Variant classification summary

NM_000051.4:c.5178-28T>A

ATM · NP_000042.3:p.? · NM_000051.4

GRCh37: chr11:108172347 T>A · GRCh38: chr11:108301620 T>A

Gene: ATM Transcript: NM_000051.4

Final call

VUS

PM2 supporting PP3 supporting

Variant details

Gene

ATM

Transcript

NM_000051.4

Protein

NP_000042.3:p.?

gnomAD AF

ClinVar

OncoKB

Classification rationale

Interpretation summary

Generated evidence synthesis

1

c.5178-28T>A is a rare deep intronic variant in ATM, located 28 bp upstream of the exon 35 acceptor splice site, that is absent from all queried population databases including gnomAD v2.1, v4.1, and gnomAD-Canada (>800,000 alleles screened).

2

SpliceAI in silico analysis predicts a potential splicing impact with a max delta score of 0.27 (DS_AL=0.27, DS_DL=0.26), exceeding the ATM VCEP 1.5.0 threshold of ≥0.2, consistent with the possible creation of an alternative splice acceptor or donor site.

3

The variant has not been reported in ClinVar and no functional studies (RNA splicing assays, ATM kinase activity, or radiosensitivity testing) have been published; thus no functional confirmation of the predicted splicing effect is available.

4

No A-T probands carrying this variant in trans with a pathogenic ATM variant have been reported, leaving the PM3 criterion unevaluable. Similarly, no cosegregation data (PP1) or case-control enrichment data (PS4) exist for this variant.

5

Applying the ATM VCEP 1.5.0 framework, two supporting-level pathogenic criteria are met: PM2_Supporting (absence from population databases) and PP3 (SpliceAI splicing prediction). No benign criteria are met. Under the ACMG/AMP 2015 combining rules adopted by the ATM VCEP, two supporting pathogenic criteria without any moderate, strong, or very strong criteria are insufficient to reach Likely Pathogenic. The variant is classified as a Variant of Uncertain Significance (VUS).

6

Resolution of this VUS would require: (1) RNA functional studies (RT-PCR or minigene assay) to confirm or exclude aberrant splicing, which could upgrade PVS1_Strength(RNA) or support BP7_RNA; (2) identification of the variant in trans with a pathogenic ATM variant in an A-T proband for PM3; and (3) reporting of this variant to ClinVar with clinical phenotype data.

Final determination: No criteria-combination rule matched the adjudicated criteria in the Richards et.al., 2015 - Combining rules v1.5.0 framework, so the variant remains a Variant of Uncertain Significance pending human review.

Criteria assessment

ACMG/AMP criteria review

Criteria shown when status is available

All criteria require review: For research and educational purposes only.

Criterion	Status	Rationale	Evidence used
PVS1	Not met	c.5178-28T>A is an intronic variant located 28 bp upstream of the exon 35 acceptor splice site, outside the canonical ±1,2 splice dinucleotide. Per the ATM VCEP 1.5.0 PS1 splicing table, intronic variants outside the donor/acceptor ±1,2 dinucleotide use PP3 as the baseline computational code, not PVS1. PVS1 (predicted splice defect) is reserved for variants at donor/acceptor ±1,2 positions in the ATM VCEP framework. No RNA-level confirmation of aberrant splicing is available to support PVS1_Strength(RNA).	spliceai cspec pvs1_variant_assessment
PS1	Not assessed	No known pathogenic or likely pathogenic ATM variant has been reported at the identical nucleotide position (c.5178-28) or within the same donor/acceptor motif with identical predicted splicing outcome. The ATM VCEP PS1 Splicing table requires a comparator reference variant classified as P/LP by VCEP specifications with matching predicted splice event. Without such a comparator, PS1 cannot be evaluated.	cspec clinvar
PS2	N/A	The ATM VCEP 1.5.0 explicitly states PS2 is not applicable for autosomal dominant or autosomal recessive disease: 'Do not use for AD or AR disease: Informative de novo occurrences have not yet been observed and de novo AR conditions are unlikely to be informed by phase.'	cspec
PS3	Not assessed	No variant-specific functional studies assessing ATM kinase activity, phosphorylation of ATM-specific targets (e.g., KAP1, p53), or radiosensitivity have been published for c.5178-28T>A. The ATM VCEP 1.5.0 requires functional assays that evaluate both an ATM-specific feature and radiosensitivity (PS3_Moderate) or an ATM-specific feature alone (PS3_Supporting). No such data exists for this deep intronic variant.
PS4	Not met	No case-control studies comparing the frequency of c.5178-28T>A in affected individuals versus controls have been performed. The ATM VCEP 1.5.0 requires a case-control study with p≤0.05 AND (odds ratio ≥2 OR lower 95% CI ≥1.5). The variant's extreme rarity (absent from gnomAD v2.1 and v4.1 with >800,000 alleles) precludes meaningful enrichment analysis without a very large disease-specific cohort. PS4_Moderate (proband counting) is explicitly not used per VCEP guidance.	gnomad_v2 gnomad_v4 cspec
PS5	Not assessed	PS5 is a placeholder for the ACMG/AMP 2015 framework. No evidence was identified or evaluated for this criterion. The variant type (intronic substitution) is not amenable to the typical applications of PS5.
PM1	N/A	The ATM VCEP 1.5.0 explicitly states PM1 is not applicable: 'Do not use: Benign and pathogenic variants are known to occur within the same domains and germline mutational hotspots are not well defined at this time.'	cspec
PM2	Met	c.5178-28T>A is absent from all queried population databases: gnomAD v2.1 (exomes), gnomAD v4.1 (genomes+exomes), and gnomAD-Canada v1.0 (HostSeq genomes). Per the ATM VCEP 1.5.0, PM2_Supporting applies when the variant frequency is ≤0.001% in the gnomAD v4 dataset. Complete absence in >800,000 alleles satisfies this threshold.	gnomad_v2 gnomad_v4 gnomad_canada cspec
PM3	Not assessed	No observations of c.5178-28T>A occurring in trans with a known pathogenic or likely pathogenic ATM variant in an individual with Ataxia Telangiectasia (A-T) have been reported. The ATM VCEP 1.5.0 PM3/BP2 table assigns points per unrelated A-T proband based on zygosity and phenotype confidence (confirmed in trans: 2.0–4.0 points; phase unknown: 1.0–2.0 points). No proband data is available for this variant in ClinVar, LOVD, or the published literature.	cspec clinvar
PM4	N/A	The ATM VCEP 1.5.0 restricts PM4 to stop-loss variants only: 'Do not use for in-frame insertions or deletions less than a single exon; Use for stop-loss variants, only.' c.5178-28T>A is a single-nucleotide intronic substitution, not a stop-loss variant.	cspec
PM5	N/A	The ATM VCEP 1.5.0 restricts PM5_Supporting to frameshifting/truncating variants with premature termination codons upstream of p.Arg3047, or to splice variants where PVS1_VS(RNA) is applied based on high-quality observed splicing impact that is NMD-prone. This variant is an intronic substitution with no confirmed truncating effect and no RNA-level splicing data. The pm5_candidates.json confirms the PM5 mode is 'truncation_cutoff_gene_specific' with no eligible comparators.	cspec pm5_candidates
PM6	N/A	The ATM VCEP 1.5.0 explicitly states PM6 is not applicable for autosomal dominant or autosomal recessive disease: 'Do not use for AD or AR disease: Informative de novo occurrences have not yet been observed and de novo AR conditions are unlikely to be informed by phase.'	cspec
PP1	Not assessed	No family cosegregation data are available for c.5178-28T>A. The ATM VCEP 1.5.0 applies PP1 only to the autosomal recessive (A-T) condition, requiring affected relatives to carry both variants identified in the proband. For the autosomal dominant (cancer predisposition) condition, PP1 is explicitly not used because cosegregation analysis in low-penetrance genes can produce false positive results (PMID 32773770). No published pedigrees include this variant.	cspec
PP2	N/A	The ATM VCEP 1.5.0 explicitly states PP2 is not applicable: 'Do not use: ATM does not have a defined low rate of missense benign variation.'	cspec
PP3	Met	SpliceAI predicts a splicing impact for this intronic variant with a max delta score of 0.27 (DS_AL=0.27, DS_DL=0.26), exceeding the ATM VCEP 1.5.0 threshold of ≥0.2 for PP3. This criterion applies to intronic variants outside the donor and acceptor ±1,2 sites. The SpliceAI prediction suggests the variant may create an alternative splice acceptor or donor site. Per the VCEP, PP3 for splice predictions is applied at the Supporting strength level only.	spliceai cspec
PP4	N/A	The ATM VCEP 1.5.0 explicitly states PP4 is not applicable. For the autosomal dominant condition, breast cancer has multiple genetic etiologies and no features distinguish hereditary from sporadic causes. For the autosomal recessive condition, such evidence is built into the PM3/BP2 table.	cspec
PP5	N/A	The ATM VCEP 1.5.0 explicitly states PP5 is not for use as recommended by the ClinGen Sequence Variant Interpretation VCEP Review Committee. Additionally, the variant is absent from ClinVar, so no reputable source has classified it as pathogenic.	cspec clinvar
BA1	Not met	c.5178-28T>A is absent from gnomAD v4.1 (0 alleles across all populations). The ATM VCEP 1.5.0 BA1 threshold requires a Grpmax Filtering AF >0.5% in the gnomAD v4 dataset. Complete absence fails this threshold by a wide margin.	gnomad_v4 cspec
BS1	Not met	c.5178-28T>A is absent from gnomAD v4.1. The ATM VCEP 1.5.0 BS1 threshold requires a Grpmax Filtering AF >0.05% in the gnomAD v4 dataset. Complete absence fails this threshold.	gnomad_v4 cspec
BS2	N/A	The ATM VCEP 1.5.0 explicitly states BS2 is not applicable: 'Do not use: ATM has incomplete penetrance.'	cspec
BS3	Not assessed	No well-established functional studies demonstrating that c.5178-28T>A has no damaging effect on ATM protein function or splicing have been published. The ATM VCEP 1.5.0 requires either rescue of both an ATM-specific feature (e.g., phosphorylation of ATM targets) AND radiosensitivity (BS3_Moderate), or rescue of either (BS3_Supporting). No functional data of any kind exists for this variant.
BS4	N/A	The ATM VCEP 1.5.0 explicitly states BS4 is not applicable. For the autosomal dominant condition, cosegregation analysis in low-penetrance genes can lead to false positive results (PMID 32773770). For the autosomal recessive condition, informative instances of lack of cosegregation in A-T families are too rare to consider at this time.	cspec
BP1	N/A	The ATM VCEP 1.5.0 explicitly states BP1 is not applicable: 'Do not use: Missense pathogenic variants are known for ATM.'	cspec
BP2	Not assessed	No phased genotype data have been reported showing c.5178-28T>A occurring in cis (on the same allele) with a known pathogenic or likely pathogenic ATM variant in an unaffected individual ≥18 years old. The ATM VCEP 1.5.0 BP2 table assigns negative points based on zygosity and phenotype context (confirmed in trans in an unaffected individual: -4.0 points; homozygous in a database setting: -1.0 point). No such observations exist.	cspec clinvar
BP3	N/A	The ATM VCEP 1.5.0 explicitly states BP3 is not applicable: 'Do not use.' This criterion concerns in-frame deletions/insertions in repetitive regions, which does not apply to this single-nucleotide intronic substitution.	cspec
BP4	Not met	SpliceAI predicts a splicing impact for this variant with a max delta score of 0.27, which exceeds the ATM VCEP 1.5.0 BP4 threshold of ≤0.1 for 'no predicted impact via splicing.' The positive splice prediction precludes application of BP4. REVEL is not applicable as this is not a missense variant.	spliceai cspec
BP5	N/A	The ATM VCEP 1.5.0 explicitly states BP5 is not applicable: 'Do not use: Cases with multiple pathogenic variants have been observed with no noticeable difference in phenotype (e.g. BRCA1 and BRCA2). In addition, ATM has low penetrance and will naturally occur with other pathogenic variants more frequently due to higher tolerance/presence in the general population.'	cspec
BP6	N/A	The ATM VCEP 1.5.0 explicitly states BP6 is not for use as recommended by the ClinGen Sequence Variant Interpretation VCEP Review Committee. Additionally, the variant is absent from ClinVar, so no external benign classification exists.	cspec clinvar
BP7	Not met	Although c.5178-28T>A is positioned at acceptor -28, which qualifies as 'deep intronic' (>21 bp from the acceptor site) per the ATM VCEP 1.5.0 BP7 definition, SpliceAI predicts a splicing impact with a max delta score of 0.27. BP7 is intended for deep intronic variants without predicted splice effects. The positive SpliceAI prediction argues against a benign interpretation via BP7. No RNA data are available to apply BP7_Strong(RNA) or BP7_Moderate(RNA) (observed lack of aberrant splicing).	spliceai cspec

Disclaimer: The content and results provided by LYFE Sciences are for research and educational purposes only and must not be used as a substitute for professional medical judgment, diagnosis, or treatment. Always consult a qualified healthcare professional before making any clinical decisions.