rs2672598 — HTRA1

Promoter variant at -487 in HTRA1 that elevates HTRA1 serine protease expression, disrupting TGF-β signaling and extracellular matrix remodeling in the retina and increasing risk of neovascular (wet) age-related macular degeneration

See your personal result for HTRA1

Upload your DNA data to find out which genotype you carry and what it means for you.

Upload your DNA data

Works with 23andMe, AncestryDNA, and other DNA test exports. Results in under 60 seconds.

HTRA1 — The Retinal Protease Variant That Drives Wet AMD

The 10q26 chromosomal locus is the second strongest genetic risk factor for age-related macular degeneration (AMD), second only to the complement factor H (CFH) gene on chromosome 1. Within the 10q26 locus sit two genes in strong linkage disequilibrium: ARMS2 (also called LOC387715) and HTRA1. rs2672598 is a promoter variant in HTRA1 — specifically a T-to-C change at position −487 relative to the transcription start site — that fundamentally alters how much HTRA1 serine protease the retina produces.

HTRA1 encodes a secreted serine protease that degrades extracellular matrix components and inhibits TGF-β signaling11 TGF-β signaling
transforming growth factor beta, which controls blood vessel growth and tissue remodeling. In the eye, excess HTRA1 activity disrupts Bruch's membrane integrity and promotes the abnormal choroidal neovascularization that defines wet AMD — the rapidly vision-threatening form of the disease.

The Mechanism

The C allele at rs2672598 alters transcription factor binding at the HTRA1 promoter. The wild-type T allele creates binding sites for STAT4, NFκB, c-Ets-1, RelA, Elk-1, and WT1. The C allele replaces these with binding sites for GR-alpha, AP-2αA, and Sp1 — a different set of transcription factors that drive substantially higher HTRA1 transcriptional output. Luciferase reporter assays showed C-containing haplotypes had 1.78–1.99 fold higher promoter activity than wild-type T-containing haplotypes.

At the protein level, this translates directly to higher HTRA1 enzyme in the eye: vitreous humor HTRA1 protein in CC homozygotes was 2.56-fold higher than in TT homozygotes (P=0.02)22 vitreous humor HTRA1 protein in CC homozygotes was 2.56-fold higher than in TT homozygotes (P=0.02)
Nakanishi et al. HTRA1 promoter variant differentiates polypoidal choroidal vasculopathy from exudative AMD. Sci Rep. 2016. This excess protease degrades extracellular matrix proteins including fibronectin and vitronectin33 vitronectin
a glycoprotein that stabilizes Bruch's membrane and regulates complement activation in the subretinal space, weakening the barrier that separates the retinal pigment epithelium from the choroidal vasculature. Excess HTRA1 also cleaves LTBP-1, releasing TGF-β from its latent complex and amplifying pro-angiogenic signaling — the molecular basis of choroidal neovascularization.

rs2672598 is in strong linkage disequilibrium with rs11200638 (−625G>A in the same promoter), the most-studied HTRA1 AMD variant. Both variants are part of a risk haplotype that spans the HTRA1 promoter and exon 1 (ACCTT haplotype, P=6.68×10⁻¹⁴ for AMD association).

The Evidence

The most comprehensive single-study data for rs2672598 comes from Tam et al. 200844 Tam et al. 2008
HTRA1 variants in exudative age-related macular degeneration and interactions with smoking and CFH. Invest Ophthalmol Vis Sci. 2008, which genotyped 163 AMD cases and 183 controls from a Hong Kong Chinese population. The CC genotype conferred a 25.6-fold increased AMD risk compared to TT (OR=25.61, 95% CI: 3.36–195.05, P=3.03×10⁻¹⁰). The CT heterozygous genotype showed intermediate risk. Population-attributable risk for the HTRA1 haplotype alone was 53.1%.

Meta-analyses of the more-studied rs11200638 variant (in near-perfect LD with rs2672598) yield consistent effect sizes across all populations: AA homozygotes have 7.5–8.5-fold increased AMD risk versus GG controls; AG heterozygotes have 2.1–2.2-fold increased risk. A 2020 comprehensive meta-analysis of 15,316 subjects55 2020 comprehensive meta-analysis of 15,316 subjects
HTRA1 rs11200638 variant and AMD risk: comprehensive analysis. BMC Ophthalmol. 2020 confirmed these effects across European and Asian populations.

The association is specific to neovascular (wet) AMD — not dry AMD or geographic atrophy in isolation. Within wet AMD subtypes, Nakanishi et al. 201666 Nakanishi et al. 2016
HTRA1 promoter variant differentiates polypoidal choroidal vasculopathy from exudative AMD. Sci Rep. 2016 demonstrated that rs2672598 CC specifically predicts exudative AMD (OR=10.53, P=1.31×10⁻⁴) much more strongly than polypoidal choroidal vasculopathy (OR=1.94 in meta-analysis, P=0.02), distinguishing it from other ARMS2 region variants.

Practical Actions

The CC genotype represents one of the highest monogenic risk increases for any common late-onset disease — a more-than-25-fold elevation in wet AMD risk. The most important actions are early detection and monitoring. Anti-VEGF therapy (ranibizumab, bevacizumab, aflibercept) is highly effective for wet AMD when started before significant vision loss — making the time between first symptoms and treatment the critical window.

Smoking is an independent multiplicative risk factor: HTRA1 risk allele carriers who smoke have compounded risk (OR ~15.7 for the combination vs baseline). Smoking cessation is therefore specifically relevant — not generically healthy advice, but a direct interaction with the HTRA1 pathway.

Lutein, zeaxanthin, and omega-3 fatty acids have been studied in the AREDS2 trial for AMD prevention and progression reduction. These are particularly relevant for high-HTRA1-risk genotypes because they protect the macular photoreceptors and retinal pigment epithelium that HTRA1-mediated matrix degradation makes vulnerable.

Interactions

rs2672598 is in strong LD with rs11200638 (HTRA1 promoter −625G>A) — both are part of the same risk haplotype and effects are not independent. Users carrying risk alleles at both loci are carrying the same underlying haplotype, not an additive combination.

The 10q26 HTRA1/ARMS2 locus interacts multiplicatively with the complement factor H gene (CFH rs1061170/rs800292, chromosome 1q32). The joint effect of homozygous risk at HTRA1 and CFH reaches OR=23.3 (95% CI: 2.5–218.2) with a combined population- attributable risk of 78.4% — meaning these two loci together explain the majority of genetic AMD risk in Western populations. Individuals carrying high-risk genotypes at rs106117077 rs1061170 in addition to rs2672598 CC should be considered at very high risk and are strong candidates for regular ophthalmic screening.

ARMS2 rs10490924 (the other major 10q26 variant, G>T) tags an overlapping haplotype and shows similar effect sizes to HTRA1. Because ARMS2 and HTRA1 are in high LD, rs10490924 and rs2672598 are correlated but non-identical — distinct SNPs capturing slightly different aspects of the 10q26 risk architecture.

Genotype Interpretations

What each possible genotype means for this variant.

TT Standard HTRA1 Expression Normal

Reference genotype — lowest HTRA1 promoter activity and lowest AMD risk at this locus

The TT genotype at rs2672598 means the HTRA1 promoter binds STAT4, NFκB, c-Ets-1, RelA, Elk-1, and WT1 — the standard set of transcription factors that drive normal, regulated HTRA1 expression in retinal tissue. Without the C allele, there is no recruitment of the higher-activity GR-alpha, AP-2αA, and Sp1 combination. Vitreous HTRA1 protein levels in TT homozygotes are approximately 2.56-fold lower than in CC homozygotes (measured directly in human vitreous samples from AMD surgery).

The 10q26 locus overall (including ARMS2 rs10490924, the neighboring gene) contributes substantially to AMD risk in the general population even at baseline, because the population-level lifetime prevalence of AMD is significant (~8–10% of people over 65 develop some form of AMD). However, the rs2672598 TT genotype specifically does not add risk above baseline — the massive amplification of risk seen in CT and CC carriers does not apply to you. The dominant risk factors for AMD in TT individuals remain age, smoking history, and family history of AMD.

CT Elevated HTRA1 Expression Intermediate

One C allele elevates HTRA1 promoter activity and confers intermediate elevated AMD risk

The CT genotype produces an intermediate HTRA1 promoter output — one allele with the standard T-driven transcription factor binding and one allele with the higher-activity C-driven binding. In codominant fashion, this yields HTRA1 protein levels between the TT and CC extremes.

Meta-analysis data from 14 case-control studies yields consistent heterozygote odds ratios of 2.13–2.27 for AMD risk. These effects are replicated across Caucasian, East Asian, and Indian populations. The risk is specifically for neovascular (wet) AMD — the form that causes rapid, severe vision loss through choroidal new vessel growth into the subretinal space.

An important environmental interaction: smoking interacts multiplicatively with HTRA1 risk alleles. CT carriers who smoke face substantially higher AMD risk than either risk factor alone would predict. This is among the strongest gene-environment interactions documented in AMD genetics (Tam et al. 2008).

rs2672598 CT is distinguishable from the polypoidal choroidal vasculopathy (PCV) subtype of AMD: the C allele specifically predicts classic exudative AMD, not PCV, which is a clinically relevant distinction for treatment planning.

CC High HTRA1 Expression High Risk

Homozygous C allele — highest HTRA1 promoter activity, 2.56-fold higher retinal HTRA1 protein, and markedly elevated wet AMD risk

The CC genotype at rs2672598 represents maximal HTRA1 promoter activity through the highest-output transcription factor combination. The 2.56-fold elevation in vitreous HTRA1 protein is a direct biological measurement — not an association signal — confirming the functional consequence of this genotype in the target tissue.

Mechanistically, excess HTRA1 serine protease in the subretinal space degrades Bruch's membrane extracellular matrix components (fibronectin, vitronectin, fibrillin), weakens the structural barrier between the retinal pigment epithelium and choroidal vasculature, and cleaves LTBP-1 to release active TGF-β — all of which promote the pathological angiogenesis of wet AMD.

The specific AMD subtype matters clinically. The C allele at rs2672598 is more strongly associated with classic exudative AMD (OR~10.5 in one study) than with polypoidal choroidal vasculopathy (PCV), which has a distinct pathophysiology. This distinction influences treatment planning: PCV responds differently to anti-VEGF monotherapy versus combination photodynamic therapy.

The joint risk with CFH variants is substantial: CC at rs2672598 combined with homozygous risk at CFH (rs800292 or rs1061170) yields OR=23.3 for AMD. These two loci together account for ~78% of the population-attributable risk of wet AMD in Western populations. If your CFH genotype is also high-risk, this represents one of the highest genetic risk profiles for a common adult-onset disease that genetic testing can identify.

The smoking interaction is multiplicative: CC homozygotes who smoke face approximately 15-fold higher risk than non-smoking TT individuals (based on data from the closely linked rs11200638 variant in the same haplotype).