 PP1: AR condition; Supporting = segregation in 1 affected relative, Moderate = segregation in 2 affected relatives, Strong = segregation in ≥3 affected relatives.

Evidence considered

• ClinGen HBOP ATM VCEP version 1.5: PP1 is applicable for the autosomal-recessive condition; Supporting requires segregation in 1 affected relative, Moderate in 2 affected relatives, and Strong in at least 3 affected relatives.
• PMID:19638463 reports ATM p.L480F heterozygously in 1 of 21 high-radiation-toxicity patients and in none of 20 low-toxicity patients, without affected-relative segregation, parental testing, or a statistically significant variant-specific association; this is not PP1 evidence.

Still missing

• Documented affected relatives carrying the variant and their ATM-related phenotype
• Parental genotypes and phase information
• Number of informative meioses or affected relatives
• Family-level segregation analysis

Sources

Reasoning

The ClinGen HBOP ATM VCEP v1.5 specification requires a variant-specific functional result showing rescue of an ATM-specific feature for BS3_Supporting, and rescue of both an ATM-specific feature and radiosensitivity for BS3_Moderate. The approved-assay table is a methodology/calibration resource and does not establish which variants were tested. No exact p.L480F rescue result, ATM-specific phosphorylation result, or qualifying combined kinase/radiosensitivity result is present. The Sun et al. Table S1 entry is computationally predicted rather than directly assayed (DeepATM_predicted=Yes), and the exact-variant observation in the radiosensitivity cohort is not a rescue assay. Therefore BS3 cannot be assigned.

Rule

ATM VCEP v1.5: BS3 Supporting requires rescue of either an ATM-specific feature or radiosensitivity; BS3 Moderate requires rescue of both.

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 specifies BS3_Supporting when a variant rescues either an ATM-specific feature or radiosensitivity, and BS3_Moderate when it rescues both.
• The VCEP-approved functional-assay table calibrates approved assay types but explicitly states that it is not a per-variant results table; the variant must be confirmed in the cited assay paper.
• PMID:19638463 reports p.L480F as an observed patient variant in a radiation-toxicity cohort, not as an engineered rescue experiment, and therefore does not demonstrate benign functional rescue.
• Sun et al. Table S1 (PMID 40580951 source file) lists c.1440A>C/p.L480F with DeepATM_predicted=Yes and computational-prediction confidence; this is not direct assay evidence and cannot establish BS3.
• No variant-specific normal ATM kinase-substrate phosphorylation result or validated radiosensitivity rescue result is available in the case bundle.

Still missing

• A variant-specific validated result showing rescue of an ATM-specific functional feature is not documented.
• A variant-specific radiosensitivity rescue result, and especially a concordant ATM-specific feature plus radiosensitivity result, is not documented.
• The exact p.L480F construct/readout in one of the VCEP-approved assay papers has not been confirmed.

Sources

Reasoning

The ClinGen HBOP ATM VCEP v1.5 BP2 framework assigns negative points only for qualifying unaffected, non-A-T individuals carrying the assessed variant with a P/LP ATM variant in trans, or qualifying homozygosity, with laboratory/database-setting distinctions. The bundle provides no such individual, no co-occurring P/LP ATM variant, and no phase or age/absence-of-A-T documentation. The exact-variant report in PMID:19638463 documents only heterozygous p.L480F in a prostate brachytherapy cohort and does not document a second P/LP ATM variant or phase. Therefore BP2 cannot be assigned.

Rule

ClinGen HBOP ATM VCEP v1.5 PM3/BP2 points table; assign BP2 only from qualifying unaffected non-A-T individuals with a VCEP-classified P/LP ATM variant in trans or qualifying homozygosity, converting total negative points to Supporting (-1), Moderate (-2), or Strong (≤-4).

Evidence considered

• ClinGen HBOP ATM VCEP v1.5 assigns BP2 points for unaffected non-A-T individuals with a P/LP ATM variant in trans: -4.0 for confirmed trans, -2.0 for phase unknown, and -2.0 in a laboratory setting or -1.0 in a database setting for qualifying homozygosity (maximum -2.0 for homozygosity). BP2 strength is Supporting at -1, Moderate at -2, and Strong at ≤-4 points.
• The ATM VCEP instructions require an unaffected individual used for this BP2 path to be at least 18 years old with no evidence of A-T; these details are absent from the case bundle.
• PMID:19638463 reports a heterozygous ATM p.L480F observation but does not report a co-occurring P/LP ATM variant, phase, unaffected A-T status, or qualifying BP2 individual.

Still missing

• An unaffected non-A-T adult carrying NM_000051.4:c.1440A>C together with a P/LP ATM variant.
• Confirmation that the co-occurring ATM variant is P/LP under the ATM VCEP specification.
• Parental or other testing establishing the variants are in trans, or qualifying homozygosity with the required unaffected status and setting documentation.
• Age and clinical documentation showing the individual is at least 18 years old with no evidence of A-T.

Sources