Skip to content

Abul-Husn NS, Cheng X, Li AH, et al. A Protein-Truncating HSD17B13 Variant and Protection from Chronic Liver Disease. N Engl J Med. 2018;378(12):1096-1106. PMID 29562163

One-paragraph summary

Exome sequence data linked to electronic health records from 46,544 DiscovEHR participants were searched for variants associated with serum ALT and AST; hits were replicated in three further cohorts (12,527 people), tested against clinical diagnoses of chronic liver disease in two independent cohorts (37,173 people total), and tested against histopathological severity in 2,391 human liver samples. A splice variant in HSD17B13 — rs72613567:TA, encoding the hepatic lipid-droplet protein hydroxysteroid 17-beta dehydrogenase 13 — was associated with reduced ALT (p = 4.2 × 10⁻¹²) and AST (p = 6.2 × 10⁻¹⁰), and with dose-dependent reductions in risk of alcoholic liver disease (42% in heterozygotes, 53% in homozygotes), nonalcoholic liver disease (17%, 30%), alcoholic cirrhosis (42%, 73%) and nonalcoholic cirrhosis (26%, 49%). In liver tissue the variant was associated with reduced risk of nonalcoholic steatohepatitis but not steatosis, and it mitigated the injury associated with the PNPLA3 I148M risk allele. The variant produces an unstable, truncated protein with reduced enzymatic activity.

Key findings

Outcome Heterozygote risk reduction (95% CI) Homozygote
Alcoholic liver disease 42% (20–58) 53% (3–77)
Nonalcoholic liver disease 17% (8–25) 30% (13–43)
Alcoholic cirrhosis 42% (14–61) 73% (15–91)
Nonalcoholic cirrhosis 26% (7–40) 49% (15–69)
  • Protects against steatohepatitis but not steatosis — it blocks the transition rather than the accumulation.
  • Mitigates PNPLA3 I148M-associated injury.
  • Loss of function is the protective direction, making the target pharmacologically straightforward to mimic.
  • Protection is not aetiology-specific: it covers alcohol-related as well as metabolic disease.

Limitations

  • Discovery is health-record-linked and association-based; the histological component (n=2,391) is cross-sectional.
  • Effect sizes for homozygotes carry wide intervals because homozygotes are uncommon.
  • The enzymatic function of HSD17B13 and the mechanism of protection were not established by the paper and are still incompletely characterised.
  • Funded by a pharmaceutical company with a direct interest in the target — disclosed, and the result has since replicated independently.

Why it matters

It is the counterweight to PNPLA3 and the clearest example in this disease of human genetics generating a druggable target with a pre-specified direction of effect. The protective allele reduces fibrosis progression (aHR 0.69, 95% CI 0.51–0.92) and increases both MASH resolution (1.58, 1.13–2.22) and fibrosis regression (1.42, 1.09–1.85) in serial-biopsy data (Vilar-Gomez 2026, PMID 40998180), and it is a component of the polygenic scores that predict HCC beyond fibrosis stage (Bianco 2021, PMID 33248170). Pharmacologically, rapirosiran — a GalNAc-conjugated siRNA against HSD17B13 — achieved a median 78% reduction in liver mRNA at six months in a phase 1 study with biopsy-confirmed target engagement and no drug-induced liver injury (Sanyal 2025, PMID 40581300), and is in phase 2 (NCT05519475). The linked phase 1 paper reports Parts A/B n=58/n=46, whereas the live NCT04565717 record says terminated with actual enrollment 6; that unresolved publication–registry discrepancy is not reconciled by substituting either source's number. The strategic logic — validate the target in human genetics before spending on a programme — is set out explicitly for this field in Lindén 2023 (PMID 37207913).

Cited by wiki pages

  • genetics.md
  • overview.md