rs6048 — F9 Factor IX Malmö

Factor IX Malmö — The X-Linked DVT Modifier with an Unknown Mechanism

Coagulation factor IX sits at the centre of the intrinsic pathway of blood clotting. When activated by factor XIa, it forms the tenase complex with factor VIIIa, which in turn activates factor X to trigger fibrin clot formation. Severe loss-of-function mutations in F9 cause haemophilia B (Christmas disease). Factor IX Malmö — rs6048, the common p.Thr194Ala missense variant¹¹ p.Thr194Ala missense variant
An amino acid change from threonine to alanine at position 194 of the canonical pre-pro-protein; also reported as Ala148Thr in older literature using mature-protein numbering that excludes the signal peptide and propeptide — is a far subtler story: a common polymorphism that modestly shifts clotting risk in the general population without causing haemophilia.

Because F9 is located on the X chromosome, males carry only one copy (hemizygous), while females can carry zero, one, or two copies of the protective G allele. Genotyping chips typically report hemizygous males as homozygous in the raw data, so a male reported as GG carries exactly one G allele, and a male reported as AA carries one A allele.

The Mechanism

At position 194 in the pre-pro-factor IX sequence, the common A allele encodes threonine; the protective G allele encodes alanine. This substitution is located in the EGF-1 domain of the factor IX protein, a region involved in calcium binding and factor VIIIa interaction. Despite this structurally interesting location, a key finding from the primary association study is that factor IX antigen levels and factor IX activation peptide levels — markers of factor IX production and activation — did not differ significantly between AA and GG carriers²² factor IX antigen levels and factor IX activation peptide levels — markers of factor IX production and activation — did not differ significantly between AA and GG carriers
Bezemer et al. 2009 measured both factor IX antigen and FIXa activation peptide in controls from the LETS (n=191) and MEGA (n=823+484) studies; no significant genotype effect. This means the mechanism of the G allele's protective effect is genuinely unknown — it is not simply producing less factor IX or generating less activated factor IX. Whether the alanine substitution subtly alters factor IX's interaction with factor VIIIa, phospholipid membranes, or inhibitory proteins such as antithrombin remains to be determined.

An adjacent intronic variant (rs422187) showed similar association to DVT in the Bezemer study, raising the possibility that rs6048 may partly or wholly tag a functional non-coding variant in linkage disequilibrium, rather than being the causal change itself.

The Evidence

The key association study is the 2009 analysis by Bezemer and colleagues³³ 2009 analysis by Bezemer and colleagues
Irene D Bezemer et al., "F9 Malmö, factor IX and deep vein thrombosis," Haematologica 2009;94(5):693–9. Combined LETS and MEGA case-control studies with 380+1,469 male cases and factor IX measurements in 191+823+484 controls, which combined two Dutch case-control studies (LETS and MEGA) to demonstrate that the G allele of rs6048 was associated with a 20% reduction in DVT odds: OR 0.80 (95% CI 0.69–0.93). This finding was subsequently replicated in large GWAS datasets. The Klarin 2019 genome-wide association study⁴⁴ genome-wide association study
Derek Klarin et al., "Genome-wide association analysis of venous thromboembolism identifies new risk loci and genetic overlap with arterial vascular disease," Nature Genetics 2019;51:1574–1579. Over 650,000 participants across the Million Veteran Program and UK Biobank identified the rs6048-A allele among 33 genomic loci associated with VTE risk (OR ~1.07). The cross-ancestry replication by Thibord et al. 2022⁵⁵ Thibord et al. 2022
Florian Thibord et al., "Cross-Ancestry Investigation of Venous Thromboembolism Genomic Predictors," Circulation 2022;146:1225–1242. 135 VTE loci identified across European, African, and Hispanic cohorts confirmed rs6048 across European, African, and Hispanic populations.

Importantly, a study of VTE recurrence in 2,185 patients⁶⁶ study of VTE recurrence in 2,185 patients
Roach et al. 2015, J Thromb Haemost 2015;13(10):1815–22. Four European cohorts: MEGA, Leiden Thrombophilia Study, Scottish Thrombophilia Study, Vienna Thrombosis Research Group found that Factor IX Malmö did not explain the observed sex difference in VTE recurrence risk. This suggests the variant's modest protective effect applies to first VTE events but does not substantially alter recurrence dynamics.

The effect size is modest — OR ~0.80 for DVT first event — and the clinical significance classification by the ClinGen Coagulation Factor Deficiency Variant Curation Expert Panel (4-star review, February 2024) is benign for haemophilia B and protective for DVT. This is not a variant that eliminates thrombosis risk; it shifts it modestly on a population scale.

Practical Implications

For carriers of the GG genotype (females with two protective alleles) or males with the G allele, the practical implication is limited: a modest downward adjustment in baseline DVT risk. This does not substitute for standard DVT prevention measures during high-risk situations (surgery, prolonged immobility, hormonal contraceptives), particularly if other prothrombotic variants are present (Factor V Leiden, prothrombin G20210A).

For AA homozygotes and AG heterozygotes, no specific intervention is warranted based on this variant alone, since the common A allele simply represents population-average coagulation factor IX function.

The G allele frequency varies dramatically by ancestry — approximately 30% in Europeans and Ashkenazi Jews, 18% in South Asians, 13% in Africans, and less than 0.2% in East Asians. This means the protective variant is essentially absent in East Asian populations, and population-level DVT risk attributable to this locus differs substantially by ancestry.

Interactions

The most clinically relevant interactions are with other coagulation pathway variants. Factor V Leiden (rs6025) and prothrombin G20210A (rs1799963) are the two most common inherited thrombophilias and act at different points in the coagulation cascade. If a carrier of rs6048-G also carries prothrombotic variants, the modest protective effect of the G allele would be expected to partially, but incompletely, offset the elevated risk. The magnitude of this offset has not been directly quantified in published studies.

The adjacent intronic SNP rs422187 is in strong linkage disequilibrium with rs6048 and showed comparable DVT association in the Bezemer cohort. These variants cannot be distinguished functionally with current evidence.