Combining Supplements for Longevity - What Does the Evidence Say? | Kamil Pabis

#1

Gemini Pro AI Summary and Analysis of the Video:

Here is the corrected summary and analysis of the provided transcript, reflecting the correct speaker identity.

Combinatorial Pharmacotherapy & Supplementation for Longevity

A. Executive Summary

This presentation by Kamil Pabis (Research Fellow at NUS Singapore, working with Prof. Brian Kennedy) addresses the controversial topic of “stacking” supplements and drugs for lifespan extension. Pabis challenges the prevailing medical skepticism that polypharmacy is inherently toxic, arguing instead that dietary patterns (vegetables + exercise) naturally expose us to hundreds of bioactive compounds without catastrophic failure, suggesting the body can handle complexity.

The core thesis is one of “Additivity with Diminishing Returns.” Citing animal data (worms, flies, and mice), Pabis demonstrates that combinations generally yield better lifespan extension than single agents, though rarely synergistic. He highlights historical data from Stephen Spindler showing that while most supplement stacks are neutral, a specific “Algae Combo” was toxic, underscoring the risk of untested formulations.

Pabis presents original human data from a cohort of 4,000 healthy individuals: while taking some supplements (1–4) correlated with lower biological age compared to zero, taking more (5+) offered no additional benefit (a plateau effect). He advocates for a “Game Theoretic” approach to selection: prioritize compounds that exist at the intersection of animal lifespan data (e.g., ITP results) and human clinical utility (e.g., blood pressure reduction), such as Cocoa Powder or Rapamycin.

B. Bullet Summary

  • Risk/Reward Volatility: Some longevity interventions may increase early-life mortality risk while extending maximum lifespan (high volatility), forcing users to choose between safety and “escape velocity” potential.
  • The “Dietary Bioactive” Argument: The speaker argues that since a standard healthy diet (vegetables + meat) exposes the liver to thousands of compounds daily, the fear of “stacking” 5–10 supplements is likely overstated.
  • Diminishing Returns: In a human cohort of 4,000, users taking 1–4 supplements had lower biological ages than non-users, but users taking 5+ showed no further improvement.
  • Worm/Fly Data: Combinatorial treatments in C. elegans and Drosophila are consistently additive or neutral, rarely toxic. Linda Partridge’s lab showed 3-drug combos worked “beautifully” in flies.
  • Spindler’s Mouse Data: A historical review of mouse studies showed most 2000s-era supplement stacks were neutral (did not extend life). One “Algae Combo” was toxic/shortened life.
  • Polypharmacy Paradox: “Polypharmacy” is deemed bad in the elderly because sick people take drugs; however, “Polypills” (low dose combinations for blood pressure/lipids) are highly effective and reduce side effects in preventative medicine.
  • Toxicity of Concentration: Bananas contain 1g potassium and are safe; a 1g potassium supplement is illegal due to risk of gut perforation. The delivery method alters the safety profile of bioactives.
  • Selection Strategy: Choose compounds that have both a longevity signal in mice (ITP) and a verified biomarker improvement in humans (e.g., lowered ApoB or BP).
  • Rapamycin + Trametinib: A specific combination mentioned as successful in flies, despite both targeting anabolic pathways (mTOR and Ras/ERK).
  • Vitamin D: The speaker prefers a moderate dose (2,000 IU/day) targeting blood levels of ~30 ng/mL, rejecting the “megadose” trend (5k–10k IU) based on recent cohort data.

D. Claims & Evidence Table (Adversarial Peer Review)

Claim from Video Speaker’s Evidence Scientific Reality (Best Available Data) Evidence Grade Verdict
“Supplement users (1-4) are biologically younger.” Cohort of 4,000 healthy individuals (internal data). Confounded. The “Healthy User Bias” is strong; people who take supplements also exercise, sleep better, and are wealthier. Correlation is not causation. C (Observational) Weak/Confounded
“Combinations work better than single agents in animals.” Cites Jan Gruber (worms), Linda Partridge (flies), and ITP (mice). Supported. The ITP (Interventions Testing Program) confirmed Rapamycin + Acarbose extends life more than either alone. A (Animal Meta-analysis) Strong Support
“Most supplements are neutral (don’t work).” Cites Stephen Spindler’s mouse data (early 2000s). Likely True. The vast majority of commercially available “anti-aging” supplements fail in rigorous lifespan tests (e.g., Resveratrol, Curcumin often fail ITP). B (Animal RCT) Strong Support
“Isotretinoin (Accutane) caused long-term dry eye.” Personal anecdote (N=1). Supported. Meibomian gland atrophy is a known, sometimes permanent side effect of isotretinoin therapy. A (Clinical Consensus) Strong Support
“Cocoa Powder extends lifespan.” Mentions rat data and human BP data. Cocoa flavanols reduce CVD risk (COSMOS trial), but robust lifespan extension data in mammals is weak/limited compared to Rapamycin. C (Mixed) Plausible (Healthspan)
“High Dose Vitamin D (5k-10k IU) is unnecessary.” Cites VITAL study (cancer mortality) and cohort data. Consensus. The VITAL trial showed no major benefit for high dose Vit D in general populations. 2,000 IU is widely considered safe/sufficient. A (Human RCT) Strong Support

Verdict Key:

  • Strong Support: Consensus in literature.
  • Plausible: Emerging data.
  • Weak/Confounded: Likely bias or lack of rigor.

E. Actionable Insights (Pragmatic & Prioritized)

Top Tier (High Confidence / Low Risk)

  • The “Intersection” Strategy:
  • Protocol: Only take supplements that pass two filters:
  1. Does it extend life in mice (ITP data)? (e.g., Rapamycin, Acarbose, 17-alpha-estradiol, Glycine).
  2. Does it improve a clinical biomarker in you? (e.g., Acarbose lowers your HbA1c; Glycine improves your sleep).

  • Action: If a supplement does neither, drop it.

  • Moderate Vitamin D:

  • Protocol: 2,000 IU/day aims for ~30-40 ng/mL blood levels. Avoid megadosing unless treating specific deficiency.

  • Cocoa Flavanols:

  • Protocol: High-quality Cocoa powder (non-alkalized) for blood pressure management and vascular health.

Experimental (Risk / Reward)

  • Combinatorial Therapy (The “Stack”):

  • Insight: Based on the speaker’s data, there is a “Sweet Spot” of 1–4 high-impact interventions. Taking 20+ supplements likely hits diminishing returns and increases risk of liver irritation.

  • Protocol: Focus on a core stack (e.g., Rapamycin + Acarbose + Exercise) rather than a “kitchen sink” approach.

  • Intermittent Rapamycin:

  • Protocol: The speaker endorses the Matt Kaeberlein/Biohacker consensus of pulsed dosing (e.g., weekly) to minimize side effects while maintaining longevity benefits.

AVOID (Safety Flags)

  • “Kitchen Sink” Proprietary Blends:

  • Warning: The “Algae Combo” in Spindler’s study shortened lifespan. Multi-ingredient stacks with obscure dosing carry higher toxicity risks than single agents.

  • Potassium Supplements:

  • Warning: Do not take concentrated potassium pills (gut perforation risk). Get it from food (bananas, potatoes).

H. Technical Deep-Dive

1. The Volatility of Longevity Interventions

The speaker introduces a financial risk model for longevity.

  • Mean vs. Variance: A drug might increase average lifespan by 10% but increase early mortality risk by 1% (e.g., immune suppression leading to infection).
  • The “Brian Johnson” Bet: Some biohackers accept high upfront risk (volatility) for the chance of reaching “Escape Velocity” (living long enough for AGI/Nanotech to cure aging).
  • Conservative Bet: Others prefer interventions with lower upside but zero downside (e.g., exercise, olive oil).

2. Polypharmacy vs. Polypills

  • Polypharmacy (Geriatrics): Uncoordinated prescribing of drugs that interact negatively (e.g., anticholinergic burden).
  • Polypill (Preventative): A designed combination (e.g., Statin + ACE Inhibitor + Aspirin) where doses are lowered to reduce side effects while attacking a pathology (CVD) from multiple mechanistic angles. The speaker argues longevity medicine should adopt the Polypill model.

3. Animal Model Discrepancies

  • Worms (C. elegans): Simple genetics. Combinations almost always work.
  • Flies (Drosophila): More complex. Some labs (Partridge) find additivity; others (Moskalev) find interference.
  • Mice (Mammals): The ITP is the gold standard.
  • Success: Rapamycin + Acarbose.
  • Failure: Resveratrol (single), Curcumin (single), Green Tea Extract (single).
  • Lesson: Most “popular” supplements fail in mammals. Stick to the few that work.

I. Fact-Check

  • Speaker Identity: Kamil Pabis, Senior Research Fellow at NUS (National University of Singapore), Center for Healthy Longevity. He works directly with Brian Kennedy (a major figure in geroscience, former Buck Institute CEO) and specializes in combinatorial interventions.
  • Spindler’s Data: TRUE. Stephen Spindler was a pioneer in lifespan screening. His data famously showed that caloric restriction worked, but most mimetics available at the time failed.
  • Bananas vs. Potassium Pills: TRUE. The FDA limits OTC potassium supplements to <100mg (a tiny fraction of a banana) specifically because enteric-coated potassium chloride tablets caused small bowel lesions in the 1960s/70s.
  • Linda Partridge Fly Data: TRUE. Her lab published “A combination of drugs targeting the IIS and TOR pathways extends lifespan in Drosophila” (2019), showing Rapamycin + Lithium + Trametinib was additive.