rs564398 — CDKN2B-AS1

Secondary T2D risk variant at the 9p21 CDKN2A/B locus within ANRIL (CDKN2B-AS1); the T risk allele impairs glucose-stimulated pancreatic beta-cell proliferation, conferring a modest but independently replicated ~8% increased type 2 diabetes risk per allele

See your personal result for CDKN2B-AS1

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.

ANRIL and the 9p21 Beta-Cell Clock

The region of chromosome 9 known as 9p21 harbors one of the most consistently replicated type 2 diabetes risk signals in the human genome. rs564398 sits within CDKN2B-AS1 (ANRIL)11 CDKN2B-AS1 (ANRIL)
ANRIL stands for "antisense non-coding RNA in the INK4 locus" — a long non-coding RNA transcribed in the opposite direction from the CDKN2A and CDKN2B protein-coding genes at 9p21. lncRNAs regulate nearby gene expression through chromatin remodeling and other epigenetic mechanisms. — a long non-coding RNA that modulates expression of the adjacent tumor suppressor genes CDKN2A (p16) and CDKN2B (p15). These inhibitors act as brakes on cell division, and in pancreatic beta cells they control how readily the insulin-producing mass can renew itself.

rs564398 is a secondary signal at this locus, independent of but weaker than the primary variant rs10811661 (~200 kb away). The T risk allele carries a per-allele T2D odds ratio of approximately 1.08, compared to ~1.24 for the primary signal. Both act through the same broad mechanism — reducing glucose-stimulated beta-cell proliferative capacity — but may tag distinct regulatory elements within the locus.

The Mechanism

The 9p21 locus regulates how effectively pancreatic beta cells can replicate in response to metabolic demand. Unlike T2D variants that impair acute insulin secretion per cell (e.g., TCF7L2 or SLC30A8), rs564398 acts at the level of beta-cell mass maintenance22 beta-cell mass maintenance
Beta cells must periodically replace themselves as they age or are damaged by metabolic stress. The total number of functional beta cells declines over decades if renewal capacity is impaired, eventually reducing the pancreas's ability to secrete enough insulin..

A functional study by Kong et al. 201833 Kong et al. 2018
Kong Y et al. CDKN2A/B T2D Genome-Wide Association Study Risk SNPs Impact Locus Gene Expression and Proliferation in Human Islets. Diabetes 2018. PMID:29432124 measured BrdU incorporation (a direct index of cell division) in 43 human islet preparations cultured at high versus normal glucose concentrations. Islets carrying one or two T risk alleles at rs564398 showed significantly lower glucose-stimulated proliferation than CC homozygous islets — directly linking the variant to impaired beta-cell renewal under metabolic challenge. Notably, rs564398 did not alter expression of p14, p15, p16, or MTAP in islets, suggesting its mechanism operates through ANRIL at a regulatory level distinct from simple CDKN2A/B transcript abundance.

The Evidence

The T risk allele at rs564398 was first reported as a T2D susceptibility signal in the landmark WTCCC GWAS44 WTCCC GWAS
Zeggini et al. Replication of genome-wide association signals in UK samples reveals risk loci for type 2 diabetes. Science 2007. PMID:17463249, where it reached genome-wide significance with OR 1.13 (95% CI 1.08–1.19, p = 1×10⁻⁶). This was one of the largest initial T2D GWAS, covering ~1,900 cases from the WTCCC plus replication in ~3,700 additional cases.

The Cugino et al. 2012 meta-analysis55 Cugino et al. 2012 meta-analysis
Cugino et al. Type 2 diabetes and polymorphisms on chromosome 9p21: a meta-analysis. Nutr Metab Cardiovasc Dis 2012. PMID:21315566 of 22 chromosome 9p21 studies quantified the rs564398 signal at OR 1.08 (95% CI 1.05–1.12) with a population attributable risk of 6% — roughly half the PAR of the primary rs10811661 signal (15%). A further meta-analysis66 further meta-analysis
Peng et al. The relationship between five widely-evaluated variants in CDKN2A/B and CDKAL1 genes and the risk of type 2 diabetes: a meta-analysis. Gene 2013. PMID:24012816 of 16 studies (20,029 cases, 24,419 controls) found a significant association in Caucasians (OR 1.19, p=0.012) but not in Asians (OR 1.01, p=0.868), with marked ethnic heterogeneity at this SNP.

The C allele frequency varies substantially across populations — ~41% in Europeans, ~34% globally, ~13% in East Asians, and ~7% in Africans. This population stratification mirrors the primary 9p21 signal and may partially explain the stronger T2D associations observed in European cohorts.

Practical Actions

Because the 9p21 locus reduces beta-cell renewal capacity rather than impairing acute insulin secretion, the relevant intervention strategy centers on reducing cumulative metabolic demand on a beta-cell pool that has limited ability to self-replenish. This means:

  • Keeping postprandial glucose excursions small by choosing low-glycemic-load carbohydrates (legumes, lentils, non-starchy vegetables) reduces the secretory burden on each individual beta cell.
  • Periodic fasting glucose and HbA1c monitoring allows detection of gradually declining beta-cell reserve before frank diabetes develops.
  • Minimizing direct beta-cell stressors (excess fructose, saturated fat overload) is specifically relevant when renewal capacity is genetically constrained.

Interactions

rs564398 and the primary 9p21 SNP rs10811661 are approximately 104 kb apart and are not in complete linkage disequilibrium — they may represent independent regulatory signals within the locus. Carrying risk alleles at both SNPs could compound the impairment of beta-cell renewal through distinct regulatory elements within ANRIL and flanking chromatin.

Beyond the 9p21 locus, the 9p21 beta-cell mass signal acts through a mechanism completely distinct from the TCF7L2 pathway (rs7903146), which impairs incretin-stimulated insulin secretion per cell. Individuals carrying risk alleles at both loci face additive T2D susceptibility from two independent mechanisms: reduced beta-cell mass (9p21) and reduced per-cell insulin output (TCF7L2).

The ANRIL locus also carries well-established cardiovascular disease associations; rs564398 has been linked to coronary artery disease risk in a Turkish cohort, particularly in females, highlighting the pleiotropic role of this genomic region.

Genotype Interpretations

What each possible genotype means for this variant.

CC Protective Genotype Normal

Protective genotype — lowest T2D risk at this secondary 9p21 signal

You carry two copies of the protective C allele at rs564398. About 12% of people globally share this genotype, which confers the lowest type 2 diabetes risk at this secondary 9p21 signal. Functional studies show that CC homozygous islets have the highest glucose-stimulated beta-cell proliferation of the three genotypes, suggesting your insulin-producing cells retain greater renewal capacity at this locus.

The C allele is substantially more common in Europeans (~41%) than in East Asians (~13%) or Africans (~7%), so CC is notably more prevalent in European-ancestry populations.

CT Intermediate Risk Intermediate

One T risk allele — modestly elevated T2D susceptibility

The functional study by Kong et al. 2018 (PMID:29432124) directly measured cell division rates in human islet preparations stratified by rs564398 genotype. CT carriers showed intermediate beta-cell proliferation between CC (highest) and TT (lowest) under glucose stimulation, consistent with the additive model seen at the primary 9p21 SNP rs10811661.

The effect is modest compared to the primary locus variant, but it is independently replicated across multiple cohorts and meta-analyses. The PAR of ~6% means that if this variant were removed from the population, roughly 6% of T2D cases attributable to this locus would be prevented.

TT Elevated Risk High Risk

Two T risk alleles — highest T2D susceptibility at this secondary 9p21 signal

Kong et al. 2018 (PMID:29432124) demonstrated that human islet preparations homozygous for the T allele at rs564398 showed the lowest BrdU incorporation (a measure of cell division) under glucose stimulation among the three genotypes. The effect was specific to glucose-stimulated (15 mmol/L) conditions rather than baseline (5 mmol/L), confirming that the impairment is in the metabolic response of beta cells rather than baseline proliferation.

The Cugino 2012 meta-analysis placed the per-allele OR at 1.08 with a PAR of 6%. Ethnic heterogeneity is marked — the association is stronger in Caucasian cohorts (OR ~1.19 in some analyses) than in Asian populations, partly reflecting differences in C allele frequency (41% in Europeans vs 13% in East Asians).

Unlike variants that impair acute insulin secretion, the 9p21 risk is thought to accumulate over decades as the beta-cell pool gradually shrinks without adequate renewal — consistent with why age modifies the T2D association at this locus.