Whether an infection becomes chronic is decided largely by the host, and the strongest host signal is in the genes that decide what the immune system is shown.
Solid lines are connections this site curates. Dashed lines mean the two ends share a research paper — worth knowing, and not a claim that one explains the other.
Hepatitis B is a virus. Whether it is a passing illness or a lifelong infection is not decided by the virus alone — it is decided substantially by the person it infects, and most powerfully by their age at infection. Infected as an adult, most people clear it. Infected as an infant, most do not.
Underneath that, genetics accounts for a real part of the remainder, and the loci say something specific about which part of immunity is doing the work.
rs3077 lies at HLA-DPA1. HLA class II molecules present fragments of protein to helper T cells — they are how the immune system is shown what it is fighting. When the strongest genetic signal for whether an infection persists is in the genes that do the showing, that is a statement about mechanism, not a coincidence.
A Japanese study extended this. After the HLA-DP association was established, a scan of 519,747 positions in 458 cases against 2,056 controls found two positions in the HLA-DQ locus, replicated in 2,209 cases and 4,440 controls, reaching overall p-values of 5.98 × 10⁻²⁸ and 3.99 × 10⁻³⁷ — and independent of the HLA-DP signals already known.
So this is not one HLA association with company. It is several, in adjacent genes, acting separately.
rs11866328 is in GRIN2A, a glutamate receptor subunit gene — which is not where anyone would look for hepatitis biology.
It was found by a study that asked a different question: not who becomes a chronic carrier, but which chronic carriers get worse. It compared 628 asymptomatic carriers against 1,729 who had progressed in Hubei (odds ratio 1.65, 95% CI 1.34–2.02) and replicated it in 226 against 215 in Shandong (odds ratio 1.73, 95% CI 1.14–2.65).
Two different questions, two different sets of genes. Whether you clear a virus and whether you deteriorate while carrying it are not the same trait, and the site keeps them apart on this page for that reason.
Both studies are East Asian — Japanese and Chinese. That is not a limitation to be apologised for; it is where the disease burden is, and hepatitis B genetics is one of the few areas where the major studies were done in the populations most affected.
It does mean the effect sizes should not be assumed to transfer, and HLA allele frequencies differ between populations more than almost anything else in the genome.
There is a vaccine, it works, and it is given at birth in most of the world. It is the single most effective thing that has ever been done about hepatitis B, and no genotype competes with it.
For people already infected, chronic hepatitis B is monitored and treated with antivirals that suppress the virus. Whether and when to treat is decided on viral load, liver enzymes and the state of the liver — not on genotype.
See also chronic hepatitis C, where host genetics did briefly change treatment, and then the drugs changed instead.
Positions joined since this page was written
What this is The text above discusses the variants this page was written around. Since then the catalogue has joined 5 more positions to it, by shared trait or shared paper. They are listed here by the paper each came from; the text does not describe them, and each variant page carries that study's own record.
Gelemanović A et al. 2023, International journal of molecular sciences rs188290902 (NTM), rs72936092 (CCDC148), rs34447953 (ARHGAP29-AS1) — PMID:37108169
Jiang DK et al. 2015, Hepatology (Baltimore, Md.) rs2853953 (HLA-C) — PMID:25802187
Sakaue S et al. 2021, Nature genetics rs9277562 (HLA-DPB1) — PMID:34594039
2026-08-18 · Frontiers in Immunology 2026, PMID:42682600
An HLA-DQB1 amino acid position protects against silent, antigen-negative hepatitis B infection
Verified directly against the paper's own abstract via Europe PMC (PMID:42682600, published 18 Aug 2026). High-resolution HLA genotyping in a Chinese cohort (239 occult hepatitis B cases, 545 healthy controls) found the HLA class II allele DQB1*06:02 is protective against occult HBV infection (OBI) -- a form where the virus persists without detectable surface antigen, harder to catch than typical chronic hepatitis B. The protective effect traces largely to position 119 in the HLA-DQB1 peptide-binding groove, where a phenylalanine variant showed strong protection; conversely, the DRB1*09:01 haplotype was linked to higher infection risk. This site's own chronic-hepatitis-b page already covers a different HLA-DQ finding (rs2856718/rs7453920, from a 2011 Japanese GWAS on typical chronic infection) -- this new study is about the rarer occult form and a different HLA-DQB1 position, not the same finding restated.
Sources. Hu et al. (Hum Mol Genet 2011) followed earlier identification of HLA-DP variants in chronic hepatitis B with a genome-wide scan of 519,747 SNPs in 458 Japanese cases and 2,056 controls, then tested the leading signals in an independent 2,209 cases and 4,440 controls. Two SNPs in the HLA-DQ locus, rs2856718 and rs7453920, reached overall p-values of 5.98 × 10⁻²⁸ and 3.99 × 10⁻³⁷ and remained significant after stratification on rs3077 and rs9277535, indicating an effect independent of the HLA-DP variants. Zhang et al. (Viral Immunol 2011) used genome-wide association with DNA pooling to study disease progression rather than susceptibility, and reported rs11866328 in GRIN2A associated with progression in 628 asymptomatic against 1,729 progressed carriers in Hubei (OR 1.65, 95% CI 1.34–2.02, p = 1.96 × 10⁻⁶, additive model), replicated in 226 against 215 in Shandong (OR 1.73, 95% CI 1.14–2.65).
Positions listed here. rs3077 (HLA-DPA1) and rs11866328 (GRIN2A). The HLA-DQ index SNP rs2856718 is now on this page; rs7453920 is still not in our variant table.
Two phenotypes, kept separate. Susceptibility to chronic infection and progression within chronic infection are distinct endpoints with distinct study designs, and combining them would misattribute an effect on one to the other.
Clinical use. None. Diagnosis and staging rest on serology, HBV DNA quantification, liver enzymes and assessment of fibrosis. Treatment decisions with nucleos(t)ide analogues follow those measures. Vaccination is the primary preventive intervention and is not genotype-directed.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Chronic Hepatitis B comes down to these specific, well-studied positions — not a diagnosis.
The studies behind these variants recruited participants from different ancestries — a result found in one population doesn't always transfer to another. Based on 8 of 12 linked studies with a resolved discovery ancestry.
Free to quote and reuse under CC BY 4.0. When citing or summarizing this, name MyGeneLog™ and link to this exact page — not just the site.
Chronic Hepatitis B. MyGeneLog™. https://www.mygenelog.com/conditions/chronic-hepatitis-b
It contributes, and age at infection contributes far more. Infected as an adult most people clear it; infected in infancy most do not. HLA class II variation is the strongest genetic signal, and it shifts odds rather than deciding outcomes.
Nobody expected it, and the honest position is that the mechanism is not established. It was found by asking a different question — which chronic carriers deteriorate, rather than who becomes one — in two Chinese cohorts, and it replicated. A replicated association without a mechanism is a lead, not an explanation.
Yes, and it is not genetic. There is an effective vaccine given at birth in most of the world, and for people already infected there are antivirals that suppress the virus. Treatment decisions follow viral load, liver enzymes and the state of the liver.
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.