ERBB2c.3582delUncertain significance

Print report

Variant interpretation report

ERBB2 c.3582del, p.(Glu1195SerfsTer3)

NM_004448.3:c.3582del NP_004439.2:p.(Glu1195SerfsTer3) View identifiers

Classification

Uncertain significance

Variant description:

No variant summary was recorded for this case.

Five-step scale from benign to pathogenic. This variant is uncertain. Select a step to read what it means.

Uncertain.The recorded classification is uncertain significance.

In plain language

Gene-level background from the recorded interpretation.

About ERBB2

No gene-level summary was recorded for this case.

Key facts

Evidence used: 1 moderate (pathogenic) · 1 supporting (pathogenic)
Expert rules: Not recorded
Pipeline record: 28 Jul 2026 · version 7.0.4

Evidence considered

View criteria

Applied evidence criteria

PVS1 (moderate): NM_004448.3:c.3582del is a frameshift deletion in exon 27/27 (the last exon) of ERBB2, predicted to result in p.(Glu1195SerfsTer3) with loss of 58 C-terminal residues (1195-1255). The variant is located in the last exon and is not predicted to undergo nonsense-mediated decay. Under PMC6185798, for frameshift variants where NMD is not expected and the truncated region (<10% of the protein) does not remove a critical functional domain (the kinase domain, aa 720-987, remains intact), PVS1 is applied at moderate strength. Germline ERBB2 loss-of-function is supported as a disease mechanism by targeted literature review, though the supporting evidence is predominantly derived from related ERBB-family genes and somatic cancer contexts.
PM2 (supporting): This variant is absent from gnomAD v2.1, v4.1, and gnomAD-Canada, consistent with a rare pathogenic variant. Under generic ACMG/AMP, absence from population databases (allele frequency <0.1%) supports PM2 at supporting strength.

LYFE Sciences is an AI system, and it can make mistakes. Criteria may be applied incorrectly, sources may be misread, and a confident-looking classification can still be wrong. Double-check every criterion and its underlying evidence before relying on any call.

01The reasoning

How it was classified

The recorded framework is Not recorded. The criteria and strengths shown below come from the saved assessment; this page does not recalculate the classification.

Recorded applied criteria

The applied criteria and their strengths are shown as assessed under the stated framework.

Pathogenic evidence criteria: 2 Benign evidence criteria: 0

Pathogenic criteria applied · 2

PVS1Moderate

PVS1 · moderate

NM_004448.3:c.3582del is a frameshift deletion in exon 27/27 (the last exon) of ERBB2, predicted to result in p.(Glu1195SerfsTer3) with loss of 58 C-terminal residues (1195-1255). The variant is located in the last exon and is not predicted to undergo nonsense-mediated decay. Under PMC6185798, for frameshift variants where NMD is not expected and the truncated region (<10% of the protein) does not remove a critical functional domain (the kinase domain, aa 720-987, remains intact), PVS1 is applied at moderate strength. Germline ERBB2 loss-of-function is supported as a disease mechanism by targeted literature review, though the supporting evidence is predominantly derived from related ERBB-family genes and somatic cancer contexts.

PM2Supporting

PM2 · supporting

This variant is absent from gnomAD v2.1, v4.1, and gnomAD-Canada, consistent with a rare pathogenic variant. Under generic ACMG/AMP, absence from population databases (allele frequency <0.1%) supports PM2 at supporting strength.

Benign criteria applied · 0

No benign criterion marked as applied in the recorded assessment.

Recorded classification: Uncertain significance

gene-specific framework lacked a usable explicit final combination framework, so generic ACMG/AMP 2015 final-combination rules were applied as fallback; applied criteria: PVS1 moderate, PM2 supporting; combination = 1 moderate + 1 supporting, which maps to VUS.

02The sources

Evidence at a glance

Evidence summaries and captured source records, where available. Missing entries should not be interpreted as negative findings.

gnomAD v4.1

gnomAD v2.1

Population frequency

Not recorded

gnomAD v4.1 allele counts were not recorded for this variant.

gnomAD v2.1

Absent from gnomAD v2.1.

gnomAD-Canada

Absent from gnomAD-Canada v1.0.

Applied · PM2 →

gnomAD v4.1 ↗gnomAD v2.1 ↗gnomAD-Canada ↗

SpliceAI score

0.000 maximum delta

Recorded SpliceAI maximum delta score: 0.000. Refer to the recorded PP3/BP4 rationale for interpretation.

No listed criterion applied SpliceAI ↗

ClinVar submissions

Not recorded

A ClinVar laboratory submission summary was not recorded.

Context only ClinVar ↗

Functional evidence

Statement recorded

OncoKB identified curated literature and non-variant-specific oncogenicity context for review; listed oncogenicity label: Unknown Oncogenic Effect.

No listed criterion applied OncoKB ↗

COSMIC

Cancer Hotspots

Somatic evidence

No matching record reported

This variant has not previously been reported in somatic cancers (COSMIC).

Cancer Hotspots

This variant does not lie in a statistically significant hotspot.

Context only

COSMIC ↗Cancer Hotspots ↗

For specialists

Technical detail

ACMG/AMP criteria under Not recorded, with the data and sources behind each one.

28 criteria with recorded statuses: 2 met · 18 not met · 0 not assessed · 8 not applicable. Open any row for the full reasoning, the rule and the evidence behind it.

Pathogenic

PVS1 PS1 PS2 PS3 PS4 PM1 PM2 PM4 PM5 PM6 PP1 PP2 PP3 PP4 PP5 PS5

Benign

BA1 BS1 BS2 BS3 BS4 BP1 BP2 BP3 BP4 BP5 BP6 BP7

Met Not met Not assessed Not applicable

NM_004448.3:c.3582del (p.Glu1195SerfsTer3) is a frameshift deletion in exon 27 of ERBB2, the last exon, predicted to truncate 58 C-terminal residues without triggering nonsense-mediated decay.

Under the ClinGen SVI PVS1 framework (PMC6185798), the variant qualifies for PVS1 at moderate strength: a frameshift in the last exon where NMD is not predicted and the truncated region (<10% of the protein) does not contain a critical functional domain — the kinase domain remains intact.

The variant is absent from all population databases (gnomAD v2.1, v4.1, gnomAD-Canada), supporting PM2 at supporting strength.

No functional studies, case-control data, de novo observations, cosegregation data, or ClinVar classifications are available for this variant. Computational splicing prediction (SpliceAI max delta = 0.00) is neutral.

The combined evidence (PVS1_Moderate + PM2_Supporting) yields 1 moderate and 1 supporting criterion. Under generic ACMG/AMP 2015 combination rules, this does not meet the threshold for Likely Pathogenic (requires ≥1 moderate + ≥4 supporting, or ≥2 moderate + ≥2 supporting, among other combinations). The variant is classified as a Variant of Uncertain Significance (VUS).

Human review is recommended: ERBB2 is a proto-oncogene where C-terminal truncations have been associated with gain-of-function in somatic contexts, and the evidence for germline ERBB2 loss-of-function as a disease mechanism is limited.

Met · 2

PVS1 ModerateNM_004448.3:c.3582del is a frameshift deletion in exon 27/27 (the last exon) of ERBB2, predicted to result in p.(Glu1195SerfsTer3) with loss of 58 C-terminal residues (1195-1255). The variant is located in the last exon and is not predicted to undergo nonsense-mediated decay. Under PMC6185798, for frameshift variants where NMD is not expected and the truncated region (<10% of the protein) does not remove a critical functional domain (the kinase domain, aa 720-987, remains intact), PVS1 is applied at moderate strength. Germline ERBB2 loss-of-function is supported as a disease mechanism by targeted literature review, though the supporting evidence is predominantly derived from related ERBB-family genes and somatic cancer contexts. Details ▾Hide ▴

Evidence considered

Frameshift variant in last exon (27/27)
NMD not predicted
Truncation removes 58 C-terminal residues (~4.6% of protein)
kinase domain intact
PVS1_Moderate assigned per PMC6185798 downgrade for last-exon variants without NMD
Gene-level LoF mechanism supported by literature (PMC6185798 gate: eligible)

Still missing

Supporting literature for germline ERBB2 LoF disease mechanism is predominantly from ERBB3 and somatic cancer contexts
dedicated germline ERBB2 LoF syndrome evidence is limited
C-terminal truncations in ERBB2 have been associated with gain-of-function in somatic contexts
the disease mechanism (LoF vs GoF) warrants expert review

Sources

pvs1_generic_framework ↗pvs1_gene_contextpvs1_variant_assessmentgnomad_v2 ↗gnomad_v4 ↗

PM2 SupportingThis variant is absent from gnomAD v2.1, v4.1, and gnomAD-Canada, consistent with a rare pathogenic variant. Under generic ACMG/AMP, absence from population databases (allele frequency <0.1%) supports PM2 at supporting strength. Details ▾Hide ▴

Evidence considered

Absent from gnomAD v2.1 (exomes)
Absent from gnomAD v4.1 (exomes)
Absent from gnomAD-Canada v1.0

Sources

gnomad_v2 ↗gnomad_v4 ↗gnomad_canada ↗

Not met · 18 · evaluated, threshold not reached

PS2 No de novo occurrence data is available for this variant. Details ▾Hide ▴

Still missing

No parental testing data available

PS3 No functional studies testing NM_004448.3:c.3582del or a systematically characterized range that includes this position were identified in the literature. Details ▾Hide ▴

Still missing

No functional data available for this variant or a systematically characterized range encompassing it

PS4 No case-control data or statistical evidence of enrichment in affected individuals is available for this variant. Details ▾Hide ▴

Still missing

No case-control studies involving this variant

PM1 The variant does not lie within a statistically significant mutational hotspot as determined by cancerhotspots.org. The C-terminal tail of ERBB2, while containing regulatory phosphorylation sites, is not a well-established critical functional domain for PM1 purposes under generic ACMG/AMP — no domain-level pathogenic variant cluster has been demonstrated in this region. Details ▾Hide ▴

Evidence considered

cancerhotspots.org: residue not significant
No established mutational hotspot or domain-level pathogenic variant cluster at the C-terminal tail

PM6 No de novo occurrence data is available for this variant. Details ▾Hide ▴

Still missing

No de novo testing data available

PP1 No cosegregation data with disease is available for this variant. Details ▾Hide ▴

Still missing

No family cosegregation data available

PP3 SpliceAI predicts no splicing impact (max delta score = 0.00). REVEL and BayesDel are not applicable to non-SNV variants. No computational evidence supports a deleterious effect. Details ▾Hide ▴

Evidence considered

SpliceAI max delta = 0.00 (no predicted splice impact)
REVEL/BayesDel not applicable (not an SNV)

Sources

spliceai ↗

PP4 No patient-specific phenotype or family history data is available for assessment. Details ▾Hide ▴

Still missing

No clinical phenotype data provided

PP5 This variant is absent from ClinVar; no reputable source has reported it as pathogenic. Details ▾Hide ▴

Evidence considered

ClinVar: no variation IDs
no classifications
no submissions

Sources

clinvar ↗

BA1 The variant is absent from gnomAD v2.1, v4.1, and gnomAD-Canada. Population allele frequency is well below the 1% threshold for BA1. Details ▾Hide ▴

Evidence considered

Allele frequency: 0% (absent from all gnomAD datasets)

Sources

gnomad_v2 ↗gnomad_v4 ↗

BS1 The variant is absent from gnomAD. Population allele frequency is well below the 0.3% threshold for BS1. Details ▾Hide ▴

Evidence considered

Allele frequency: 0% (absent from all gnomAD datasets)

Sources

gnomad_v2 ↗gnomad_v4 ↗

BS2 Not observed in a healthy adult homozygous or hemizygous state. No such observations are reported in population databases. Details ▾Hide ▴

Evidence considered

gnomAD homozygote count: 0

Sources

gnomad_v2 ↗gnomad_v4 ↗

BS3 No well-established functional studies demonstrating a benign effect for this variant are available. Details ▾Hide ▴

Still missing

No functional data available

BS4 No cosegregation data demonstrating absence of segregation with disease is available. Details ▾Hide ▴

Still missing

No family cosegregation data available

BP2 No evidence of this variant occurring in trans with a known pathogenic variant in a recessive disorder. Details ▾Hide ▴

Still missing

No phase data available

BP4 SpliceAI predicts no splicing impact (max delta = 0.00), but NM_004448.3:c.3582del is a frameshift variant with a clear predicted protein-truncating effect (p.Glu1195SerfsTer3). Computational evidence does not support a benign interpretation — a frameshift with premature termination is inherently predicted to be damaging. Details ▾Hide ▴

Evidence considered

SpliceAI max delta = 0.00 (splicing neutral
but not informative for a frameshift variant)

Sources

spliceai ↗

BP5 No alternative molecular cause has been identified in a case harboring this variant. Details ▾Hide ▴

Still missing

No clinical case data available

BP6 This variant is absent from ClinVar; no reputable source has reported it as benign. Details ▾Hide ▴

Evidence considered

ClinVar: no variation IDs
no classifications
no submissions

Sources

clinvar ↗

Not applicable · 8· recorded as not applicable Show reasons ▾Hide reasons ▴

PS1 N/APS1 applies when the same amino acid change has been independently established as pathogenic. This variant is a frameshift deletion, not a missense variant. Details ▾Hide ▴

No further detail was recorded.

PM4 N/APM4 applies to in-frame deletions/insertions in non-repeat regions and stop-loss variants. NM_004448.3:c.3582del is a frameshift (out-of-frame) deletion, which is evaluated under PVS1 rather than PM4. Details ▾Hide ▴

No further detail was recorded.

PM5 N/APM5 requires a same-residue missense comparator with established pathogenicity. This is a frameshift variant; no same-residue missense comparator candidates were identified. The automated PM5 candidate harvest returned no eligible comparators. Details ▾Hide ▴

Evidence considered

pm5_mode: unknown
eligible_for_classic_pm5_search: false
No same-residue candidate variants found in ClinVar

Sources

pm5_candidates

PP2 N/APP2 applies to missense variants in genes with a low rate of benign missense variation. NM_004448.3:c.3582del is a frameshift deletion, not a missense variant. Details ▾Hide ▴

No further detail was recorded.

BP1 N/ABP1 applies to missense variants in genes where only truncating variants cause disease. NM_004448.3:c.3582del is itself a truncating (frameshift) variant, not a missense. Details ▾Hide ▴

No further detail was recorded.

BP3 N/ABP3 applies to in-frame deletions/insertions in repetitive regions. NM_004448.3:c.3582del is an out-of-frame (frameshift) deletion. Details ▾Hide ▴

No further detail was recorded.

BP7 N/ABP7 applies to synonymous variants without predicted splicing impact. NM_004448.3:c.3582del is a frameshift deletion, not a synonymous variant. Details ▾Hide ▴

No further detail was recorded.

PS5 N/APS5 is not part of the standard ACMG/AMP 2015 criteria set. Details ▾Hide ▴

No further detail was recorded.

Identifiers

Gene: ERBB2
Transcript: NM_004448.3
Coding (HGVS c.): NM_004448.3:c.3582del
Protein (HGVS p.): NP_004439.2:p.(Glu1195SerfsTer3)
GRCh38: chr17:39727854 AC>A
GRCh37: chr17:37884107 AC>A

Gene diagram

Every interpreted variant in ERBB2, mapped onto the exon structure of NM_004448.3. Select a row to find it on the plot.

ERBB2NM_004448.3

Fetching transcript structure from UCSC…

Exons

—

Transcript span

—

Strand

—

Variants mapped

—

Source

UCSC ncbiRefSeqCurated

All variants in ERBB2 — Select a row to find it on the plot; use the link to open its page.

	Protein	Location	Classification	Link

gnomAD v4.1 Allele count data not recorded

gnomAD v2.1 Allele count data not recorded

gnomAD-Canada v1.0 Allele count data not recorded

Statements

gnomAD v4.1

Absent from gnomAD v4.1.

gnomAD v2.1

Absent from gnomAD v2.1.

gnomAD-Canada

Absent from gnomAD-Canada v1.0.

SpliceAI and Pangolin

Pangolin splice lossmagnitude shown

-0.050

Pangolin splice gain

0.050

Other predictors

REVEL: Not available
BayesDel: Not available

How the predictions were used

Evidence considered

SpliceAI max delta = 0.00 (no predicted splice impact)
REVEL/BayesDel not applicable (not an SNV)

Sources

spliceai ↗

Evidence considered

SpliceAI max delta = 0.00 (splicing neutral
but not informative for a frameshift variant)

Sources

spliceai ↗

BP7 N/ABP7 applies to synonymous variants without predicted splicing impact. NM_004448.3:c.3582del is a frameshift deletion, not a synonymous variant. Details ▾Hide ▴

No further detail was recorded.

Statement

In silico

SpliceAI predicts no significant splice impact for this variant (max delta score = 0.00).

0 paper records listed. No publication is explicitly linked to an applied criterion in the recorded metadata.

Screenshots represent source records captured during evidence collection. Select one to view the recorded source content.

gnomAD v4.1 Captured gnomAD v4.1 source record Open source ↗

gnomAD v2.1 Captured gnomAD v2.1 source record Open source ↗

SpliceAI Lookup Recorded maximum delta 0.000 Open source ↗

ClinVar Captured ClinVar source record Open source ↗

COSMIC Captured COSMIC source record Open source ↗

Cancer hotspots Captured Cancer Hotspots source record Open source ↗

OncoKB Captured OncoKB source record Open source ↗

All source links

Statements for reports

The evidence statements recorded by the interpretation pipeline. They have not been reinterpreted by the page.

gnomAD v4.1

Absent from gnomAD v4.1.

gnomAD v2.1

Absent from gnomAD v2.1.

gnomAD-Canada

Absent from gnomAD-Canada v1.0.

ClinVar

This variant is absent from ClinVar.

In silico

SpliceAI predicts no significant splice impact for this variant (max delta score = 0.00).

Functional

OncoKB identified curated literature and non-variant-specific oncogenicity context for review; listed oncogenicity label: Unknown Oncogenic Effect.

Somatic

This variant does not lie in a statistically significant hotspot. This variant has not previously been reported in somatic cancers (COSMIC).

Cancer Hotspots

This variant does not lie in a statistically significant hotspot.