A common East Asian genetic variant that leaves the body unable to fully break down a toxic alcohol byproduct, causing facial flushing and, with regular drinking, a substantially elevated risk of esophageal cancer.
When the body processes alcohol, it does so in two main steps. First, an enzyme converts ethanol into a compound called acetaldehyde — itself toxic and classified as a human carcinogen. Second, a different enzyme, ALDH2, quickly breaks that acetaldehyde down further into a harmless substance the body can clear. In most people this second step happens fast enough that acetaldehyde never builds up much. In people carrying a common genetic variant in the ALDH2 gene, that second enzyme works far less efficiently, so acetaldehyde accumulates rapidly after even modest drinking.
The buildup of acetaldehyde produces a recognizable set of symptoms, usually within minutes of drinking:
This is often called "Asian flush" or "Asian glow," reflecting the fact that the underlying variant is heavily concentrated in people of East Asian ancestry — particularly Chinese, Japanese, and Korean populations — and is rare in people of European or African ancestry.
It's tempting to treat alcohol flush as a harmless, if awkward, party inconvenience. It isn't. The same acetaldehyde buildup that causes flushing is directly implicated in a well-documented and substantially elevated risk of esophageal cancer among people who carry this variant and continue to drink regularly. Acetaldehyde damages DNA, and when the enzyme that would normally clear it away is impaired, tissues lining the esophagus are exposed to much higher local concentrations for much longer after each drink. This is one of the clearer examples in human genetics of a common variant with a well-established, mechanistically explained cancer association tied to a specific, modifiable behavior (drinking alcohol).
Because the flush reaction itself is such an effective early-warning signal — it's hard to ignore — many carriers naturally drink less or avoid alcohol altogether, which is protective. The people at the highest measured risk tend to be those who override the reaction and continue drinking regularly despite it, sometimes because of social or workplace drinking pressure, or because tolerance to the flushing develops somewhat with continued exposure even though the underlying enzyme problem does not go away.
A second, related variant affects the first step of alcohol breakdown instead of the second. Certain versions of the ADH1B gene make that first enzyme unusually fast, so acetaldehyde is produced more quickly than usual. On its own, a fast-acting ADH1B variant is also linked to some protection against alcohol dependence, likely because it makes drinking more unpleasant. But when a fast ADH1B variant is combined with an impaired ALDH2 variant, acetaldehyde is generated quickly and cleared slowly — a combination associated with the highest cancer risk of all, particularly with regular drinking.
Estimates vary, but the ALDH2 variant is thought to affect several hundred million people worldwide, overwhelmingly concentrated in East Asian populations, where a substantial minority to a large share of people carry at least one copy depending on the specific population studied. It is one of the most common single functionally significant genetic variants in humans.
Gene and variant. ALDH2 (OMIM *100650), on chromosome 12q24.2, encodes mitochondrial aldehyde dehydrogenase 2, the primary enzyme responsible for oxidizing acetaldehyde to acetate. The clinically relevant variant is rs671, a missense change historically described as Glu487Lys (using numbering that excludes the initiator methionine) and reported in current transcript-based nomenclature as NM_000690.4:c.1510G>A, p.(Glu504Lys) — commonly called the ALDH2*2 allele. This corresponds to OMIM phenotype entry #610251, Alcohol Sensitivity, Acute.
Mechanism. ALDH2 functions as a homotetramer. The p.Glu504Lys substitution disrupts the enzyme's active site and acts in a dominant-negative fashion within tetramers that mix normal and variant subunits, so even heterozygotes show markedly reduced enzyme activity — not simply half of normal. Homozygotes for the variant allele have close to no functional ALDH2 activity. As a result, both heterozygotes and homozygotes accumulate acetaldehyde rapidly after drinking, though the effect is more pronounced in homozygotes, who often experience such severe symptoms that they avoid alcohol almost entirely. Because both genotype classes show clinically apparent effects, the trait behaves as functionally dominant for flushing even though "deficiency" at the enzymatic level is most complete in homozygotes.
Population frequency. The rs671 A allele is essentially absent outside East Asia (frequency well under 1% in European and African reference populations) but reaches an estimated 20–30% allele frequency in East Asian populations, with meaningful variation between Chinese, Japanese, and Korean cohorts. Given an allele frequency in this range, roughly 35–50% of individuals in some East Asian populations are expected to carry at least one copy (1-(1-p)² for p ≈ 0.2–0.3), consistent with reported carrier estimates in the literature.
Esophageal cancer association. A 2023 meta-analysis and Mendelian randomization study (Zhang B, et al., Cancer Medicine, PMID 37795758) pooling more than 15,000 esophageal squamous cell carcinoma (ESCC) cases found that, considered without stratifying by drinking status, the rs671 variant allele was associated with lower ESCC risk overall (OR approximately 0.60, 95% CI 0.50–0.73) — reflecting that many carriers drink little or no alcohol because of the aversive flush reaction. However, among people who do drink alcohol, the picture reverses: a separate Chinese case-control study of 1,190 cases and 1,883 controls (Suo C, et al., J Thorac Oncol 2019, PMID 30639619) reported that drinkers carrying the rs671 variant allele had substantially increased ESCC risk (adjusted OR 1.98, 95% CI 1.51-2.60) compared with drinkers without it. That study also found the ADH1B fast-metabolizing variant, rs1229984, associated with increased ESCC risk overall (OR approximately 2.50, 95% CI 1.70–3.69), consistent with a "fast production, slow clearance" model of acetaldehyde-driven carcinogenesis in individuals carrying both variants who drink regularly. Earlier Mendelian randomization work using ALDH2 genotype as a proxy for alcohol exposure (Lewis & Davey Smith, Cancer Epidemiol Biomarkers Prev, PMID 16103445) similarly supports a causal role for alcohol-driven acetaldehyde exposure in esophageal cancer risk among ALDH2-deficient drinkers, rather than a purely correlational one.
Clinical counseling point. Genetic or family-history evidence of ALDH2 deficiency (a strong flush reaction to modest alcohol intake) is a reasonable basis for a clinician to discuss reducing or avoiding alcohol, given the compounded esophageal cancer risk specifically among ALDH2-deficient individuals who drink regularly; this is a behavioral/lifestyle discussion point rather than a pharmacologic one.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about ALDH2 Deficiency (Alcohol Flush Reaction) comes down to these specific, well-studied positions — not a diagnosis.
It's a real warning sign, not just cosmetic. The flushing reflects a buildup of acetaldehyde, a toxic, DNA-damaging compound. People with this variant who continue to drink regularly despite the flush have a substantially higher risk of esophageal cancer than drinkers without the variant.
Not necessarily — flushing intensity varies, and some people develop a degree of tolerance to the visible flush with repeated exposure even though the underlying enzyme deficiency does not change. A genetic test for rs671, or a strong flush history in close relatives, is more informative than how noticeable your own reaction currently feels.
No, it is not an allergy. It is an enzyme deficiency that slows the breakdown of a toxic byproduct of alcohol metabolism (acetaldehyde), which is what causes the flushing and other symptoms — a metabolic issue rather than an immune one.
The variant arose and became common in East Asia; it is thought to have spread partly because it discourages heavy drinking, which may have offered some survival advantage historically. It remains rare in European and African populations.
Yes — this combination is linked to the highest esophageal cancer risk among regular drinkers, because acetaldehyde is produced quickly by ADH1B and then cleared slowly by the impaired ALDH2 enzyme, prolonging tissue exposure to the toxic byproduct.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.