Variant classification summary

NM_004972.3:c.1849G>T

JAK2 · NP_004963.1:p.(Val617Phe) · NM_004972.3

GRCh37: chr9:5073770 G>T · GRCh38: chr9:5073770 G>T

Gene: JAK2 Transcript: NM_004972.3

Final call

Pathogenic

PS3 strong PS4 strong PM1 moderate PM2 supporting PP3 supporting PP5 supporting

Variant details

Gene

JAK2

Transcript

NM_004972.3

Protein

NP_004963.1:p.(Val617Phe)

gnomAD AF

0.0002932901928990148 (v4.1)

ClinVar

Pathogenic

OncoKB

Oncogenic

Classification rationale

Interpretation summary

Generated evidence synthesis

1

JAK2 c.1849G>T (p.Val617Phe) is the canonical activating mutation in myeloproliferative neoplasms, identified in >80% of polycythemia vera cases and a substantial proportion of essential thrombocythemia and primary myelofibrosis cases.

2

Well-established functional studies across multiple independent laboratories demonstrate that V617F disrupts JH2 pseudokinase domain autoinhibition, resulting in constitutive JAK2 kinase activity, STAT-mediated transcriptional activation, cytokine-independent cell growth, and erythrocytosis in murine models.

3

The variant is located in the JH2 pseudokinase domain, a critical regulatory region and established mutational hotspot for gain-of-function variants in myeloproliferative neoplasms.

4

Population databases show the variant at very low frequency (gnomAD v2.1: 0.034%; v4.1: 0.029%), well below the 0.1% PM2 threshold. Observed homozygotes are consistent with somatic clonal hematopoiesis rather than germline events.

5

REVEL in silico prediction score of 0.881 supports a deleterious effect on protein function.

6

ClinVar reports the variant as Pathogenic by 14 clinical diagnostic laboratories and Likely pathogenic by 5, with variation ID 14662.

7

Two strong pathogenic criteria (PS3 + PS4) satisfy the '2 Strong' pathogenic combination rule under the generic ACMG/AMP 2015 framework. Additional moderate (PM1) and supporting criteria (PM2, PP3, PP5) provide further confirmatory evidence.

Final determination: Generic ACMG/AMP 2015 fallback rules support a Pathogenic classification based on the observed combination of very strong, strong, moderate, and supporting pathogenic criteria.

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	N/A	NM_004972.3:c.1849G>T is a missense substitution (p.Val617Phe), not a null variant (nonsense, frameshift, or canonical splice). The variant does not fall into the default generic PVS1 null-variant buckets defined by ClinGen SVI PVS1 recommendations (PMC6185798).	pvs1_generic_framework
PS1	Not assessed	No alternate nucleotide change at c.1849 resulting in the same amino acid change (p.Val617Phe) has been identified in the case data. Only the G>T transversion is documented.
PS2	Not met	No de novo observation with confirmed maternity and paternity is documented for this variant in any source. The literature_pass flags PMID:25741868 for PS2 but this is the ACMG/AMP framework guideline paper, not a de novo case report.
PS3	Met	Well-established functional studies demonstrate JAK2 V617F is a gain-of-function mutation. James et al. (2005) showed that V617F induces constitutive STAT5 transcriptional activation in luciferase reporter assays, transforms Ba/F3 cells to cytokine-independent growth, and causes erythrocytosis in a mouse bone marrow transplant model. Haan et al. (2009) confirmed constitutive activation of STAT1/3/5, ERK1/2, and p38 pathways, with SOCS-mediated feedback regulation. Sanz et al. (2011) demonstrated altered catalytic activity via peptide microarray. McKenney et al. (2018) confirmed oncogenicity in a Jak2V617F-driven MPN mouse model. Multiple independent studies with diverse functional assays consistently support a gain-of-function pathogenic mechanism.	PMID:15793561 PMID:19543316 PMID:21533163 PMID:29355841
PS4	Met	JAK2 V617F is the hallmark mutation in myeloproliferative neoplasms, reported in >80% of polycythemia vera, ~50% of essential thrombocythemia, and ~50% of primary myelofibrosis cases. It is virtually absent from the general germline population — gnomAD v2.1 shows global AF of 0.034% (97/281,626 alleles) with 5 homozygotes, and v4.1 shows 0.029% (471/1,605,918 alleles) with 17 homozygotes, frequencies that almost certainly represent somatic clonal hematopoiesis rather than true germline variation. The extreme case enrichment meets the PS4 threshold at the strong level.	gnomad_v2 gnomad_v4 clinvar PMID:15793561
PS5	Not assessed	While ClinVar reports a 'criteria provided, single submitter' classification of Likely pathogenic from Labcorp/Invitae, the full ACMG criteria trail from that submission is not available for independent evaluation. PP5 is applied instead to capture the ClinVar pathogenic consensus.
PM1	Met	The p.Val617Phe substitution is located in the JH2 pseudokinase domain (amino acids ~545-809), a well-established critical regulatory domain that mediates autoinhibition of JAK2 kinase activity. The JH2 domain is a recognized mutational hotspot for gain-of-function variants in myeloproliferative neoplasms, with multiple activating mutations described at this locus (e.g., K539L, R564Q, E695K). Position 617 is specifically identified as a residue whose alteration disrupts autoinhibitory interactions with the JH1 kinase domain activation loop.	PMID:15793561 PMID:19543316 PMID:21533163
PM2	Met	The variant is present at very low frequency in gnomAD population databases: global AF 0.034% in v2.1 and 0.029% in v4.1, both below the 0.1% threshold for PM2. The maximum subpopulation frequency is 0.075% (Ashkenazi Jewish, v4.1). The presence of homozygotes (5 in v2.1, 17 in v4.1) is noted but is consistent with somatic clonal hematopoiesis in aging individuals rather than true germline homozygosity.	gnomad_v2 gnomad_v4
PM5	N/A	PM5 candidate harvesting did not identify valid same-residue comparator variants. The pm5_candidates.json indicates eligibility for classic PM5 search could not be confirmed.
PM6	N/A	No de novo observation data available. PM6 requires a de novo observation with confirmed maternity and paternity.
PP1	Not assessed	No segregation data available. PP1 requires co-segregation with disease in multiple affected family members.
PP2	Not assessed	Gene-level constraint metrics (Z-score, pLI, missense depletion) for JAK2 are not evaluated in the case materials. PP2 requires a low rate of benign missense variation with missense variants as a common disease mechanism. While JAK2 gain-of-function missense variants are a well-known disease mechanism, formal constraint metric assessment is needed.
PP3	Met	REVEL score of 0.881 is above the established deleterious threshold (≥0.75). Multiple in silico predictors support a deleterious effect. SpliceAI predicts no splicing impact (max delta = 0.01). HCI Prior is not available for JAK2.	revel spliceai
PP4	Not assessed	Patient phenotype and family history data are not provided in the case materials. PP4 requires the patient's phenotype or family history to be highly specific for a disease with a single genetic etiology.
PP5	Met	ClinVar reports the variant as Pathogenic by 14 clinical laboratories and Likely pathogenic by 5 laboratories (ClinVar variation ID 14662). Multiple reputable clinical diagnostic laboratories have independently classified this variant as pathogenic, supporting a disease-causing role.	clinvar
BA1	Not met	gnomAD global allele frequency is 0.034% (v2.1) and 0.029% (v4.1), far below the 1% threshold for BA1. Maximum subpopulation frequency is 0.075% (Ashkenazi Jewish, v4.1).	gnomad_v2 gnomad_v4
BS1	Not met	gnomAD global allele frequency is 0.034% (v2.1) and 0.029% (v4.1), below the 0.3% threshold for BS1. Maximum subpopulation frequency is 0.075% (Ashkenazi Jewish, v4.1), also below the threshold. The variant is not observed at a frequency greater than expected for the disorder.	gnomad_v2 gnomad_v4
BS2	Not assessed	BS2 requires observation of the variant in healthy adults with full penetrance expected at an early age. No such data are available in the case materials.
BS3	N/A	Functional evidence consistently demonstrates a gain-of-function pathogenic effect. BS3 requires well-established functional studies showing no deleterious effect — the opposite of what is observed.
BS4	Not assessed	No segregation data available. BS4 requires lack of segregation in affected family members.
BP1	N/A	JAK2 disease mechanism is predominantly gain-of-function via activating missense variants, not loss-of-function via truncating variants. BP1 applies when a missense variant occurs in a gene where primarily truncating variants cause disease.
BP2	Not assessed	No phase data available. BP2 requires observation in trans with a pathogenic variant for a fully penetrant dominant disorder, or in cis with a pathogenic variant for recessive disorders.
BP4	Not met	REVEL score of 0.881 predicts a deleterious effect, contradicting the BP4 requirement that multiple lines of computational evidence suggest no impact on gene product. SpliceAI also shows no splicing impact, but the primary in silico metric (REVEL) strongly supports pathogenicity.	revel
BP5	Not assessed	No data available regarding an alternate molecular basis for disease in a case harboring this variant. BP5 requires the variant to be found in a case with an alternate molecular basis for disease.
BP6	N/A	No reputable source has classified this variant as benign. ClinVar classifications are exclusively pathogenic or likely pathogenic.
BP7	N/A	NM_004972.3:c.1849G>T is a missense variant (p.Val617Phe), not a synonymous variant. BP7 applies only to synonymous variants with no predicted splice impact.
BP3	N/A	Skipped per instruction.
PM3	N/A	Skipped per instruction.
PM4	N/A	Skipped per instruction.

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.