Healthspan Weekly — Mar 10, 2026

Two things happened this week that, taken together, tell you where this field is actually headed: scientists built a biological aging clock that finally stitches together proteins, metabolites, and DNA methylation into one score that predicts death better than anything before it — and a completely separate study showed that the difficult person in your life is measurably accelerating your epigenetic aging. Better measurement and harder-to-ignore evidence that the "soft" stuff is biology. That's the week.

![Lab team analyzing blood and DNA samples — CDC scientist transfers H7N9 / Photo Credit: James Gathany

The problem with biological age clocks has always been fragmentation. One measures chemical tags on your DNA. Another reads proteins in your blood. A third tracks metabolites. They don't always agree, and none was trained on enough real health data to be truly useful in a clinic.

A new paper in Nature Aging introduces OMICmAge, and it takes a different approach. Built from roughly 31,000 electronic health records in the study dataset, it integrates proteomic and metabolomic information through epigenetic proxies — essentially using multiple biological data streams to train a single DNA-methylation score. The result predicts mortality at least as well as, and often better than, existing clocks.

What makes this more than incremental: rather than forcing you to choose between proteomics or epigenomics, OMICmAge reconciles them. Different organs age at different speeds — your liver clock and your brain clock can diverge sharply — and tools like this are steps toward capturing that complexity in something a clinician can actually act on.

The near-term question is adoption. If commercial platforms like TruDiagnostic integrate multi-omic methodology, and if ARPA-H's new healthspan trial platforms (more on that below) standardize which clocks get used in large studies, OMICmAge or its successors could become the default readout for whether an intervention is actually working. Meanwhile, clinics are already using pace-of-aging clocks like DunedinPACE for shorter-term feedback — a complementary tool that responds to lifestyle tweaks within months. Two clocks, two timescales, one goal.

You already know this person. The family member who leaves you hollowed out. The colleague who manufactures emergencies. The friend who takes and takes. New research puts molecular numbers behind what your nervous system has been telling you.

A study covered by The Washington Post this week found that "hasslers" — the researchers' term for chronically difficult people in your social network — are associated with elevated epigenetic markers of accelerated aging. Nearly 30 percent of people in the study sample had at least one. Researchers suggest stress physiology may be involved: chronic interpersonal friction was associated with elevated cortisol and inflammatory signals, which are linked to epigenetic changes that aging clocks detect.

This isn't soft wellness advice. In this study, who you spend time with was associated with accelerated cellular aging signals.

The corollary matters just as much: high-quality social connection appears protective. A University of Kansas study published this week found that counties with stronger civic engagement show better aggregate health outcomes — though the gains aren't equitably distributed across racial groups. And longitudinal work continues to show that midlife relationships powerfully predict late-life thriving, echoing the Harvard 85-year studies and Blue Zones research.

The practical question this raises and doesn't yet answer: is creating distance from a hassler as powerful as adding more positive connection? Worth tracking.

This study published in Science late last year, but it's having its viral moment now — 400 points on Hacker News this week — so let's be precise about what it does and doesn't show, because most headlines got it wrong.

Stanford researchers identified a protein called 15-PGDH that, in their data, increases with age by about twofold. They're calling it a "gerozyme" — an enzyme of aging — and it appears to suppress the regenerative capacity of cartilage-producing cells in joints. When they injected an inhibitor into old mice, cartilage regrew. Critically, the chondrocytes (cartilage cells) shifted their gene expression to a more youthful state without any stem cell involvement. That's mechanistically novel: it's not adding new cells, it's reprogramming the ones already there.

Human knee tissue collected during joint replacement surgeries also responded — samples began forming new, functional cartilage. A pill-based version is in early clinical trials for age-related muscle weakness.

Now the honest calibration: the preliminary Phase I human trial underway focuses on safety, not efficacy. No human trial has begun testing cartilage regeneration specifically. Additional animal model coverage this week — including rabbit early-damage models — helps explain the media spike but doesn't change the evidence tier. This is a genuinely exciting mechanistic finding, not a treatment. The bigger idea — that specific proteins accumulate with age and suppress tissue function, and that blocking them can restore it — is the concept worth watching.

If you've ever wished someone would just run a proper, large-scale trial to find out whether a longevity drug actually adds healthy years, the U.S. government nudged that future closer. ARPA-H announced up to $144 million in funding for projects aimed explicitly at extending healthspan — not treating single diseases.

The money isn't going to one moonshot. It's seeding platforms: multi-site clinical networks, standardized biomarkers (including epigenetic clocks like DunedinPACE), and modeling tools that simulate how slowing biological aging would impact frailty, healthcare costs, and independence over decades.

Separately, the Buck Institute launched Healthspan Horizons this week — a data platform designed to aggregate wearables, labs, clinical notes, and long-term outcomes into an AI-ready system for population-scale discovery. If ARPA-H builds the trial networks and Buck provides the observational baseline, the gap between "promising mouse study" and "we know this works in humans" could shrink dramatically.

For you, this is the infrastructure layer. It means future rapamycin, senolytic, or combination trials can be bigger, faster, and measured by outcomes that matter — walking speed, independence, time to first major disease — instead of just single biomarkers. The next thing to watch: which specific interventions get slotted into these platforms.

This landed on Hacker News with nearly 300 points, and it sits at the exact intersection of purpose science and aging biology that rarely gets covered well.

Stanford's Center on Longevity makes the case that purposeful engagement in later life isn't just good for morale — it's one of the most underestimated biological levers we have. Continued work, whether full-time, part-time, or project-based, is associated with better cognitive resilience, cardiovascular health, and overall longevity. Mental and social challenges stimulate brain metabolism and neuroplasticity the way exercise trains muscles. People with strong purpose scores show lower chronic inflammation and reduced risk of early death.

A Yale analysis reinforces this from a different angle: people with more positive expectations about aging were likelier to gain physical or cognitive function over a decade, not just decline more slowly. Beliefs about aging predict real biological outcomes.

Purpose isn't a luxury you earn after the health work is done. It may be part of the health work itself. The practical question for any reader: what does continued purposeful engagement look like for your next decade?