As dozens of therapies targeting the biology of aging move toward human clinical trials, researchers increasingly need reliable biomarkers that can rapidly determine whether these interventions are working without waiting decades for clinical outcomes such as chronic disease, disability, or mortality.
A new study by researchers at Yale School of Medicine provides an evidence-based roadmap for answering that question. By systematically evaluating more than 100 DNA methylation (epigenetic) biomarkers across more than 50 putative longevity interventions tested in humans, the researchers identified the biomarkers best suited to measure the effects of interventions designed to improve healthy aging.
The study, published in Nature Medicine, systematically evaluated 111 DNA methylation biomarkers, including 16 of the field's most widely used epigenetic clocks, across 51 longitudinal human intervention studies. These studies encompassed a range of putative longevity interventions, including calorie restriction, intermittent fasting, exercise, Mediterranean and plant-based diets, metformin, semaglutide, rapamycin, anti-TNF therapies, bariatric surgery, smoking cessation, hyperbaric oxygen therapy, dietary supplements, and other pharmacological and lifestyle approaches proposed to influence biological aging.
Rather than asking whether any single intervention slows aging, the researchers asked a broader question: Which biological aging biomarkers consistently detect meaningful biological changes across human longevity interventions?
The analysis revealed substantial differences among existing biomarkers. While many biological aging measures accurately predict future health outcomes, only a subset consistently detected changes following interventions.
Biomarkers designed to measure the pace of aging or predict mortality consistently demonstrated strong responsiveness across multiple interventions, suggesting they are promising candidates for use as endpoints in future geroscience clinical trials.
To ensure these findings are broadly accessible, the researchers included the complete results in TranslAGE, an open framework for evaluating biological aging biomarkers. They also launched TranslAGE.io, an interactive platform that allows researchers to explore the performance of more than 100 biomarkers across over 50 putative longevity interventions, compare biomarker responsiveness across studies, and identify the most appropriate biomarkers for future clinical trials.
"Our field has developed an extraordinary number of biological aging biomarkers, and an equally rapidly growing pipeline of interventions aimed at slowing or modifying biological aging," said Raghav Sehgal, PhD, associate research scientist and lead author of the study. "Yet the field has lacked a rigorous, evidence-based framework for determining which biomarkers are sufficiently prognostic, responsive, and reliable to serve as endpoints in clinical trials.
“We developed TranslAGE to address that gap,” he said. “This study focuses on the responsiveness component of the framework, systematically evaluating more than 100 aging biomarkers across over 50 human longevity interventions to identify which measures consistently detect biological changes following intervention. Together with upcoming studies examining the prognostic validity and technical and biological reliability of these biomarkers, TranslAGE provides a comprehensive roadmap for selecting aging biomarkers for geroscience research and clinical translation. Our hope is that it becomes a foundational framework for the next generation of longevity clinical trials."
The researchers believe the resource will help standardize biomarker selection across the field, improve comparisons between studies, and reduce one of the major barriers to developing therapies that target the biology of aging.
"As longevity therapeutics move toward larger human clinical trials, choosing the right biomarker becomes just as important as choosing the right intervention," said Albert Higgins-Chen, MD, PhD, assistant professor of psychiatry and senior author of the study. "TranslAGE gives researchers a practical roadmap for selecting the biomarkers that are most appropriate for evaluating therapies designed to extend healthy lifespan. By doing so, the field can minimize unnecessary multiple testing, ignore false positive results that would mislead the field, and mitigate publication bias. By making these analyses openly available through TranslAGE and TranslAGE.io, we hope to accelerate progress across the entire field."
The authors hope the study will serve as a roadmap for the field, helping researchers select biomarkers for future geroscience clinical trials, standardize outcome measures across studies, and advance the development of therapies that target the biology of aging.
Additional Yale authors include Daniel Borrus, Jessica Kasamato, Jenel F. Armstrong, John Gonzalez, and Yaroslav Markov. The work also included collaborators Michael Corley from UCSD, Jessica Lasky-Su from Harvard, and Ryan Smith, Natalia Carreras, and Varun B. Dwaraka of TruDiagnostic, all of whom contributed data and/or algorithms that were analyzed at Yale. This complemented publicly available data and algorithms contributions from the broader DNA methylation biomarker community and the longevity intervention research community.
The study was supported by the National Institute on Aging (R01AG065403), the Impetus Grant, the Gruber Science Fellowship at Yale University, and the Thomas P. Detre Fellowship Award in Translational Neuroscience Research at Yale University.