By: Geoffrey Feld, Ph.D. | Geocyte LLC

What if you could take a blood test that could tell you how you and your organs are aging (Goeminne, 2024) and provide recommendations on how to slow your specific aging mechanisms? What if there were “longevity clinics” you could then attend to help you implement those changes?

Does this sound like science fiction? Academic researchers, companies, and governments are lining up to make it a reality. The Biomarkers of Aging Consortium is the research hub for communicating the current science behind tracking and treating age-related decline.

Increasingly, proteomics is taking center stage as the lead role in the blood test pageantry. Epigenetic bio-age clock pioneer Steve Horvath asked, “Why do we like proteomics? Interpretability.”

Our previous biomarkers of aging blog introduced biological clocks and the ProteoAge clock, which was developed using the UK Biobank cohort. Here, we delve deeper into the discussion on measuring resilience and its impact on longevity, proteomics’ role in determining the Consortium Challenge winners, and validating biomarkers of aging with functional measurements.   

Frailty and Resilience

Lynne Cox, Ph.D., the Program Director for the Wellcome Leap Dynamic Resilience Program, defined frailty as “the loss of biological resilience to stress.” In effect, the mission of the Biomarkers of Aging Consortium can be reimagined as quantitatively defining frailty in the context of resilience mechanisms. What pathways are activated during stress that enable the body to cope? Why do some people appear more resilient than others? Can superior resilience be translated into therapies?

To understand the physiology of resilience, Dr. Cox studies zebrafish, which exhibit similar conserved mechanisms as mammals. It turns out that proteins are fantastic interpretable markers of stress and resilience. Her lab identified heritable resilience mechanisms imprinted during embryonic development and modulated by neuropeptides and components of the innate immune system (Swaminathan, 2023).

The idea of brain and immune system age is conveniently linked to UK Biobank proteomics research in humans, spearheaded by Stanford Professor Tony Wyss-Coray, Ph.D. Surveying plasma proteomics data, Oh et al. report in a preprint that neurofilament light chain (NEFL), myelin oligodendrocyte protein (MOG), glial fibrillary acidic protein (GFAP), and brevican (BCAN) significantly contribute to brain age, the latter of which is a brain extracellular matrix (ECM) component. Overall, brain aging strongly predicted mortality, and individuals with “youthful” brains and immune system signatures demonstrated significant longevity (Oh, 2024). This group also showed lower levels of neuroinflammation marker and ECM degrader metalloproteinase-9 (MMP9), further implicating immune function and ECM hardiness as resilience mechanisms “crucial for promoting longevity.”

Accepting the Challenge

Grand scientific challenges, such as the XPRIZE and the Critical Assessment of protein Structure Prediction (CASP), represent powerful incentives for empowering scientific initiatives in their infancy. The Biomarkers of Aging Challenge Series was established similarly to “stimulate innovation and collaboration” and “foster technological advancements” that drive bio-aging clock development in three phases:

1. Chronological Age Prediction ($30K awarded in 2024)
2. Longevity Prediction ($40K awarded at the Conference, November 2024)
3. Healthspan Prediction ($100K in awards TBD 2025)

For Phase 2, entrants were tasked with developing a biological aging clock that most closely predicts a unique cohort comprising comprehensive plasma biomarker data and longitudinal health outcomes information on over 500 individuals. In collaboration with the Methuselah Foundation, Alamar Biosciences donated NULISAseq Inflammation 250 and CNS 120 panel data analysis on all samples to the cause so research teams can accelerate aging research with high-quality and industry-leading proteomics sensitivity.

Alamar CTO Xiao-Jun Ma, Ph.D., had presented earlier in the conference on the power of NULISA in predicting biological age. The proteomic age signature developed with the Methuselah-Alamar cohort predicts chronological age with 94% accuracy (see figure).

“Team Kuai,” composed of doctoral students Jakob Träuble (University of Cambridge), Raphael Lermer (Helmholtz Munich), and Raphael Lermer (German Heart Center Munich), earned first place. Their model was trained on the UK Biobank proteomics and follow-up data, which included over 52,000 samples and 229 proteins. A key factor in their success
was recognizing that the training data (from the UK Biobank) and the challenge data (from the NULISA cohort) were not equivalent, necessitating the alignment and normalization of dataset distributions for each protein. Team Kuai then developed a risk score for each individual based on a Cox proportional hazards model of the protein-plus-follow-up data and applied these risk scores to a Gompertz distribution, commonly used to describe lifespans, to predict mortality ages.

Are you surprised to learn that NEFL topped the list of predictive proteins? Note that Mr. Träuble also placed 3^rd in the Phase 1 prize, so pay attention when he tells you how old you really are!

Form Follows Function

While refining their prize-winning model, Team Kuai compared its performance against other bio-age clocks in the Biolearn database. Biolearn is an open-source GitHub designed to evaluate and harmonize the various clocks under development (Ying, 2024). It represents a critical component of validating biomarkers of aging for clinical and consumer utility. 

Before gaining widespread appeal, biomarkers of aging must achieve an acceptable level of standardization in their development, measurement, and validation. For instance, leading regulatory agencies such as the FDA and EMA have yet to issue guidelines on these processes (Moqri, 2024). Including demographics, performance-based metrics, outcomes, and other metadata for individuals in study designs is essential for discerning and validating highly predictive biomarkers. Today’s biological clocks rely heavily on retrospective cohorts (e.g., the UK Biobank) and are restricted to the metadata that has already been collected. The significance of longitudinal collections and robust “functional biomarkers” that measure frailty is gaining recognition and being incorporated into prospective study designs (Biomarkers of Aging Consortium, 2024).

As the Biomarkers of Aging Challenge Series moves to Phase 3: Healthspan Prediction and biological clocks provide actionable insights into lifestyle modifications (Perlmutter, 2024), functional biomarkers become paramount. Routine evaluations of motor and cognitive function, including gait speed, grip strength, and balance, inform an individual’s frailty and reflect their body’s resilience in response to aging. While such functional biomarkers are ubiquitous in the aging field, they desperately deserve a refresher.

“Many functional biomarkers used in the field are subjective and haven’t changed in decades,” reflects John Ralston, Ph.D., CEO and Founder of Neursantys. “For multiomic fluid biomarker methods to positively impact seniors’ lives, they need benchmarking to quantitative, objective, and functional biomarkers that directly report on physiology.” Neursantys developed a wearable biosensor called Phybrata that digitally reports on an individual’s balance and gait instabilities, effectively taking these tests into the 21^st century (Hope, 2021). At the same time, Neursantys offers a noninvasive electrostimulation device called Neurvesta that turns the Phybrata signal into balance-restoring therapy for age, injury, and microgravity-related impairments (Ralston, 2023).

Aging is finally getting recognition as the strongest risk factor for the chronic ailments facing developed countries. Government funding initiatives, such as ARPA-H PROSPR and Singapore’s Centre for Healthy Longevity, global nonprofits, including the Hevolution Foundation and Methuselah Foundation, and innovative biotech companies, such as BioAge Labs, are investing heavily in longevity research and identifying interventions that prolong healthspan. Combining cutting-edge ‘omics technologies like Alamar’s NULISA proteomics with quantitative functional measurements, prospective cohort collections, artificial intelligence, and well-placed investments paints a bright picture of humanity’s future, leading longer, more productive, and disease-free lives.  

References

Goeminne et al. (2025) Cell Metab. Plasma protein-based organ-specific aging and mortality models unveil diseases as accelerated aging of organismal systems. 37(1): P205–222.E6.

Swaminathan et al. (2023) Cell Rep. Stress resilience is established during development and is regulated by complement factors. 42(1): 111973.

Oh et al. (2024) bioRxiv. Plasma proteomics in the UK Biobank reveals youthful brains and immune systems promote healthspan and longevity.

Ying et al. (2024) bioRxiv. A unified framework for systematic curation and evaluation of aging biomarkers.

Moqri et al. (2024) Nat Med. Validation of biomarkers of aging. 30(2): 360–372.

Biomarkers of Aging Consortium et al. (2024) Nat Aging. Challenges and recommendations for the translation of biomarkers of aging. 4: 1372–1383.

Perlmutter et al. (2024) Front Nutr. The impact of a polyphenol-rich supplement on epigenetic and cellular markers of immune age: a pilot clinical study. 11: 1474597.

Hope A et al (2021) Sensors. Phybrata sensors and machine learning for enhanced neurophysiological diagnosis and treatment. 21: 7417.

Ralston JD et al. (2024) Phybrata biomarker assessments of age-related balance impairments and EVS balance restoration. Biomarkers of Aging Conference, Harvard Medical School, Cambridge, MA. November 1–2, 2024.