Models & Researchaging clockbiomarkerscohortsliver

Researchers Develop Liver Aging Index That Predicts Mortality

|June 16, 2026

5.0

Relevance Score

Photo: fightaging.org · rights & takedowns

FightAging summarises a study that developed the Liver Aging Index (LAI) using data from the China Kadoorie Biobank (CKB, N = 21,629), combining three clinical factors, eight blood biomarkers, and two imaging biomarkers (fat attenuation parameter and liver stiffness measurement). Per the FightAging report, external validation in NHANES (N = 3,412) and the VCTE-Prognosis cohort (N = 12,170, 16 centres) yielded discrimination metrics of 0.764 and 0.759, respectively. Each 1-unit increase in liver aging acceleration (LAA) was associated with 22%-85% higher all-cause mortality risk and 34%-170% higher liver-related event risk, per FightAging. Genetic analyses reportedly link predisposition to accelerated liver aging with elevated risks including liver cancer (odds ratio 7.82), with validation in Biobank Japan. The primary academic paper is not yet independently indexed; these figures are sourced from the FightAging commentary.

What happened

FightAging, a longevity-research aggregator, summarises a study that developed the Liver Aging Index (LAI) using data from the China Kadoorie Biobank (CKB, N = 21,629), according to the FightAging post. The post reports the LAI combined three clinical factors (including blood pressure), eight blood biomarkers (including lipoprotein cholesterol), and two imaging biomarkers - fat attenuation parameter and liver stiffness measurement. External validation is reported in NHANES (N = 3,412) and the VCTE-Prognosis cohort (N = 12,170, 16 centres), with discrimination metrics of 0.764 in NHANES and 0.759 in VCTE-Prognosis, per FightAging. Liver aging acceleration (LAA), defined as LAI minus chronological age, was reportedly associated with a 22%-85% higher all-cause mortality risk and 34%-170% higher liver-related event or mortality risk per 1-unit increase. The post notes genetic analyses linking predisposition to accelerated liver aging with a liver cancer odds ratio of 7.82, validated in Biobank Japan.

Technical context

Organ-specific aging indices built from multimodal clinical data (routine labs plus imaging) are an active area in translational bioinformatics. A 2026 study in npj Digital Medicine (nature.com/articles/s41746-026-02488-7) established seven imaging-based organ aging clocks - including a liver clock - that predict disease and mortality, providing broader context for this line of research. The FightAging summary does not include model architecture, training procedure, or complete biomarker list; those specifics are necessary to assess overfitting risk, calibration, and cross-population transportability.

Editorial note

The primary peer-reviewed paper did not surface in independent searches at the time of this audit. All quantitative figures above (discrimination scores, hazard ratio ranges, OR 7.82) are drawn from the FightAging commentary and should be treated as unverified until the source manuscript is confirmed. For practitioners, full methods, training-validation splits, subgroup calibration, and open code release are the standard bar before a new biological aging index is adopted in research or clinical pipelines.

Scoring Rationale

A longevity-blog summary of a liver aging index study with multi-cohort validation. Relevant context for practitioners applying ML to biomedical survival data, but sourced from a secondary aggregator with the primary paper unverified at audit time. Specific figures require confirmation against the source manuscript. Score reflects single secondary source, indirect ML angle, and unconfirmed primary publication.