rs2968864 — KCNH2 KCNH2 QT interval GWAS variant (7q36.1)

The Independent hERG Repolarization Signal at 7q36.1

The KCNH2 gene at 7q36.1 encodes Kv11.1, universally known as the hERG potassium channel¹¹ hERG potassium channel
Human Ether-à-go-go Related Gene; carries the rapid delayed-rectifier current IKr that drives phase 3 repolarization of the cardiac action potential. IKr is the dominant current terminating each heartbeat. When IKr is reduced — by rare loss-of-function mutations, by common variants, or by drugs that block the hERG channel pore — the QTc interval lengthens, and the heart's vulnerability to life-threatening ventricular arrhythmias (including Torsades de Pointes, TdP) increases.

rs2968864 sits in the intergenic region approximately 19 kilobases upstream of KCNH2 at GRCh38 chr7:150,925,074. It was identified as an independent second signal at the KCNH2 locus in the landmark QTGEN genome-wide association study of 13,685 Europeans²² QTGEN genome-wide association study of 13,685 Europeans
Newton-Cheh C et al. Common variants at ten loci influence QT interval duration in the QTGEN Study. Nat Genet. 2009;41(4):399–406, alongside rs4725982 as the first signal. The C allele at rs2968864 decreases QTc by approximately 1.4–1.8 ms per allele — a modest but reproducible effect that reaches genome-wide significance (P = 8×10⁻¹⁶ in GWAS Catalog meta-analysis).

The Mechanism

rs2968864 has no annotated gene consequence — it lies outside any coding exon — but its location ~19 kb upstream of KCNH2 places it within a region that likely contains regulatory elements influencing KCNH2 transcription. The C allele is associated with a small reduction in QTc, consistent with mildly increased IKr activity (more rapid repolarization). The precise regulatory mechanism has not been characterized; the variant may alter transcription factor binding, chromatin accessibility, or enhancer activity at the KCNH2 promoter region. Its independence from the companion first signal at the locus (rs4725982, which tags a different haplotype) suggests at least two separate regulatory modules modulate KCNH2 expression levels in human cardiac tissue.

The Evidence

The primary evidence comes from two parallel 2009 GWAS consortia. The QTGEN study³³ QTGEN study
Newton-Cheh et al. Nat Genet 2009 analyzed 13,685 Europeans across three cohorts (Framingham Heart Study, Rotterdam Study, Cardiovascular Health Study) and identified rs2968864 as a second independent QT locus signal. The parallel QTSCD study⁴⁴ QTSCD study
Pfeufer et al. Common variants at ten loci modulate the QT interval duration in the QTSCD Study. Nat Genet. 2009;41(4):407–14 analyzed 15,842 Europeans across five cohorts and independently confirmed the KCNH2 locus. Together, 14 independent variants across 10 loci explain 5.4–6.5% of QTc variability in the population.

The multiethnic PAGE study⁵⁵ PAGE study
Seyerle AA et al. Evidence of heterogeneity by race/ethnicity in genetic determinants of QT interval. Epidemiology. 2014;25(6):790–8 examined 21 GWAS variants across five population groups (European n = 16,398; African n = 5,437; American Indian n = 5,032; Hispanic n = 1,143; Asian n = 932) and found that effect sizes were heterogeneous across ancestries for many loci — important because the C allele frequency varies substantially: ~24% in Europeans, ~5% in Africans and East Asians, ~20% in South Asians. Clinical interpretation should account for this ancestry stratification.

The C allele shows a clear dose-response relationship with QTc: each C allele decreases QTc by ~1.4 ms. This means TT individuals (reference allele homozygotes) have, on average, a ~2.8 ms longer QTc than CC individuals from this locus alone. While 2.8 ms is smaller than clinical thresholds (QTc ≥450 ms for men, ≥460 ms for women), it contributes to an individual's position within the population distribution of QTc — and in the context of drug-induced QT prolongation, a longer baseline QTc leaves less buffer before clinically dangerous thresholds are crossed.

Practical Actions

For TT homozygotes: while the population-wide risk attributable to this locus alone is modest, the variant contributes to baseline QTc length. In the context of QT-prolonging medications (antiarrhythmics, some antipsychotics, macrolide antibiotics, fluoroquinolones, azole antifungals), a longer QTc baseline reduces the available pharmacological safety margin. A baseline 12-lead ECG before starting any hERG-inhibiting drug is the key actionable step.

CT heterozygotes have intermediate QTc and intermediate considerations.

CC carriers have a slightly shorter baseline QTc from this locus and are modestly more protected against drug-induced QTc prolongation — though this provides no protection against primary short QT syndrome or other arrhythmia risks.

Interactions

rs2968864 is one of two independent signals at the KCNH2/7q36.1 locus; rs4725982 is the companion first signal on a distinct haplotype. The related variant rs2968863 (also at 7q36.1, ~1.1 kb from rs2968864) is in partial LD and tags the K897T nonsynonymous variant rs1805123 in KCNH2 — associated with early-onset lone atrial fibrillation at OR 2.40 for TT homozygotes (PMID 24074973). The NOS1AP variant rs10918594 provides an independent QTc modifier at chromosome 1q23.3; carrying risk alleles at both loci produces additive QTc lengthening. Individuals who carry the TT genotype at rs2968864 and also carry risk alleles at NOS1AP loci may have meaningfully elevated drug-induced QT prolongation risk.