HSD17B13

HSD17B13 rs6834314 — A Tag SNP Telling the Same Story as rs72613567

In 2011, a landmark genome-wide association study scanning 61,089 individuals of Caucasian
and Asian Indian descent identified an intergenic variant near the HSD17B13 gene on
chromosome 4 as one of the strongest signals for plasma ALT concentrations — a key marker
of liver injury:
Chambers JC et al. Genome-wide association study identifies loci influencing concentrations of liver enzymes in plasma. Nature Genetics, 2011.
That variant was rs6834314, and its protective G allele was associated with lower liver
enzyme levels in the general population. However, subsequent functional work established
that rs6834314 itself has no direct effect on HSD17B13 gene expression or enzyme activity.
Its associations are entirely a consequence of
linkage disequilibrium | Two variants are in LD when they co-occur on the same chromosome more often than expected by chance; r² measures how predictively one tags the other
with the causal splice variant rs72613567 (D'=0.995, r²=0.93). Knowing your
rs6834314 genotype tells you almost exactly what your rs72613567 genotype is —
these two variants are nearly perfect proxies for each other.

The Mechanism

rs6834314 is located approximately 11 kb downstream of the HSD17B13 gene in an intergenic
region and has no predicted regulatory or coding consequence. The variant's associations
with liver disease risk, liver inflammation, and liver enzyme levels are entirely attributable
to its co-inheritance with rs72613567, the
adenine insertion adjacent to the splice donor site of intron 6 | rs72613567 disrupts the splice donor site, generating an aberrant truncated transcript (isoform D) that encodes a non-functional HSD17B13 protein with reduced enzymatic activity
in HSD17B13. This was confirmed by Ma et al. (2019), who showed that rs6834314 genotype
was not associated with hepatic HSD17B13 expression levels after accounting for
the rs72613567 splice variant.

The Evidence

The clinical signal captured by rs6834314 mirrors the rs72613567 findings precisely,
because the two variants tag the same underlying haplotype. In 768 Caucasian patients
with biopsy-proven NAFLD, Ma et al. found the rs6834314 G allele (tagging the
HSD17B13 loss-of-function haplotype) was associated with lower risk of liver
inflammation (OR 0.77), reduced hepatocyte ballooning (OR 0.67), fewer
Mallory-Denk bodies (OR 0.68), and lower serum transaminases and GGT. In the
general population cohort (Michigan Genomics Initiative), the G allele was associated
with a 21% reduction in cirrhosis risk (OR 0.79, p=7.5×10⁻⁴):
Ma Y et al. 17-Beta Hydroxysteroid Dehydrogenase 13 Is a Hepatic Retinol Dehydrogenase Associated With Histological Features of Nonalcoholic Fatty Liver Disease. Hepatology, 2019.

In a multi-ethnic Asian cohort, Seto et al. followed 165 biopsy-proven NAFLD patients
for a mean of 89 months and found that each rs6834314 G allele was associated with
lower odds of NAFLD (adjusted OR 0.59, 95% CI 0.40–0.87) and NASH
(adjusted OR 0.48, 95% CI 0.31–0.75). Homozygous GG individuals showed markedly
lower liver-related complication rates during follow-up. However, the protective
effect was observed only in ethnic Chinese, not in Malays or Indians within the
same cohort — likely reflecting population-specific LD patterns between rs6834314
and the causal rs72613567 insertion:
Seto WK et al. Loss-of-function HSD17B13 variants, non-alcoholic steatohepatitis and adverse liver outcomes: results from a multi-ethnic Asian cohort. Clinical and Molecular Hepatology, 2021.

In Japanese patients with biopsy-proven NAFLD (n=290), rs6834314 G allele carriage
abolished the fibrosis-promoting effect of PNPLA3 I148M: among HSD17B13 AA carriers,
PNPLA3 GG conferred significantly higher advanced fibrosis risk (OR 2.4, p=0.041);
among HSD17B13 AG/GG carriers, no such PNPLA3 effect was detected. The G allele was
also associated with lower inflammation and ballooning prevalence:
Seko Y et al. Attenuated effect of PNPLA3 on hepatic fibrosis by HSD17B13 in Japanese patients with non-alcoholic fatty liver disease. Liver International, 2020.

Practical Implications

The clinical interpretation of rs6834314 is structurally identical to that of rs72613567,
because they tag the same biological state. If your genome file includes rs6834314 but
not rs72613567 (a scenario most likely on older Illumina arrays that predated specific
addition of the HSD17B13 locus), your rs6834314 genotype tells you the same information:
whether you carry the HSD17B13 loss-of-function haplotype. The G allele protective
association with liver inflammation and fibrosis is real — it just originates from a
functionally upstream cause.

One important caveat specific to rs6834314: because protection depends on LD, and LD
patterns can differ between ancestral populations, the G allele is a reliable proxy
for rs72613567 in European and East Asian populations (r²=0.93) but may be a
less accurate proxy in other ancestries, particularly in populations not well-represented
in the original LD reference panels.

Interactions

The most clinically important interaction is between the HSD17B13 loss-of-function
haplotype (tagged by rs6834314 G) and PNPLA3 rs738409 (I148M). The high-risk
PNPLA3 GG genotype strongly predicts advanced fibrosis in HSD17B13 AA carriers
but not in G allele carriers — meaning the HSD17B13 protective haplotype functionally
overrides much of the PNPLA3 fibrosis risk. This same interaction was originally
described for rs72613567 in the Abul-Husn NEJM 2018 cohort and replicated here
using rs6834314 as the proxy.

See also rs58542926 (TM6SF2 E167K), which operates in the same hepatic lipid
droplet biology axis and further stratifies liver disease risk when combined with
PNPLA3 and HSD17B13 genotypes.