Skip to content
Preprint · CC-BY via OpenAlex

A generator-matrix model quantifies the limited contribution of measured biomarkers to human mortality acceleration

Masato Tanigawa, Takafumi Iwaki

medRxiv · 2026

Abstract

Aging clocks and biomarkers are increasingly used as if they were the mechanism that drives mortality. But a mechanism, unlike a thermometer, must satisfy three conditions: it must account for the mortality signal, be causal, and be the layer that reverses when aging is reversed. Using public or provider-restricted de-identified data, we test all three. First (accounting), a Markov generator-matrix model with death as an absorbing state, fitted jointly to biomarkers and mortality in NHANES (n=23,512) and replicated in the Health and Retirement Study, assigns most of the Gompertz rise in mortality to a component latent to measured blood biomarkers (92.7%; 89.1% under a mean-field re-specification; 91.5% on replication). Second (causation), a positive-control-calibrated, two-platform cis-pQTL Mendelian-randomization design (UKB-PPP, deCODE) detects known causal proteins (LPA, IL6R) yet finds the measurable inflammatory, renal and growth-signalling markers null. Third (reversibility), at donor level reprogramming reverses a chronological clock (-9.7 and -22.4 yr), whereas a causality-enriched damage clock shows no detectable reversal while somatic identity is retained and moves only as pluripotency is approached. On these data the biomarkers meet none of the three conditions; aging measures appear to track mortality risk rather than its cause, a caution for their use as surrogate endpoints.

◌ CITATION ONLY
Full text is not openly licensed for redistribution here. Read it at the source:

Read at source →

Provenance

Source
OpenAlex
DOI
10.64898/2026.07.05.26356402
Canonical
link ↗
Fetched
2026-10-16 MST

Cite this

APA
Tanigawa, M., &amp; Iwaki, T. (2026). A generator-matrix model quantifies the limited contribution of measured biomarkers to human mortality acceleration. <em>medRxiv</em>. https://doi.org/10.64898/2026.07.05.26356402
Vancouver
Tanigawa M, Iwaki T. A generator-matrix model quantifies the limited contribution of measured biomarkers to human mortality acceleration. medRxiv. 2026. doi:10.64898/2026.07.05.26356402.
BibTeX
@unpublished{masato2026Agener, title = {A generator-matrix model quantifies the limited contribution of measured biomarkers to human mortality acceleration}, author = {Masato Tanigawa and Takafumi Iwaki}, journal = {medRxiv}, year = {2026}, doi = {10.64898/2026.07.05.26356402}, }

Research neighborhood

References, citing works, and semantically nearest findings. Click a node to open it.

Related findings