Which biomarkers should I actually track if I'm running a longevity peptide protocol — and which are overrated?

The short answer

You don't need a 200-marker panel. The high-value set is short: standard bloodwork (CBC, CMP, lipids, HbA1c, hsCRP, ApoB, fasting insulin, IGF-1, hormones, thyroid, vitamin D), annual DEXA, and one epigenetic clock — DunedinPACE specifically. Total cost in 2026: $400–$800 annually with reasonable insurance coverage on the standard bloodwork.

This piece walks through the biomarkers that actually predict longevity outcomes, what each tells you, how often to re-test, and which marketed measures are overrated.

For the broader picture see the Peptides for longevity (2026) cornerstone and the healthspan vs lifespan framing piece.

Evidence tier: 2 for standard bloodwork markers (decades of outcome literature); Tier 2 for ApoB (strong cardiovascular prediction); Tier 2 for DunedinPACE specifically (Belsky 2022 eLife methodology + replication); Tier 3 for the calendar epigenetic clocks (Horvath, GrimAge, PhenoAge — useful for trend but less responsive to intervention).

Standard bloodwork — the foundation

Evidence tier: 2 — established clinical biomarkers with extensive outcome literature.

The standard annual panel is more important than any "longevity-specific" add-on. It catches the conditions that actually kill people in middle age and beyond: cardiovascular disease, diabetes, kidney decline, liver disease, thyroid dysfunction, anemia, vitamin deficiencies.

Core annual panel:

• CBC — complete blood count; catches anemia, infection, occult blood loss
• Comprehensive metabolic panel (CMP) — kidney function (eGFR), liver enzymes (ALT, AST, GGT, alk phos), electrolytes, fasting glucose
• Lipid panel — total cholesterol, LDL-C, HDL-C, triglycerides
• HbA1c — chronic glycemic load
• Fasting insulin + fasting glucose — early dysglycemia (more sensitive than HbA1c)
• hsCRP — chronic inflammation
• TSH + free T4 + free T3 — full thyroid panel; TSH alone misses subclinical issues
• Vitamin D (25-hydroxy)
• Ferritin + B12 + folate
• Homocysteine — methylation pathway + cardiovascular risk

Sex-specific: - Men: total testosterone + free testosterone + SHBG (early morning draw) - Women: estradiol + FSH if perimenopausal; cycle-day-specific for younger - All adults: PSA (men over 50), Pap / HPV (women)

Optional, higher-yield for peptide users: - ApoB — atherogenic particle count; better than LDL-C alone (see below) - Lp(a) — inherited cardiovascular risk; check once lifetime - IGF-1 — baseline + monitoring if on GH-axis peptides - DHEA-S — adrenal output; weakly predictive of healthspan in some cohorts - Cystatin C — alternative kidney function marker, more sensitive than creatinine in muscular adults

For most readers in their 40s–60s, that's the full panel. ~$200–$500 with insurance reimbursement on most components when ordered for medication-management or preventive-care reasons.

ApoB > LDL-C for cardiovascular risk

Evidence tier: 2 — strong outcome literature; ApoB is the better atherogenic-particle marker.

ApoB measures the number of atherogenic particles (LDL + VLDL + IDL + chylomicron remnants combined). Standard LDL-C measures the cholesterol content of LDL particles only. The number of particles predicts cardiovascular events better than the cholesterol carried inside them (Sniderman 2019 JAMA Cardiol, PMID 31389986).

In ~25% of adults, ApoB and LDL-C diverge meaningfully — often in metabolic syndrome where LDL-C looks fine (small dense particles carry less cholesterol per particle) but ApoB is elevated (many small dense particles = high atherogenicity).

Practical translation: order ApoB instead of or in addition to standard LDL-C for any cardiovascular-relevant decision. Most clinical labs run it for $20–$50 add-on. Some preventive cardiologists now use ApoB as the primary lipid target rather than LDL-C.

Body composition — DEXA annually

Evidence tier: 2 — gold standard for body composition; outcome literature on lean mass / visceral fat / bone density is extensive.

DEXA (dual-energy X-ray absorptiometry) scans give you three measurements that matter for longevity:

• Lean mass (regional + total) — the strongest single predictor of late-life function and fall risk
• Visceral fat — the inflammation-driving fat depot; better predictor of metabolic risk than waist circumference
• Bone density (T-score and Z-score) — osteoporosis risk; relevant for both sexes from 50 onward

Annual DEXA costs $100–$250 cash; not always insurance-covered for screening. The trend over time is more informative than a single snapshot.

Surrogate measures if DEXA is unavailable: - Grip strength — proxies overall lean mass, predicts late-life mortality - Waist-to-height ratio — rough visceral fat proxy - Bioimpedance scales — useful for trend tracking, less accurate for absolute values

Epigenetic clocks — DunedinPACE is the one to use

Evidence tier: 2 for DunedinPACE specifically; Tier 3 for the calendar clocks (Horvath, GrimAge, PhenoAge).

Epigenetic clocks measure DNA methylation patterns that change predictably with age. There are four major classes:

1. First-generation calendar clocks (Horvath, Hannum) — predict chronological age from methylation. Useful for proof-of-concept; less useful for personal tracking because they're tuned to predict chronological age, not biological age.

2. Second-generation calendar clocks (PhenoAge, GrimAge) — predict biological age and mortality risk; better outcome literature than Horvath. GrimAge has the strongest mortality prediction in published cohorts.

3. DunedinPACE — measures rate of aging (the pace per chronological year) rather than current biological age. Developed in the Dunedin longitudinal cohort. The most-replicated rate-of-aging measure as of 2026 (Belsky 2022 *eLife*, PMID 35029144).

4. Glycan-based clocks (GlycanAge) — measures aging via IgG glycosylation patterns. Captures inflammation-aging differently than methylation clocks. Some evidence it responds to intervention faster.

Practical guidance: - Order DunedinPACE if you're running interventions and want a single rate-of-aging marker that moves on intervention timescales. - Optional: pair with PhenoAge or GrimAge for biological-age trend lines. - Skip GrimAge-only or Horvath-only providers — they're using superseded tools. - Available 2026 providers: TruDiagnostic (includes DunedinPACE), Elysium (Index test), several specialty longevity clinics. - Cost: $250–$500 per test in 2026. - Re-test frequency: every 6–12 months. DunedinPACE specifically moves on faster timescales than calendar clocks.

Honest caveat: the epigenetic-clock literature is evolving fast. The clock you use in 2026 may not be the recommended clock in 2028. Treat the current best practice as best-available, not final.

Cognitive markers — underused

Evidence tier: 2 — cognitive testing has extensive outcome literature; the longevity community under-tracks it.

Most longevity-protocol tracking focuses on metabolic and body-composition markers. Cognitive performance is at least as important for healthspan but gets less measurement attention because the tools are less obvious.

Useful self-tracking cognitive measures:

• Cambridge Brain Sciences (or similar) — short battery of working memory, executive function, processing speed
• Stroop test — executive function / cognitive flexibility
• Symbol-digit substitution — processing speed; sensitive to subclinical cognitive change
• Verbal fluency — sit, say as many words starting with a letter as you can in 60 seconds; tracks executive function

Run quarterly for trend. Cognitive markers can identify subtle decline 3–5 years before clinical symptoms; catching it early opens intervention windows.

Sleep architecture — wearable-tracked is good enough

Evidence tier: 2 — sleep / mortality association well-established; wearables have validated against polysomnography for stage-level tracking.

Sleep tracking via Oura, Whoop, Apple Watch, Garmin etc. has matured to the point that wearable-tracked metrics are useful for trend monitoring even if absolute values diverge from gold-standard PSG. The metrics worth tracking:

• Total sleep time — target 7–9 hours adult
• Sleep efficiency — time asleep / time in bed; target >85%
• Slow-wave sleep proportion — GH release happens here; declines with age but is intervention-responsive
• REM proportion — cognitive consolidation
• Resting heart rate during sleep — long-term trend predicts cardiovascular risk
• HRV trend (not daily — weeks-to-months trend)

Sleep changes under almost every meaningful longevity intervention (training, GH-axis peptides, GLP-1, diet shifts). Track it.

Wearable metrics — useful and overrated

Evidence tier: 3 — wearables are good for trend, less reliable for absolute thresholds.

Useful wearable metrics: - Resting heart rate (RHR) — clinically meaningful; tracks fitness + recovery + illness - HRV trends over weeks (not days) — autonomic balance - VO2 max estimates — from Garmin, Apple Watch with cardio data; reasonable trend accuracy - Step count + active time — for activity volume baseline

Less useful: - Daily "recovery scores" / "readiness scores" — useful as summary, but you can derive most of it from the underlying metrics - Continuous glucose monitoring (CGM) for non-diabetics — usually drives anxiety, not action; useful only if dysglycemia is suspected - "Stress scores" from wearables — broadly correlate with HRV; track HRV directly

The general rule: wearables for trends, lab tests for absolute values. Both have a role.

Retest cadence

Evidence tier: 2 — derives from established preventive-care frequency recommendations adapted for longevity-protocol monitoring.

How often to repeat each test:

| Test | Default cadence | Adjust when | |------|------------------|-------------| | Standard bloodwork | Annual | Every 3–6 months when adjusting protocol or after a new intervention | | DEXA | Annual | Skip a year if no body composition changes targeted | | Fasting insulin / glucose | Annual within CMP | Every 3 months if running GLP-1 or dysglycemia present | | ApoB | Annual | Every 6 months if on lipid-modifying therapy | | IGF-1 | At baseline + every 3 months if on GH-axis peptides | Otherwise annual | | DunedinPACE | Every 6–12 months | Every 6 months if running active intervention | | PhenoAge / GrimAge | Every 12–24 months | Slower-moving; annual is enough | | PGx panel | Once lifetime | Genotype doesn't change | | GlycanAge | Every 12 months | Optional; complement to methylation clocks | | Cognitive testing | Quarterly self-administered | If subjective change, see a neurologist |

Over-testing wastes money and can drive over-intervention. The goal is informing decisions, not chasing perfect numbers.

What to skip

Evidence tier: 4–5 — marketing-driven measures with weak outcome literature.

• "Telomere length" consumer tests — high measurement noise, not actionable, marketing-driven
• Generic "longevity panels" at premium pricing — usually a standard bloodwork panel + a wellness-themed interpretation layer at 3–5x the clinical cost
• Microbiome sequencing for "gut health optimization" — interesting science, no validated intervention algorithm yet
• CGM monitoring for non-diabetic adults — usually drives anxiety, not behavior change
• Hair-mineral analysis — clinically not validated
• Food sensitivity IgG panels — discredited methodology, fixed cost, no clinical action
• HOMA-IR as a primary measure — calculate-derived from fasting glucose + insulin; both components are more informative than the derived ratio

If a service offers a "100-marker longevity panel" for $1500+, it's almost always overdelivering on markers and underdelivering on actionable interpretation. Cheaper, smaller, clinically-grade testing beats premium consumer panels.

Limitations

This is biomarker guidance, not personalized medical advice.

• Biomarker interpretation needs clinical context. A single elevated marker isn't actionable without trend lines and concurrent medical context.
• Over-testing has costs — financial, but also the cost of acting on noise rather than signal.
• Reference ranges vary by lab. What's "high" at one lab may be "normal" at another. Track trends within one lab when possible.
• Some biomarkers move under intervention more than others. Set realistic expectations — DunedinPACE moves on months, GrimAge moves on years.
• Marko Maal, MSc Pharmacy reviewed this article. Reviewer attribution does not constitute a doctor-patient relationship.

The bottom line

The high-value biomarker panel is shorter than premium-longevity marketing suggests. Standard annual bloodwork + DEXA + DunedinPACE covers most of what you need to inform a peptide protocol. ApoB is the upgrade most readers should make on standard lipid testing. The rest — calendar epigenetic clocks, GlycanAge, cognitive self-testing, wearable trends — fills out the picture but isn't the foundation.

Order the panel through your physician where insurance helps. Skip premium "longevity testing" services that charge multiples for the same underlying tests with a wellness-themed report wrapped around them.

Related on this site

• Peptides for longevity (2026) cornerstone
• Healthspan vs lifespan peptides
• Senolytics vs senomorphics
• Pharmacogenomics + peptide therapy cornerstone
• Personal genomics services compared 2026
• Should I get a PGx panel before peptides?
• Longevity pillar hub
• Finnrick vendor testing

References

• Belsky DW, Caspi A, Corcoran DL, et al. 2022. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 11:e73420. PMID 35029144 — DunedinPACE methodology; primary reference for rate-of-aging measure.
• Sniderman AD, Thanassoulis G, Glavinovic T, et al. 2019. Apolipoprotein B Particles and Cardiovascular Disease: A Narrative Review. JAMA Cardiol. 4(12):1287-1295. PMID 31389986 — ApoB > LDL-C evidence.
• Levine ME, Lu AT, Quach A, et al. 2018. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 10(4):573-591. PMID 29676998 — PhenoAge methodology.
• Lu AT, Quach A, Wilson JG, et al. 2019. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging (Albany NY). 11(2):303-327. PMID 30669119 — GrimAge mortality prediction.
• Studenski S, Perera S, Patel K, et al. 2011. Gait speed and survival in older adults. JAMA. 305(1):50-58. PMID 21205966 — functional-marker mortality prediction.
• Krištić J, Vučković F, Menni C, et al. 2014. Glycans are a novel biomarker of chronological and biological ages. J Gerontol A Biol Sci Med Sci. 69(7):779-789. PMID 24325898 — GlycanAge methodology.
• López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. 2023. Hallmarks of aging: An expanding universe. Cell. 186(2):243-278. PMID 36599349 — broader aging-biomarker framework.