Short telomeres, but not accelerated epigenetic or proteomic age, increase fibrotic lung disease risk: Mendelian randomisation with pleiotropy dissection and independent replication
- 1European Research Infrastructure for biobanking and biomolecular resources - BBMRI-ERIC
Abstract
BackgroundShort telomeres are an established cause of idiopathic pulmonary fibrosis (IPF), but the genetic evidence rests on seven instruments, six of which lie at telomere-maintenance loci, and so cannot separate an effect of telomere length from an effect of the telomerase pathway itself -- a distinction that determines whether telomere length is a tractable target or only a marker of the pathway. Whether other measures of biological ageing share the effect, and whether accelerated ageing causes immune-mediated disease more generally, has not been tested systematically.
MethodsTwo-sample Mendelian randomisation of six ageing measures -- GrimAge, PhenoAge, Hannum-age and intrinsic epigenetic age acceleration, proteomic age acceleration, and leukocyte telomere length -- against six immune-mediated diseases in FinnGen R12: giant cell arteritis (GCA), polymyalgia rheumatica, primary Sjogrens disease, interstitial lung disease (ILD), Crohns disease and ulcerative colitis. All 36 pairs were analysed in both directions using inverse-variance weighted (IVW), MR-Egger, weighted-median and weighted-mode estimators, with Cochrans Q, I{superscript 2}, leave-one-out and single-variant diagnostics, a pre-specified analysis excluding HLA-region instruments, and iterative Q-based outlier removal. Instruments were annotated against telomere-maintenance and pulmonary-fibrosis loci and that class excluded. Minimum detectable effects were computed for every pair before estimates were examined. Any pair surviving correction was re-estimated in two independent IPF case-control datasets. A two-step analysis tested mediation of ILD risk by seven gut microbial taxa acting through telomere length.
ResultsOne of 36 forward pairs survived correction. Longer telomeres reduced ILD risk: outlier-robust IVW odds ratio (OR) 0.59 per standard deviation (SD), 95% CI 0.49-0.71, p = 2.7 x 10-, on 77 instruments, with no residual heterogeneity (I{superscript 2} = 6%) and concordance across estimators. Four pleiotropic instruments -- in TERT, TERC, APOE and ATM -- inflated the unfiltered estimate to OR 0.46 (0.37-0.57) and accounted for all of its heterogeneity. The effect was reproduced in 5,159 clinician-diagnosed European IPF cases (OR 0.18, 0.12-0.25, p = 5.0 x 10-{superscript 2}3) and in 1,210 East Asian cases (OR 0.54, 0.35-0.83, p = 0.005). Excluding every telomere-maintenance and pulmonary-fibrosis locus left OR 0.26 (0.17-0.40), p = 1.6 x 10-, on 72 instruments, establishing that the effect does not depend on TERT or TERC. The effect was three times larger for clinically defined IPF than for the broad registry ILD endpoint (p = 1.8 x 10-), consistent with dilution of a fibrosis-specific mechanism by case mix. No methylation clock and no proteomic age measure affected any disease; smallest detectable ORs ranged from 1.29 to 5.24 per SD, so these nulls exclude only large effects. The reverse direction could not be tested at informative precision. One apparent clock-autoimmunity association was produced entirely by a single HLA-region instrument. Mediation was null at its first step.
ConclusionsTelomere attrition causally increases fibrotic lung disease risk, with an effect that concentrates in clinically defined IPF and that survives removal of every telomere-maintenance locus -- shown here for the first time -- and is therefore attributable to telomere length rather than to the telomerase genes that dominate conventional instrument sets. Four methylation clocks and proteomic age acceleration show no such effect, so these measures cannot be treated as interchangeable proxies for biological ageing.
Related
- Associations of systemic inflammatory indices with rheumatoid arthritis and all-cause mortality: a NHANES study with independent hospital-based replication
- Antihypertensive Medication Class and Incident Alzheimers Disease and Related Dementias: External Replication Across Two Independent Healthcare Data Sources-the Vanderbilt EHR Synthetic Derivative and US Medicare Claims
- Epigenetic signatures of different types of childhood maltreatment in a high-risk sample: a preregistered replication study
- Concurrent TMS-EEG to characterise interhemispheric signal propagation over the motor and prefrontal regions: An independent replication of GABAB receptor mechanisms
- cAMP Analogs Induce p53-independent p21 Expression through Sp1-responsive Promoter Activation and p300-associated Epigenetic Regulation in PC-3 Human Prostate Cancer Cells.
BibTeX
@misc{hanke2026short,
title = {{Short telomeres, but not accelerated epigenetic or proteomic age, increase fibrotic lung disease risk: Mendelian randomisation with pleiotropy dissection and independent replication}},
author = {S. Hanke},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.14.26364960},
url = {https://www.medrxiv.org/content/10.64898/2026.10.14.26364960v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Short telomeres, but not accelerated epigenetic or proteomic age, increase fibrotic lung disease risk: Mendelian randomisation with pleiotropy dissection and independent replication
AU - S. Hanke
PY - 2026
DA - 2026/10/14
PB - medRxiv
DO - 10.64898/2026.10.14.26364960
UR - https://www.medrxiv.org/content/10.64898/2026.10.14.26364960v1
AB - BackgroundShort telomeres are an established cause of idiopathic pulmonary fibrosis (IPF), but the genetic evidence rests on seven instruments, six of which lie at telomere-maintenance loci, and so cannot separate an effect of telomere length from an effect of the telomerase pathway itself -- a distinction that determines whether telomere length is a tractable target or only a marker of the pathway. Whether other measures of biological ageing share the effect, and whether accelerated ageing causes immune-mediated disease more generally, has not been tested systematically. MethodsTwo-sample Mendelian randomisation of six ageing measures -- GrimAge, PhenoAge, Hannum-age and intrinsic epigenetic age acceleration, proteomic age acceleration, and leukocyte telomere length -- against six immune-mediated diseases in FinnGen R12: giant cell arteritis (GCA), polymyalgia rheumatica, primary Sjogrens disease, interstitial lung disease (ILD), Crohns disease and ulcerative colitis. All 36 pairs were analysed in both directions using inverse-variance weighted (IVW), MR-Egger, weighted-median and weighted-mode estimators, with Cochrans Q, I{superscript 2}, leave-one-out and single-variant diagnostics, a pre-specified analysis excluding HLA-region instruments, and iterative Q-based outlier removal. Instruments were annotated against telomere-maintenance and pulmonary-fibrosis loci and that class excluded. Minimum detectable effects were computed for every pair before estimates were examined. Any pair surviving correction was re-estimated in two independent IPF case-control datasets. A two-step analysis tested mediation of ILD risk by seven gut microbial taxa acting through telomere length. ResultsOne of 36 forward pairs survived correction. Longer telomeres reduced ILD risk: outlier-robust IVW odds ratio (OR) 0.59 per standard deviation (SD), 95% CI 0.49-0.71, p = 2.7 x 10-, on 77 instruments, with no residual heterogeneity (I{superscript 2} = 6%) and concordance across estimators. Four pleiotropic instruments -- in TERT, TERC, APOE and ATM -- inflated the unfiltered estimate to OR 0.46 (0.37-0.57) and accounted for all of its heterogeneity. The effect was reproduced in 5,159 clinician-diagnosed European IPF cases (OR 0.18, 0.12-0.25, p = 5.0 x 10-{superscript 2}3) and in 1,210 East Asian cases (OR 0.54, 0.35-0.83, p = 0.005). Excluding every telomere-maintenance and pulmonary-fibrosis locus left OR 0.26 (0.17-0.40), p = 1.6 x 10-, on 72 instruments, establishing that the effect does not depend on TERT or TERC. The effect was three times larger for clinically defined IPF than for the broad registry ILD endpoint (p = 1.8 x 10-), consistent with dilution of a fibrosis-specific mechanism by case mix. No methylation clock and no proteomic age measure affected any disease; smallest detectable ORs ranged from 1.29 to 5.24 per SD, so these nulls exclude only large effects. The reverse direction could not be tested at informative precision. One apparent clock-autoimmunity association was produced entirely by a single HLA-region instrument. Mediation was null at its first step. ConclusionsTelomere attrition causally increases fibrotic lung disease risk, with an effect that concentrates in clinically defined IPF and that survives removal of every telomere-maintenance locus -- shown here for the first time -- and is therefore attributable to telomere length rather than to the telomerase genes that dominate conventional instrument sets. Four methylation clocks and proteomic age acceleration show no such effect, so these measures cannot be treated as interchangeable proxies for biological ageing.
ER -