Hi {{first name | there}},

Every few weeks, a patient sends me a screenshot of a biological-age result and asks, very reasonably, whether they should change their life or just be offended.

The number is usually two things at once: flattering enough to share on socials and too blunt to practice good medicine with.

One number cannot tell you whether the arteries are degrading, the liver is reacting, the immune system is smoldering, or the kidney, the organ nobody puts on a vision board, is setting the chemistry of the room.

The evidence is now strong enough to stop pretending biological age is just one tidy number.

A 50-year-old can have an artery clock that reads 60, a liver clock that reads 45, and a brain clock that reads 40. These are model estimates, not accurate forecasts, but the pattern keeps showing up across large plasma-proteomic, imaging, and multi-omics cohorts.

The practical question is not, "Am I young?" (or "Am I beating all my friends?")

It is, "Which system is vulnerable, and is it pulling on the rest?"

Most bio-age tests are snapshots: take blood once, estimate organ ages, then ask who develops disease later.

But while a snapshot can tell you something looks off, it cannot tell you whether the system is accelerating, stabilizing, or responding to an intervention.

That is why the new Neirynck 2026 bioRxiv preprint is interesting. The authors measured plasma proteomics in 1,250 middle-aged adults, then measured them again ten years later.

They watched the clocks move. What they found:

1. Organ aging was dynamic, but not random. People who looked like fast agers at baseline often still looked like fast agers ten years later. \

2. Organs aged partly in parallel. They were not independent scorecards. Aging in one organ tended to travel with aging elsewhere.

3. Immune and adipose systems sat near the center of the network. This fits what we already know from inflammaging, senescence, visceral fat, adipokines, insulin resistance, and metabolic-immune crosstalk. Adipose tissue is not passive storage. It is a hormonal group chat with occasional arson.

4. Early cardiorespiratory aging predicted later metabolic aging. The heart, arteries, and lungs did not just report on the system. They seemed to foreshadow where the metabolic system was going.

5. In women, menopause looked like a multi-organ transition. In that preprint, the menopausal transition was associated with acceleration across multiple organs at once. A separate 2026 structural-aging map of 25,306 postmortem biopsies across 40 tissue types also found later-life acceleration in uterus and vagina around menopause, with sex-hormone signaling linked to coordinated cross-organ deterioration.

The question is no longer just, "Which organ looks old today?"

It is, "Which systems are moving together, and which one is changing the trajectory?"

A few years ago, organ aging sounded like a very expensive way to create new anxieties. You thought your birthday was rude. Now your spleen wants a separate opinion.

But the data is becoming clearer with every research publication.

Stanford, 2023: Tony Wyss-Coray's group showed in Nature that plasma proteins could estimate the biological age of 11 organs in 5,676 people across five cohorts. Nearly 20% had strongly accelerated aging in one organ. About 1.7% were accelerated across multiple organs. Accelerated organ aging carried 20% to 50% higher mortality risk. Heart aging tracked with heart-failure risk. Brain and vascular aging predicted Alzheimer's progression independent of pTau-181.

Whitehall II, 2025: A Lancet Digital Health study followed 6,235 adults through roughly 20 years of national health records. Organ age gaps predicted 30 age-related diseases. Most associations were not even same-organ. An old organ often marked broader systemic vulnerability.

UK Biobank, 2025: In 44,498 people, plasma proteomics linked brain and immune-system aging with healthspan and longevity. A youthful brain and immune system were especially associated with longevity, while an aged brain carried a large Alzheimer's-risk signal in some analyses.

Cross-population validation, 2026: A Nature Aging paper built organ-specific proteomic clocks in 43,616 UK Biobank participants and validated them in cohorts from China and the US. Accelerated organ aging predicted disease onset, progression, and mortality beyond clinical and genetic risk. Brain and artery clocks were tightly linked to cognitive decline and dementia biology.

Genetics, 2025: The genetics tells the same story. A Nature Communications analysis mapped 119 loci to organ-specific aging, with 27 shared across multiple organs and 554 prioritized risk genes. Translation: organs have their own biology, but the body did not draw the borders as neatly as our medical specialties did.

The point isn't "your liver is older than your brain," even though that is the headline everyone wants.

The stronger point is - one organ system aging is rarely just a local problem. It is often a window into the broader aging picture.

This is where the conversation gets sloppy.

People want to know the most important organ. That is like asking which instrument matters most when the concert hall is on fire.

It all depends on what you care about: mortality, dementia, cardiovascular disease, metabolic disease, frailty, medication response, or the next test to order.

Here is the hierarchy I would actually use. I do not use this as a strict ranking. I use it as a triage map: which system changes the next question, the next lab, or the next intervention?

Vascular and cardiorespiratory systems: the prevention leverage layer. Blood vessels age early. A 2025 human tissue proteome atlas studied 516 samples from 13 tissues across five decades and found vascular tissue among the earliest-aging and most susceptible. The structural-aging preprint above also saw vascular acceleration as early as the 30s. Clinically, this maps to ApoB, blood pressure, endothelial function, arterial stiffness, sleep apnea, VO2 max, insulin resistance, and inflammation.

Immune system: the hub layer. Immune aging keeps showing up as a coordinator, mortality signal, and cross-organ amplifier. This is not one lab value you "fix." It is infections, autoimmunity, chronic activation, senescence surveillance, sleep, stress physiology, visceral fat, and metabolic inflammation all talking at once.

Adipose and metabolic system: the propagation layer. Visceral fat is not a storage unit. It is endocrine, inflammatory, immune-active tissue. For most patients, waist circumference, liver fat, fasting insulin, triglycerides, A1c, exercise capacity, and body composition tell me more than a glossy biological-age PDF.

Brain: the high-stakes sentinel. Brain age is repeatedly linked to mortality, Alzheimer's risk, cognitive decline, and longevity. But brain aging is often body aging wearing a neurologic costume. A 2026 medRxiv preprint found that accelerated multi-organ aging was detectable decades before dementia onset, with brain, heart, and muscle aging acting synergistically. tldr; brain age follows systemic age.

Kidney and liver: the internal environment. They are not the flashiest organs. But, they are the chemistry of the whole house. An eight-decade life-course study found that liver, kidney, and immune aging mediated links between early-life exposures and later mortality. Mechanistic human and mouse work also suggests the kidney may contribute to systemic aging through sex-specific glycosphingolipid and uremic-toxin biology.

Muscle: under-counted by many blood clocks. Plasma proteomics can miss what strength, power, VO2, gait speed, and body composition see directly. Muscle is not just for show. It is glucose disposal, frailty prevention, immune-metabolic health, and the difference between recovering from a stressor and not.

If your organ-age report does not change what you investigate next, it is mostly an expensive horoscope.

If you have ever bought an epigenetic test, a proteomic test, and an MRI-based test and gotten three different answers, the charitable interpretation is not that one company is lying.

It is that they are measuring different layers of the same organism.

A 2025 paired analysis of protein- and imaging-based clocks across eight organs found that molecular and structural clocks have distinct phenotypic and genetic signatures. They may be reading different stages of aging. They still often converge on the same disease pathways.

A large Nature Medicine MRI-clock paper built seven imaging-derived organ age gaps in 313,645 people and linked them to proteins, metabolites, variants, other aging clocks, disease endpoints, mortality, and cognitive trajectories.

A 2026 Aging Cell framework makes the caveat explicit: single-omics clocks can miss cross-omics biology and carry platform-specific bias.

Methylation, chromatin, proteins, metabolites, MRI structure, histology, and clinical function are not competing for the same crown. They are reading different chapters.

So no, disagreement does not automatically mean one test is wrong.

It does mean you should probably stop shopping for the lowest number 😅.

This is not hypothetical for me. I have had several biological-age readouts that were reassuring: one multi-omic clock put me a few years younger than my calendar age, one organ-system clock put most systems meaningfully younger, my GlycanAge improved over five months, and my DunedinPACE suggested a slower recent pace of aging.

But a different platform, collected earlier, flagged respiratory and immune aging as the weaker signals. A functional composite told a similar story in a different language: grip strength looked excellent, but FEV1 was the lagging piece.

That does not mean my lungs are “old” in any literal sense. In fact it pointed me toward looking at whether my almost-asthma was maybe acting up, or if my bout of COVID had a greater impact than expected.

The same lesson showed up in the cardiovascular lane. My old lipid markers were the part of my dashboard I took most seriously: ApoB was previously 128 mg/dL, and LDL-C had been as high as 192 mg/dL. After treatment, ApoB moved to 45 and LDL-C to 37 (and now my vascular age is super young).

I do not interpret that as “my arteries got younger.” I interpret it as: the intervention changed a causal risk factor, which is an important difference.

This is the part people want to jump to: can we make the organ clocks younger?

Carefully: sometimes we can move them. That is not the same as proving rejuvenation.

The cleanest human intervention signal so far comes from the 2026 CALERIE analysis. In 185 adults from the CALERIE Phase 2 randomized trial, two years of sustained caloric restriction slowed organ-age increases compared with eating freely.

The effect was not uniform: Metabolic and cardiovascular clocks moved the most. Whole-body and immune followed. Liver moved modestly and only at 24 months. Kidney did not budge.

A separate NHANES analysis of 4,890 adults found that moderate eating frequency and fasting duration were associated with lower organ-age indices and better health metrics. Extremes looked worse. But this was cross-sectional. It does not prove time-restricted eating rejuvenates organs.

Medication effects are the other trap.

In the Neirynck preprint, medication initiation altered clocks through specific drug-targeted proteins such as renin and APOB. That matters. If a patient starts a statin, ApoB drops, and an artery-related clock improves, that may be excellent medicine. Celebrate the risk reduction. But do not put a tiny birthday hat on the artery. The model may be responding to a protein it weights heavily, which isn't proof of rejuvenation.

Sometimes the drug moved the patient. Sometimes it moved one of the model's favorite inputs. Those are not the same.

Most patients encounter organ clocks through a consumer test, a clinic-grade SomaScan or Olink panel, an epigenetic report, or a whole-body MRI platform.

The rules are the same.

Use clocks as triage, not as an ego score. The point is not to be 42 (or 24) biologically. The point is to find a vulnerable system before it becomes disease.

Ask what changes if the clock is high. If nothing changes, the test is mostly entertainment.

Here is how I would translate the common patterns:

If liver age is high: look at alcohol, metabolic health, ALT, AST, GGT, ferritin, fatty-liver risk, medications, viral hepatitis when appropriate, and insulin resistance. The next step is not panic-buying a supplement powder with a name like MitoLiver Ultra Max. It is a real workup.

If artery or heart age is high: do a serious cardiovascular assessment. ApoB, BP, lipid particles, hs-CRP, Lp(a), sleep apnea screening, exercise capacity, family history, and coronary calcium when appropriate.

If lung or respiratory age is high: look at smoking/vaping, air or occupational exposures, asthma/COPD symptoms, sleep apnea, recurrent infections, reflux or aspiration risk, and VO2. If the signal repeats or matches symptoms, get spirometry instead of buying a lung-detox supplement.

If brain age is high: do not jump to a dementia diagnosis. Look at vascular risk, sleep, inflammation, metabolic health, hearing, alcohol, mood, medications, and cognitive symptoms. Cognitive testing belongs where there is an actual cognitive question.

If immune age is high: take infections, autoimmunity, chronic inflammation, sleep, stress physiology, visceral fat, and metabolic dysfunction seriously.

If kidney age is high: do not ignore it because creatinine is "normal." Look at eGFR trends, cystatin C when useful, urine albumin-to-creatinine ratio, BP, glycemic status, medication exposures, and cardiorenal risk.

If multiple organs are high at once: do not assume the report is broken. Multi-organ acceleration can reflect a life course showing up in protein: adversity, smoking, untreated metabolic disease, inflammation, kidney-liver-immune burden, sleep disruption, or medication complexity.

Pair clocks with boring physiology. If the clock and the labs disagree, the labs usually win. If they agree, your plan just got easier.

This is the rule I use with my own data: a respiratory-age signal only matters if it points me toward spirometry, exposures, asthma/allergy context, or sleep evaluation. An immune-age signal only matters if it points me toward direct immune testing, infection history, sleep, stress physiology, or inflammation. A cardiovascular signal only matters if it maps to ApoB, blood pressure, imaging, family history, and exercise capacity.

Track pace, not just snapshots. If you repeat testing, look at trajectory. A single organ-age estimate is interesting. Direction over time is more clinically useful.

Do not compare platforms like a leaderboard. A protein-based liver age and an MRI-based liver age are not the same measurement. Picking your favorite number is not science. It is vibes with a login portal.

Turns out I have a way old video on this topic! Our newsletter today is definitely more up to date, but there are still nuggets in here for those who want to go beyond the basics.

Watch on YouTube | Subscribe to The Longevity Show

A short list that almost always pays off, with or without a fancy clock:

If a plan is working, at least some of these should move.

The whole reason to measure the boring stuff is to protect the exciting stuff.

🫘 GLP-1/GIP cardiorenal: In SURPASS-CVOT's kidney analysis of 13,165 adults with T2D and ASCVD, tirzepatide had fewer composite major kidney events than dulaglutide over four years, but the signal was albuminuria-heavy and comparator-specific.

🧠 GLP-1 brain claims: A meta-analysis of 14 RCTs in non-diabetic Parkinson's, Alzheimer's, and MCI found a tiny cognition signal with only a 1% probability of clinically important benefit, which is a useful cold shower for disease-modification hype.

💊 Organ context changes drug reads: In adults with T2D after AKI requiring dialysis, SGLT2 inhibitors were associated with lower dementia and adverse kidney-event risk than GLP-1 RAs, which is interesting and very much not permission to generalize across all patients.

🦠 Bariatric surgery may rewire the ecosystem: A Nature Metabolism paper linked T2D remission after surgery to greater microbial gene richness and metabolic potential for butyrate, fermentation, and methanogenesis.

🚺 PCOS may soon be PMOS: A Lancet consensus from 56 organizations and 14,360 survey respondents proposed renaming polycystic ovary syndrome to polyendocrine metabolic ovarian syndrome, which is overdue because PCOS was never really about cysts.

The future of longevity medicine is not a single rejuvenation drug or a prettier biological-age number.

It is systems diagnosis.

Find the vulnerable node. Ask what is upstream, what is downstream, and what is propagating through the network. Then intervene on the things that map cleanly to real physiology: blood pressure, lipids, insulin resistance, visceral fat, aerobic capacity, muscle, sleep, inflammation, kidney function, liver fat, medication burden.

Treat an organ clock as a hypothesis.

Treat a clock that moved after a medication as a signal worth investigating, not a victory lap.

And please, if your biological-age app says your spleen is thriving but your ApoB is 140, do not let the spleen write the care plan.

If you want help turning organ-clock results, labs, and risk factors into a plan instead of an anxiety spiral, this is the work. We do it across longevity, cardiometabolic, immune, and cognitive risk with the same lens: which system is most vulnerable, and what changes the trajectory?

Keep learning,

Hillary Lin, MD

Co-Founder & CEO, CareCore

Was this useful? Forward The Longevity Letter to someone who has bought a biological-age test and is not sure what to do with it.

Learn more about my practice here: hillarylinmd.com/longevity-practice

The Aging Code Summit, May 26-27, Cambridge, MA. I am speaking Wednesday, May 27, at Longevity Global Boston's summit on moving aging science from discovery into interventions.

NYC Tech Week: "Sick Care is Dead. What Comes Next?", June 2, New York, NY. A brunch panel on the consumer health stack: GLP-1s by text, gyms as longevity clinics, AI triage, and what gets lost or gained when health moves outside traditional institutions.

NYC AI x Longevity Summit, June 4-5, Long Island City, Queens. I am on the AI in Clinical Practice panel on June 4 with Longevity Global.

Dry Eye Society of the Americas Conference, July 10-11, New York, NY. A clinician-focused dry eye and ocular surface meeting.

Science of Skin Summit, September 17-20, Austin, TX. Speaking on AI in dermatology and skin and hair as windows into biological age.

Livelong Women's Health Summit NYC, September 25-26, New York, NY. A two-day summit on women's health, longevity, and agency. My role is still being finalized, but it is on the fall calendar.

A few reader-supported partners that fit the longevity stack:

If you want physician-guided biological age testing, start on the testing page. It includes epigenetic age testing with consultation, plus organ-specific proteomic and SystemAge assessments when they fit the clinical question.

Have a question or topic you want covered? Reply to this email.

Was this forwarded to you? Subscribe here to get The Longevity Letter in your inbox.