How Advanced Biomarkers Can Predict Disease YEARS Before Symptoms

What if you could peek into your body's future and predict diseases years before symptoms appear? In this video, longevity physician Dr. Hillary Lin share…

Published December 29, 2024The Longevity Show

Conventional panels are designed to catch disease that has already started. Dr. Lin's framing is that by the time a standard lipid panel or fasting glucose flags something, you are past the window where cheap, non-invasive changes would have worked and into the territory of medications and procedures. The question the episode asks is what you would test if the goal were to see the trajectory decades early instead of confirming the diagnosis on schedule.

The structure moves in three tiers. First, how to read the conventional panel you already have — including the details most people never get explained, like what a low versus high MCV is pointing at, or why creatinine misleads in anyone with real muscle mass or a creatine habit. Then the advanced tier: fasting insulin, hs-CRP, ApoB and Lp(a), homocysteine, a properly timed hormone panel, omega-3 index, vitamin D, and body composition. Then the frontier — genome sequencing, microbiome profiling, heavy metal screening, oxidative stress and mitochondrial assays, CIMT and CAC.

The most useful running argument is about reference ranges. Lab 'normal' is calibrated against the average American, who is frequently overweight and often carrying fatty liver — so a liver enzyme inside the reference interval may be normal for someone with a disease you would rather catch. Dr. Lin's approach is to hold tighter optimal ranges than the lab prints. She is equally clear about where the frontier is still soft: biological age testing she calls unproven and confusing and defers to its own episode, several mitochondrial assays she says need further validation, and she notes she rarely needs advanced inflammatory markers because hs-CRP and ESR usually answer the question.

Before you watch

Reference ranges encode the population they were built on. Because AST, ALT and most other blood values are benchmarked against the average American — a population where overweight and fatty liver are common — a result inside the 'normal' band can be normal for someone with early disease. Dr. Lin's practice is to work to tighter optimal ranges than the lab reports in order to intervene earlier.
Creatinine is a bad kidney marker in exactly the people most likely to be tracking their health. It rises with muscle mass and with creatine supplementation, not only with impaired filtration — so Dr. Lin orders cystatin C instead for patients who are physically fit or supplementing creatine for muscle or cognitive benefit.
The CBC contains detail almost nobody has explained to them. A low MCV points toward iron deficiency; a high MCV points toward B12 or folate deficiency, which affects energy, nerve health and mood. On the white cell side, high neutrophils suggest bacterial infection, high lymphocytes suggest viral, and eosinophils rise with allergy or, rarely, parasitic infection.
Fasting insulin is the single test Dr. Lin says she runs on everyone, because it is the earliest signal of insulin resistance and can precede type 2 diabetes by many years. The specific pattern she is looking for is high fasting insulin with still-normal fasting glucose and HbA1c — the window where lowering refined carbohydrate intake, adding resistance training or trialling intermittent fasting is often enough on its own.
Lp(a) is genetically determined, meaningfully raises cardiovascular risk independent of every other cholesterol number, and appears on no conventional lipid panel — you have to ask for it, once in your lifetime. Dr. Lin describes a 35-year-old who ate clean, trained daily and had a broadly normal LDL, whose Lp(a) came back far above range and who is now on lipid-lowering treatment decades before he would otherwise have been.
Elevated homocysteine damages vessels through four mechanisms — endothelial dysfunction, oxidative stress, reduced nitric oxide availability, and abnormal vascular smooth muscle proliferation — and is linked to heart attack, stroke and Alzheimer's risk. It usually reflects B6, B12 or folate metabolism rather than intake, which is why the MTHFR polymorphism matters: Dr. Lin describes a patient with a strong dementia family history whose B vitamin blood levels were unremarkable while homocysteine was high, a pattern that itself points to a methylation problem and to methylated B vitamin forms.
SHBG is diagnostic, not a target. It binds and transports sex hormones, so high levels reduce how much testosterone is actually available. Dr. Lin is asked how to lower it and her answer is that you cannot usefully — the routes that lower SHBG are malnutrition, low protein intake and poor liver health. It is a clue explaining why hormone levels look the way they do, not something to intervene on.
The omega-3 index is to omega-3 status what HbA1c is to glucose: a red-cell measure of long-term status rather than a snapshot. Its practical value is dose guidance — without it there is no way to know whether a given person needs one gram or four to five. Dr. Lin describes a patient with unexplained mood swings and concentration problems whose index was extremely low, and whom she pushed to 4g a day alongside more fatty fish.
DEXA ordered at a conventional radiology office will only give you bone density, because conventional medicine cares about osteopenia and osteoporosis. The longevity use is the rest of it — lean muscle mass and visceral fat, the fat around the organs that drives cardiovascular risk. Home body composition scales, in Dr. Lin's experience, can be off by 10% on body fat, which makes her doubt everything else they report.
The CAC score's blind spot is the point. It measures calcified plaque, which is the more stable and less dangerous kind — once plaque has hardened and mineralised it is less likely to dislodge. The unstable, soft plaque that actually causes events requires CT angiography to see, with companies like Cleerly applying AI analysis to those scans. Dr. Lin's own disclosure here is direct: she had a carotid intima-media thickness test done at a young age because of her strong family heart disease history, and it came back positive for early atherosclerosis.
Oxidation is the step that makes trapped lipid dangerous. A lipid particle in the artery wall is not the problem in itself — it is when the particle oxidises that macrophages are recruited, become foam cells, and form plaque. That is why the oxidative stress markers exist: 8-OHdG for DNA damage, MDA and F2-isoprostanes for lipid peroxidation, protein carbonyls, AOPP and nitrotyrosine for protein oxidation, plus oxidised LDL and myeloperoxidase for the vascular version specifically.

Chapters

01:12

Bonus Health Tests: Elevating Your Health to the Next Level

Questions

What is the single most useful test that isn't on a standard panel?

Fasting insulin, by Dr. Lin's account — she runs it on everyone. Fasting glucose tells you whether blood sugar control has already broken down; fasting insulin tells you how hard your body is working to keep it looking normal, which is the earliest available signal of insulin resistance and can lead type 2 diabetes by many years. The pattern worth catching is high fasting insulin alongside normal fasting glucose and HbA1c, because at that stage lowering refined carbohydrates, adding resistance training or trialling time-restricted eating is frequently enough to reverse it. She describes it as the most satisfying point to intervene, precisely because nothing has broken yet.

Why does my doctor say my labs are normal when I still feel unwell?

Partly because 'normal' is a population statistic, not a health standard. Reference ranges are built against the average American — a population in which overweight and fatty liver are common — so a value inside the printed range can be perfectly typical for someone with early disease. Longevity practice narrows those ranges deliberately to catch trajectories rather than diagnoses. The other half of the answer is that standard panels omit the markers that would explain the symptoms: fasting insulin, hs-CRP, ApoB, Lp(a), homocysteine, omega-3 index and a properly timed hormone panel are all absent from a routine annual check.

What is Lp(a) and why isn't it on a standard lipid panel?

Lipoprotein(a) is a lipoprotein whose level is largely set by genetics and which independently raises cardiovascular risk regardless of what the rest of your cholesterol looks like. It is not included in conventional lipid panels, so you have to request it specifically — and because it is genetically determined and stable, testing it once in a lifetime is generally enough. Dr. Lin's case for that single test is a patient in his mid-thirties with a broadly normal LDL, a clean diet and daily exercise, whose Lp(a) came back far above range. Finding it meant starting lipid management decades earlier than any other signal would have prompted, which matters because time spent with untreated risk is time arteries spend accumulating plaque that current tools cannot easily reverse.

When should hormone tests be done during the menstrual cycle?

Timing is what makes the results interpretable. Estradiol is measured in the early follicular phase, around day 3, counting day 1 as the first day of your period — anywhere from day 2 to day 5 or 6 is generally acceptable. Progesterone is measured in the luteal phase, around day 21 of a 28-day cycle, again with some tolerance since not everyone runs a 28-day cycle. FSH is drawn on day 3 as a signal of ovarian reserve, alongside AMH, which matters for anyone thinking about conception within the next decade. Testosterone and DHEA can be drawn at any point, but both free and total testosterone are needed, since free reflects what is actually available rather than bound to SHBG. For genuinely confusing pictures Dr. Lin recommends continuous testing across a full cycle or two, often via at-home urine testing at around $300 for the month.

Should I get full genome sequencing or is 23andMe enough?

Dr. Lin's analogy is that genotyping looks up the specific words you care about while sequencing reads the whole dictionary. Genotyping — the 23andMe approach — is cheaper, faster and covers the genes with the strongest supporting data. Whole genome sequencing runs roughly $2,000–3,000, produces a volume of information that is genuinely overwhelming, and includes regions whose significance nobody yet understands. Her view is that sequencing earns its cost mainly for people who are doing everything right and still not getting results — where an unusual variant might explain it — and that it should be interpreted with a physician. For most people, targeted genotyping of APOE, MTHFR, BRCA1 and BRCA2, PARK2, COMT, FTO and ACE delivers most of the actionable value without the cost or the noise.

How reliable are advanced tests like biological age, mitochondrial function and microbiome panels?

This is where Dr. Lin is most explicitly cautious. She calls biological age testing 'still very unproven and confusing' and declines to fold it into her recommendations, noting only that PhenoAge can be derived from routine labs while telomere length, inflammaging and metabolic approaches all exist and disagree. Several mitochondrial assessments — high-resolution respirometry, membrane potential, metabolite panels, FGF21 — she says require further validation. And on advanced inflammatory markers like IL-6, TNF-α and F2-isoprostanes, she notes she rarely needs them because hs-CRP and ESR usually suffice. Her broader caution applies to this whole tier: get these interpreted by a physician experienced with them, because a single out-of-range result can look alarming or be actively misleading on its own.

The Longevity Letter