---
canonical_name: P7C3
alternate_names: Pool 7 Compound 3, P7C3-A20, P7C3-S243, (-)-P7C3-S243, 3,6-dibromo-α-[(phenylamino)methyl]-9H-carbazole-9-ethanol
canonical_topic: P7C3 for Health & Longevity
short_topic_lc: p7c3
creation_date: 2026-1011-1004
creator_ai_fullname: Opus 5.5
ep_keywords: NAMPT Activators, Nicotinamide Phosphoribosyltransferase Activators, Carbazoles, Aminopropyl Carbazoles
---

# P7C3 for Health & Longevity
<section id="top" markdown="1"></section>
Evidence Review created on 10/11/2026 using [AI4L](https://github.com/forever-healthy/AI4L) / Opus 5.5

**Also known as:** Pool 7 Compound 3, P7C3-A20, P7C3-S243, (-)-P7C3-S243, 3,6-dibromo-α-[(phenylamino)methyl]-9H-carbazole-9-ethanol
  
## Motivation

<!-- This motivation section was written last, after every other section of this review was complete, so that it reflects the full scope of the topic rather than anticipating it. -->

P7C3 (Pool 7 Compound 3) is a synthetic small molecule that came out of a search for substances able to keep newly formed brain cells alive. It is not a vitamin, a plant extract, or an approved medicine; it is a laboratory compound, and a family of chemically related versions has grown up around the original.

The molecule was picked out by testing a thousand candidate chemicals directly in living mice rather than in a test tube, an approach unusual in drug discovery. Since then it has been studied in animals for brain injury, Parkinson's-type nerve loss and memory loss with age, and it has been connected to a cell molecule used for energy transfer and repair whose levels fall as people get older. Work in other tissues followed from that same connection.

This review examines what is known about P7C3: how it is thought to work, what the animal and cell studies measured, how strong that evidence is, what safety information exists, and what remains entirely untested in people.

**[Benefits](#expected-benefits) - [Risks](#potential-risks--side-effects) - [Protocol](#therapeutic-protocol) - [Conclusion](#conclusion)**
  
## Recommended Reading

High-level sources that discuss P7C3, or the enzyme pathway it acts on, in enough depth to orient a reader.

<!-- Search performed 10/10/2026. Web searches run for "P7C3 NAMPT activator neuroprotection", "P7C3 NAMPT NAD neuroprotection", and expert-specific queries. Each of the six priority platforms was also searched on its own site: foundmyfitness.com (no results), peterattiamd.com ("Nothing Found"), hubermanlab.com (no results), chriskresser.com ("There are no search results for that term"), lifeextension.com (d-browser returned "Access Denied"; d-proxy-2 returned the genuine search page with no P7C3 results), lifespan.io (two hits). Of the two lifespan.io hits, the NAD⁺-decline feature is listed below as the one priority-platform item that treats the pathway in depth; the monthly "Rejuvenation Roundup December 2025" carries only a one-line abstract excerpt on the P7C3-A20 Alzheimer-reversal paper and is excluded as a paper-listing digest. Every other hit was a journal article, a patent document or a chemical-supplier catalogue page. Selected items are a priority-platform article, narrative reviews, a perspective article and one primary study; the single systematic review was excluded and routed to the Systematic Reviews section, as were encyclopedia entries (Grokipedia has its own section) and supplier catalogue pages (database/directory exclusion). -->

Most primary work on this compound comes from the laboratories that discovered it — Steven McKnight's and Joseph Ready's groups at UT Southwestern and Andrew Pieper's group — who are named inventors on the patents covering the compound class ([US9243281B2](https://patents.google.com/patent/US9243281B2/en), assigned to the University of Texas System and the University of Iowa Research Foundation). That financial interest applies to a large share of the evidence cited throughout this review.

* [P7C3 and an unbiased approach to drug discovery for neurodegenerative diseases](https://pubmed.ncbi.nlm.nih.gov/24514864/) - Pieper et al., 2014

  The discovery team's own account of how the molecule was found and chemically optimised, with the screening logic, the structure-activity work and the early animal results in one place.

* [Small-molecule activation of NAMPT as a potential neuroprotective strategy](https://pubmed.ncbi.nlm.nih.gov/39872755/) - Gu et al., 2022

  A perspective on activating nicotinamide phosphoribosyltransferase (NAMPT, the enzyme that recycles vitamin B3 into the cell's energy-transfer coenzyme) with small molecules, and what still blocks clinical use; its senior author is a named patent inventor.

* [Targeting Nicotinamide Phosphoribosyltransferase as a Potential Therapeutic Strategy to Restore Adult Neurogenesis](https://pubmed.ncbi.nlm.nih.gov/27018006/) - Wang et al., 2016

  Places the compound class inside the wider effort to restore new-neuron formation after stroke, injury and ageing, written by a laboratory with no stake in the patents.

* [Why NAD+ Declines During Aging – Part 1](https://lifespan.io/why-nad-declines-during-aging-part-1/) - Steve Hill

  Qualifies through the shared pathway: a plain-language survey of the human evidence that NAD⁺ (nicotinamide adenine dinucleotide, the cell's energy-transfer coenzyme) falls with age. P7C3 is meant to restore it by activating NAMPT.

* [P7C3-A20 neuroprotection is independent of Wallerian degeneration in primary neuronal culture](https://pubmed.ncbi.nlm.nih.gov/30334859/) - Hill et al., 2018

  The most useful dissenting study: a Cambridge group found no protection against nerve fibres dying back after injury, and concentration-dependent toxicity in cultured neurons.

Only Lifespan.io, among the six priority platforms, carries relevant material, and it covers the coenzyme pathway rather than the compound: the article listed above, plus a one-line excerpt on the Alzheimer-reversal paper in a monthly research digest. P7C3 has never been given to a person, is not sold as a supplement, and has no consumer following, so the other health-optimisation publications and podcasts have not covered it; coverage of the compound itself is confined to the primary scientific literature.
  
## Grokipedia

<!-- grokipedia.com searched directly 10/10/2026. Tier 1 (d-browser) succeeded: browser_navigate to https://grokipedia.com/search?q=P7C3 returned "2 results for 'P7C3'" — a dedicated P7C3 article and a passing mention inside the Nicotinamide phosphoribosyltransferase article. browser_navigate to the article page returned the dedicated page, title "P7C3 — Grokipedia". No further tiers were needed. -->

* [P7C3](https://grokipedia.com/page/p7c3)

  A dedicated encyclopedia entry covering the screening origin, its chemical family, the enzyme target and the animal models, useful as a fast structured orientation before the primary literature.
  
## Examine

<!-- examine.com searched directly 10/10/2026. Tier 1 (d-browser): browser_navigate to https://examine.com/search/?q=P7C3 returned a "Vercel Security Checkpoint" bot wall. Tier 2 (d-fetch): HTTP 429. Tier 3 (d-proxy-1): browser_navigate returned the genuine search page, title "Results for 'P7C3' - Examine", stating "Sorry, there are no search results for P7C3." No article exists. -->

No Examine article on P7C3 exists. Examine covers dietary supplements and nutrition; P7C3 is an unapproved research compound sold only for laboratory use, so it falls outside that scope.
  
## ConsumerLab

<!-- consumerlab.com searched directly 10/10/2026. Tier 1 (d-browser): browser_navigate to https://www.consumerlab.com/search/?q=P7C3 returned the genuine search page, heading "Sorry, we didn't find any results for P7C3". No further tiers were needed. No article exists. -->

No ConsumerLab article on P7C3 exists. ConsumerLab tests consumer supplement and food products; no P7C3 product is sold for human consumption, so there is nothing for it to test.
  
## Systematic Reviews

One systematic review covers the compound class; the search returned no others.

<!-- PubMed searched 10/10/2026 via pubmed_search_articles: "P7C3 OR P7C3-A20 OR P7C3-S243" (96 records, all screened) and "P7C3 AND (systematic review[pt] OR meta-analysis[pt] OR 'systematic review' OR 'meta-analysis')" (1 record). Exactly one publication is typed as a systematic review; no meta-analysis of any P7C3 outcome exists, which follows from the absence of human trials. Selection was by relevance and recency; with a single eligible paper, citation count and study size did not discriminate. -->

* [P7C3 Compounds as Targeted Mitochondrial Therapeutics for Brain Disorders](https://pubmed.ncbi.nlm.nih.gov/42698301/) - Chen et al., 2026

  Searched three databases to synthesise mechanism, structure-activity data and preclinical efficacy across six disease areas, and names the obstacles to human testing.

The trade-off here is one-sided in the literature: this review addresses the claimed benefit of neuroprotection and discusses on-target safety concerns, but no systematic review or meta-analysis exists for the principal risk of activating this enzyme pathway — its consequences for tumour biology. That side of the trade-off is unrepresented at systematic-review level and is covered below from primary studies only.
  
## Mechanism of Action

P7C3 compounds raise cellular nicotinamide adenine dinucleotide (NAD⁺), the coenzyme cells use to carry energy and to fuel the PARP family (DNA-repair enzymes) and the sirtuins SIRT1 and SIRT3 (metabolic regulators). A photocrosslinking experiment identified [nicotinamide phosphoribosyltransferase as the binding partner](https://pubmed.ncbi.nlm.nih.gov/25215490/): it converts nicotinamide (vitamin B3) into nicotinamide mononucleotide (NMN), the rate-limiting step of the salvage route that regenerates NAD⁺. Active analogues raised purified-enzyme activity and restored NAD⁺ in cells depleted by doxorubicin. Higher NAD⁺ is proposed to stabilise mitochondria against the membrane collapse that triggers cell death and to limit oxidative stress in injured neurons.

Competing explanations exist. The prototype molecule also binds the glucagon-like peptide-1 receptor (the target of drugs such as semaglutide) at a secondary site, [acting through inhibition of glycogen synthase kinase 3](https://pubmed.ncbi.nlm.nih.gov/30189238/) (an enzyme whose activity promotes nerve cell death) — a route the later analogues do not appear to use. A proteome screen identified [phosphoglycerate kinase 1, an enzyme of sugar breakdown, as a direct target](https://pubmed.ncbi.nlm.nih.gov/34221965/). In primary neuron cultures the compound [raised NAD⁺ in some neuron types but not others and failed to protect injured nerve fibres](https://pubmed.ncbi.nlm.nih.gov/30334859/), which those authors read as evidence that enzyme activation cannot explain all the protection.

Pharmacologically, the [optimised analogue](https://pubmed.ncbi.nlm.nih.gov/24697290/) is orally bioavailable, brain-penetrant, over 99% protein-bound and metabolically stable in human liver microsomes (half-life 117 minutes); clearance is hepatic, and no published study identifies the enzymes responsible. In vitro it inhibited only CYP2C19 and CYP1A2 (liver enzymes that clear many medicines), both weakly.
  
## Historical Context & Evolution

P7C3 was not repurposed from another use; it was created as a research tool and has only ever had one intended purpose — keeping neurons alive. In 2010 a group at UT Southwestern [screened a thousand small molecules directly in the brains of living mice](https://pubmed.ncbi.nlm.nih.gov/20603013/), looking for compounds that increased the number of surviving new neurons in the hippocampus. Eight worked; one, labelled by its screening plate and well, became P7C3. In the same report, prolonged dosing in aged rats was associated with more surviving new neurons and better performance on cognitive testing, and that observation is what pulled the molecule out of neurology and into the longevity conversation.

A [medicinal-chemistry campaign](https://pubmed.ncbi.nlm.nih.gov/21210688/) produced more potent, orally available analogues, notably P7C3-A20 and [(-)-P7C3-S243](https://pubmed.ncbi.nlm.nih.gov/24697290/), a single mirror-image form. Efficacy reports followed in mouse models of [Parkinson's disease](https://pubmed.ncbi.nlm.nih.gov/23027934/) and [amyotrophic lateral sclerosis](https://pubmed.ncbi.nlm.nih.gov/23027932/), then [blast brain injury](https://pubmed.ncbi.nlm.nih.gov/25220467/), and in 2014 the enzyme target was proposed.

The mechanistic story has not settled. Independent work found the enzyme-activation account insufficient to explain protection, and alternative targets have been proposed. Rather than being overturned, the original screening finding has been repeatedly reproduced while its explanation remains contested — including by the discovery group itself, which has since emphasised injury-induced tau acetylation as a [separate mechanism](https://pubmed.ncbi.nlm.nih.gov/33852912/) in brain injury.
  
## Expected Benefits

<!-- Dedicated search performed 10/10/2026 before writing this section. pubmed_search_articles: "P7C3 OR P7C3-A20 OR P7C3-S243" (96 records, all titles screened); "(P7C3) AND (NAMPT OR nicotinamide phosphoribosyltransferase) AND (activating OR salvage)" (26 records); plus web searches for expert and non-indexed commentary. Searches deliberately included null and opposite-direction findings: the retrieved set includes Hill 2018 (no protection against Wallerian degeneration, plus toxicity), the NOPARK trial of an NAD precursor in humans (worse outcomes than placebo), and the renal-cell-carcinoma and glioma studies in which the compound acts against cell growth. No human efficacy study of P7C3 of any design exists; ClinicalTrials.gov returned zero records for "P7C3", and a search for "NAMPT activator" returned no interventional trial of any direct NAMPT activator (its few text matches are nutrition studies). Every benefit below is therefore graded Speculative under the no-human-outcome-data rule, irrespective of how consistent the animal data are. -->

### High 🟩 🟩 🟩

No benefit reaches High: the class of evidence required — a human clinical endpoint or a validated clinical surrogate, replicated across independent groups — does not exist, because P7C3 has never been administered to a person in any published study.

### Medium 🟩 🟩

No benefit reaches Medium either: that would require the same class of human outcome in a single trial or in controlled observational data, and there is no human trial and no human observational cohort with a comparison group.

### Low 🟩

### Speculative 🟨

#### Survival of Newly Formed Hippocampal Neurons ⚠️ Conflicted

Across [mice](https://pubmed.ncbi.nlm.nih.gov/20603013/) and [rhesus monkeys](https://pubmed.ncbi.nlm.nih.gov/30258178/), treated animals retained more labelled new neurons. [Cultured neurons](https://pubmed.ncbi.nlm.nih.gov/30334859/) showed no protection, and toxicity instead. Net: reproducible in living animals, unexplained in dishes.

#### Preservation of Memory and Learning in Aged and Diseased Brains

Cognitive testing favoured treatment in [aged rats](https://pubmed.ncbi.nlm.nih.gov/20603013/), [chemically amnesic mice](https://pubmed.ncbi.nlm.nih.gov/26646000/), [Alzheimer-model rats](https://pubmed.ncbi.nlm.nih.gov/29246437/) and [advanced-disease Alzheimer-model mice](https://pubmed.ncbi.nlm.nih.gov/41435831/). Basis is animal behaviour only; no human cognitive data exist.

#### Protection of Dopaminergic and Motor Neurons

In mice given MPTP (a toxin that selectively kills dopamine-producing neurons), treated animals retained more of those neurons ([Parkinson model](https://pubmed.ncbi.nlm.nih.gov/23027934/)). A [motor-neuron-disease model](https://pubmed.ncbi.nlm.nih.gov/23027932/) showed preserved spinal neurons and walking gait. Rodent histology only.

#### Limitation of Axonal Degeneration After Head Injury

After blast or impact injury in rodents, treatment was associated with preserved nerve fibres and function, including [when started a year after injury](https://pubmed.ncbi.nlm.nih.gov/33087571/) ([acute model](https://pubmed.ncbi.nlm.nih.gov/25220467/)). Animal histology and behaviour only.

#### Recovery of Function After Ischaemic Stroke

Rats treated even six hours after artery occlusion [regained sensorimotor and spatial-memory function](https://pubmed.ncbi.nlm.nih.gov/28842830/) with more spared tissue; [mice improved walking and brain damage](https://pubmed.ncbi.nlm.nih.gov/30189238/) too. Rodent histology and behaviour only.

#### Protection of Retinal Ganglion Cells After Optic Nerve Injury

After optic nerve crush in rats, treatment [preserved retinal ganglion cell density and damped nerve inflammation](https://pubmed.ncbi.nlm.nih.gov/28973334/), reproducing [an earlier report](https://pubmed.ncbi.nlm.nih.gov/28032230/) from the same laboratory. Basis: rodent histology from one group.

#### Relief of Depressive-Like Behaviour

In mice, treatment [reduced depressive-like behaviour after social stress and food restriction](https://pubmed.ncbi.nlm.nih.gov/24751964/), an effect abolished by ablating hippocampal stem cells. Rodent behavioural assays only, from the discovery group.

#### Preservation of Cardiac Function

Diabetic mice showed [better ejection fraction, fewer arrhythmias and smaller infarcts](https://pubmed.ncbi.nlm.nih.gov/35680376/); rats given the compound after [myocardial reperfusion injury](https://pubmed.ncbi.nlm.nih.gov/42735786/) lost less muscle. Rodent cardiac physiology only.

#### Protection Against Bone Loss

In [ovariectomised rats](https://pubmed.ncbi.nlm.nih.gov/38477537/) (ovaries removed to model menopause), [irradiated rodents](https://pubmed.ncbi.nlm.nih.gov/37385982/), [unloaded limbs](https://pubmed.ncbi.nlm.nih.gov/42829180/) and [aged mice](https://pubmed.ncbi.nlm.nih.gov/39543818/), bone architecture and strength were better preserved. Basis: rodent imaging and mechanical testing.

#### Improved Skeletal Muscle Function and Glucose Control

Diabetic mice showed better [grip strength, running distance and insulin sensitivity](https://pubmed.ncbi.nlm.nih.gov/35060352/); [two-year-old mice](https://pubmed.ncbi.nlm.nih.gov/42764081/) showed better contractility and larger fibres. Rodent physiology only, from one laboratory.

#### Prevention of Chemotherapy-Induced Nerve Damage

In rats given paclitaxel, treatment [prevented behavioural and histological signs of neuropathy](https://pubmed.ncbi.nlm.nih.gov/29125463/), and blocking the enzyme abolished the protection; separately, tumour-bearing mice kept the drug's antitumour effect. Rodent models only.

#### Reduced Liver Fat Accumulation

In mice fed a high-fat diet, treatment was associated with [less hepatic fat, inflammation and fibrosis](https://pubmed.ncbi.nlm.nih.gov/32037512/), with the effect lost in mice lacking AMPK (the cell's fuel-gauge enzyme). Single rodent study.

#### Improved Quality of Aged Egg Cells

In [aged mouse egg cells](https://pubmed.ncbi.nlm.nih.gov/39325941/), maturation with the compound improved spindle assembly and mitochondrial energy output, reduced proton leak, and offset oxidative-stress damage. Basis: cultured rodent egg cells in a single study.

#### Suppression of Tumour Growth at Supra-Neuroprotective Concentrations ⭕️ Not Central to Health & Longevity

Above neuroprotective concentrations, the compound restrained [kidney cancer](https://pubmed.ncbi.nlm.nih.gov/38356717/) cells; it also restrained [glioma](https://pubmed.ncbi.nlm.nih.gov/34221965/) cells. This bears on oncology, not longevity; basis is cell culture plus mouse tumour implants.
  
## Benefit-Modifying Factors

* **No human pharmacogenetic data:** No variant has been tested as a modifier of response, because no human has been dosed. In rodents, [reducing the target enzyme blunted the compound's effect](https://pubmed.ncbi.nlm.nih.gov/29125463/), implying that genetically determined enzyme abundance would matter.

* **Baseline coenzyme status:** The compound acts by restoring depleted NAD⁺ rather than by pushing it above normal; protection in cells appeared where NAD⁺ had been driven down. Benefit would plausibly track baseline depletion, which rises with age and injury.

* **Substrate availability:** Adding nicotinamide, the enzyme's substrate, [raised a subthreshold dose to full effect](https://pubmed.ncbi.nlm.nih.gov/29125463/) in rats, so dietary and supplemental vitamin B3 status is a plausible modifier of any effect.

* **Sex:** The [non-human primate study](https://pubmed.ncbi.nlm.nih.gov/30258178/) used males only. Rodent work in [Alzheimer-model rats](https://pubmed.ncbi.nlm.nih.gov/29246437/) included both sexes and reported protection in each, but no formal sex comparison of effect size has been published.

* **Pre-existing conditions:** Effects were largest in animals with an active insult — injury, toxin exposure, diabetes, oestrogen loss, irradiation. Healthy young animals had little to gain in these assays, which limits extrapolation to healthy adults.

* **Age:** Benefit was most visible in aged animals, where new-neuron survival and target enzyme levels are already low. Older members of the target audience are therefore the group the preclinical data most directly address.
  
## Potential Risks & Side Effects

<!-- Dedicated safety search performed 10/10/2026 before writing this section. Drug reference sources checked: DailyMed (search for "P7C3" returned "0 results — No Drug Package Labels found"), drugs.com (d-browser returned an "Access Denied" interstitial; d-proxy-2 refused the domain under its robots policy, so no result was obtained from that site), and no FDA or EMA authorisation exists for any P7C3 compound, so there is no prescribing information, no package insert and no post-marketing surveillance. Supplier documentation (Cayman Chemical item 16682) carries the warning "This product is not for human or veterinary use." PubMed searches: "P7C3 AND (pharmacokinetics OR toxicity OR 'drug interaction' OR CYP OR bioavailability)" (17 records) plus the full 96-record P7C3 set. The search also covered findings of no excess risk: the primate study found no microscopic toxicology across 35 tissue types after 38 weeks of daily oral dosing, rodent studies reported no change in weight, behaviour or appearance at up to 40 mg/kg/day for 21 days, and in-vitro screening found no meaningful cardiac ion-channel inhibition (under 15% hERG block at 50 µmol/L). These null findings are reported in the items below. Every risk is graded Speculative: there is no human adverse-event data of any kind. -->

### High 🟥 🟥 🟥

No risk reaches High: the required class of evidence — documented adverse events in people, replicated across independent groups — cannot exist for a compound that has never been administered to a human in a published study.

### Medium 🟥 🟥

No risk reaches Medium either: that would require human adverse-event data from a single trial or controlled observational data, and neither exists. All toxicology is animal, cellular or in-vitro.

### Low 🟥

### Speculative 🟨

#### Neurite Toxicity Above a Narrow Concentration ⚠️ Conflicted

In cultured mouse neurons, [P7C3-A20 caused concentration-dependent neurite degeneration above 100 nmol/L, within six hours at 10 µmol/L](https://pubmed.ncbi.nlm.nih.gov/30334859/), while [rodent dosing studies found none](https://pubmed.ncbi.nlm.nih.gov/24697290/). Net: a real in-vitro ceiling, unmapped in living humans.

#### Unresolved Effect on Tumour Biology ⚠️ Conflicted

Tumours depend on this enzyme and inhibitors of it are developed as [anticancer agents](https://pubmed.ncbi.nlm.nih.gov/42698301/); yet P7C3 restrained [cancer cells](https://pubmed.ncbi.nlm.nih.gov/38356717/) and [did not blunt radiotherapy](https://pubmed.ncbi.nlm.nih.gov/41680036/). Net: direction unresolved, no human data.

#### Suppression of Inflammatory Immune Cells

P7C3 [blocked T helper 17 cell differentiation](https://pubmed.ncbi.nlm.nih.gov/32647376/) (immune cells that drive inflammation) and damped brain autoimmunity in mice. Therapeutic in that setting, but the same action could weaken host defence; no infection outcome was measured.

#### Interference With Drug-Metabolising Enzymes

In vitro, the optimised analogue [inhibited CYP2C19](https://pubmed.ncbi.nlm.nih.gov/24697290/) at 1.9 µmol/L and CYP1A2 at 20 µmol/L. Over 99% protein binding keeps free concentrations far below those values, so clinical relevance is unknown.

#### Off-Target Receptor Binding

[Screening 47 neuronal receptors and channels](https://pubmed.ncbi.nlm.nih.gov/24697290/) found displacement only at translocator protein (0.35 µmol/L) and the mu-opioid receptor (8.2 µmol/L). Both are weak relative to free drug levels; consequences in humans are untested.

#### Possible Lowering of Blood Glucose

The prototype molecule [activates the glucagon-like peptide-1 receptor at a secondary site and raised insulin secretion](https://pubmed.ncbi.nlm.nih.gov/30189238/) in cells; [fasting glucose fell in diabetic mice](https://pubmed.ncbi.nlm.nih.gov/35680376/). Mechanistically plausible in people, never measured there.
  
## Risk-Modifying Factors

* **Genetic variation:** Untested in humans. [Neither deleting Sarm1 nor carrying the protective Wlds mutation](https://pubmed.ncbi.nlm.nih.gov/30334859/) (genes governing how fast injured nerve fibres die back) spared cultured neurons from the compound's toxicity, so that toxicity runs through another pathway.

* **Baseline coenzyme and enzyme levels:** Tissues already rich in the target enzyme have least headroom for activation and most scope for pushing an intermediate metabolite too far. No human measurement of this margin exists.

* **Sex:** No sex-stratified safety data exist. The only [chronic non-human primate toxicology study](https://pubmed.ncbi.nlm.nih.gov/30258178/) dosed male animals exclusively, so female-specific organ findings have never been looked for.

* **Pre-existing conditions:** Active or prior cancer is the condition of greatest theoretical concern, given tumour dependence on this enzyme pathway. Liver disease matters because clearance is hepatic, and diabetes because of the glucose signal.

* **Age:** Older adults carry more undiagnosed cancer and more impaired hepatic and renal clearance, which widens exposure and compounds the unresolved tumour question — in the same group where preclinical benefit looks largest.

* **Pregnancy and reproduction:** No reproductive or developmental toxicology has been published. Rodent studies have dosed pregnant animals, but none was designed to detect birth defects.
  
## Key Interactions & Contraindications

<!-- Dedicated interaction search performed 10/10/2026: pubmed_search_articles for "P7C3 AND (pharmacokinetics OR toxicity OR 'drug interaction' OR CYP OR bioavailability)" (17 records, all screened), plus review of the in-vitro enzyme and receptor panels in Naidoo 2014 (PMID 24697290) and the primate pharmacokinetic data in Bauman 2018 (PMID 30258178). No human interaction study and no human pharmacokinetic study of any P7C3 compound exists, so every entry below is inferred from mechanism or from in-vitro data and is marked accordingly. -->

* **CYP2C19 substrates (theoretical):** Clopidogrel, omeprazole, escitalopram, voriconazole. Caution; in-vitro inhibition at 1.9 µmol/L could raise substrate levels or, for clopidogrel, reduce its activation and its antiplatelet effect. Dose separation is untested.

* **CYP1A2 substrates (theoretical):** Caffeine, theophylline, clozapine, tizanidine. Caution only; raised substrate levels, with sedation or tachycardia, are the theoretical consequence, but in-vitro inhibition was weak, at 20 µmol/L, far above plausible free concentrations.

* **Over-the-counter medications (theoretical):** Omeprazole and esomeprazole, sold without prescription, are CYP2C19 substrates; paracetamol (acetaminophen) shares the hepatic route. Monitor; raised substrate levels are the theoretical consequence, though activating this enzyme [reduced paracetamol liver injury in mice](https://pubmed.ncbi.nlm.nih.gov/29684358/).

* **Glucose-lowering drugs and supplements (theoretical):** Insulin, sulfonylureas (older blood-glucose-lowering tablets — glipizide, glibenclamide), glucagon-like peptide-1 agonists (semaglutide), berberine. Caution; additive effect could cause hypoglycaemia (blood glucose falling too low, with sweating and confusion).

* **Vitamin B3 forms (theoretical, additive):** Nicotinamide, nicotinamide riboside, nicotinamide mononucleotide and niacin supply the enzyme's substrate, and nicotinamide [potentiated a subthreshold dose in rats](https://pubmed.ncbi.nlm.nih.gov/29125463/). Caution; the consequence is an amplified compound effect rather than toxicity.

* **NAMPT inhibitors (theoretical, opposing):** FK866 and the oncology agents in this class occupy the same site. Avoid (theoretical); mutual antagonism is expected, with the inhibitor abolishing any P7C3 effect.

* **Cytotoxic chemotherapy (theoretical):** In rodents the compound protected nerves without reducing paclitaxel's antitumour effect, and [did not impair radiotherapy](https://pubmed.ncbi.nlm.nih.gov/41680036/). Monitor; whether that separation holds in human tumours is unknown.

* **Opioids (theoretical):** Weak mu-opioid receptor displacement in vitro, at 8.2 µmol/L. Caution only; altered analgesia is the theoretical consequence, improbable at achievable free concentrations and never looked for in any species.

**Populations who should avoid P7C3:**

* Everyone outside a regulated clinical trial — no P7C3 compound has marketing authorisation from any regulator, no human dose has been established, and no human-grade product exists.
* People with active or recently treated cancer — the target enzyme supports tumour metabolism, and the net effect of activating it in a tumour-bearing human is unknown.
* Pregnant or breastfeeding women — no reproductive or developmental toxicology exists.
* Children and adolescents — no paediatric data of any kind exist.
* People with hepatic impairment — clearance is hepatic and has been characterised only in animals and liver preparations.

No threshold values — for example a Child-Pugh class (a liver-function score) or a platelet floor — can be given for these groups: no product label, guideline or trial exclusion criterion exists to source them from, so the general categories are named without numbers.
  
## Risk Mitigation Strategies

The parameters below follow common practice in early-phase small-molecule research unless a citation is given; none is a clinical standard, because no clinical standard exists for this compound.

* **Confine exposure to a regulated trial:** The decisive mitigation for every risk listed above is that no human-grade P7C3 exists. Research-grade material carries supplier warnings against human use and has no toxicology package behind it.

* **Treat the in-vitro toxicity ceiling as a dose constraint:** Neurite degeneration appeared above 100 nmol/L in culture and within six hours at 10 µmol/L, so any first-in-human design would cap exposure well below that, guarding against direct nerve toxicity.

* **Anchor dose on the primate exposure record:** The only chronic primate study used 10 mg/kg/day orally for 38 weeks with peak levels of 963–2320 ng/mL and [no microscopic toxicology across 35 tissues](https://pubmed.ncbi.nlm.nih.gov/30258178/), bounding the exposure actually examined.

* **Screen for and monitor cancer:** Because activating this enzyme could in principle support tumour metabolism, age-appropriate cancer screening before exposure and surveillance during it addresses the single largest unresolved risk.

* **Monitor blood glucose early:** Fasting glucose at baseline, week 1 and week 4 would detect the hypoglycaemia predicted by the receptor and rodent glucose findings, particularly in anyone already using glucose-lowering drugs.

* **Review co-medication for the inhibited enzymes:** Checking for CYP2C19 substrates, especially clopidogrel, before exposure limits the risk of altered drug levels arising from the in-vitro inhibition signal.

* **Check liver enzymes periodically:** Alanine aminotransferase at baseline and at each monitoring visit, pausing exposure if it exceeds three times the upper limit of normal, addresses the hepatic clearance route; that threshold is conventional practice, not P7C3-specific.
  
## Therapeutic Protocol

There is no therapeutic protocol. No practitioner, clinic or guideline uses P7C3, because it has never been administered to a human being in a published study and no regulator has authorised it. The figures below are animal study regimens, reported so the tested record is visible; they are not human doses, and no human-equivalent dose has been established. Other parameters without a citation reflect common laboratory practice rather than clinical convention.

* **No established human regimen:** No dose, frequency, duration or route has been defined for humans. Converting an animal milligram-per-kilogram figure into a human dose is not valid without human pharmacokinetic data, which do not exist.

* **Longest primate regimen tested:** 10 mg/kg/day of P7C3-A20 administered orally for 38 weeks in rhesus monkeys, in corn oil with flavoured syrup, with exposure stable across the period ([Bauman et al.](https://pubmed.ncbi.nlm.nih.gov/30258178/)).

* **Rodent dose range tested:** 1 mg/kg/day was the lowest active dose and 10 mg/kg/day the efficacy ceiling in mouse neuroprotection and Parkinson-model assays ([Naidoo et al.](https://pubmed.ncbi.nlm.nih.gov/24697290/)); liver studies used 20 mg/kg/day ([Hua et al.](https://pubmed.ncbi.nlm.nih.gov/32037512/)).

* **Route:** Oral gavage and intraperitoneal injection give comparable brain levels, with oral dosing partitioning into brain better despite lower plasma availability ([Naidoo et al.](https://pubmed.ncbi.nlm.nih.gov/24697290/)).

* **Single versus split dosing:** Both have been used — once daily in the primate and most rodent work, twice daily in the 21-day Parkinson-model protocol ([Naidoo et al.](https://pubmed.ncbi.nlm.nih.gov/24697290/)). The two schedules have never been compared.

* **Elimination and dosing interval:** No human half-life exists. In monkeys, levels peaked near four hours and fell to 73–132 ng/mL by 24 hours, which supported once-daily dosing in that species ([Bauman et al.](https://pubmed.ncbi.nlm.nih.gov/30258178/)).

* **Best time of day:** Not studied in any species. The target enzyme is under circadian control ([Khaidizar et al.](https://pubmed.ncbi.nlm.nih.gov/33918226/)), so timing is mechanistically plausible as a variable but entirely theoretical.

* **Competing approaches:** The only approach in this family available to humans is oral coenzyme precursors — nicotinamide riboside and nicotinamide mononucleotide — which raise the same coenzyme indirectly. Other direct enzyme activators, such as SBI-797812, [remain preclinical](https://pubmed.ncbi.nlm.nih.gov/31324777/).

* **Mirror-image form matters:** Activity resides almost entirely in one enantiomer; the (S)-(-) form of the optimised analogue was protective while its mirror image was indistinguishable from vehicle ([Naidoo et al.](https://pubmed.ncbi.nlm.nih.gov/24697290/)).

* **Genetic polymorphisms:** No pharmacogenetic guidance exists. Variation in CYP2C19 would be the first candidate to examine, given the in-vitro inhibition signal, but the compound's own clearance route in humans is uncharacterised.

* **Sex differences:** No dosing difference has been studied. The chronic primate work used males only; rodent efficacy work has included both sexes without reporting dose-response differences between them.

* **Age:** No age-adjusted regimen exists. Aged rats and two-year-old mice tolerated the same milligram-per-kilogram doses as young animals in the published work.

* **Baseline biomarkers:** No biomarker has been validated to guide dosing. Blood coenzyme metabolites are the obvious candidate and were tracked as an exploratory measure in human precursor trials, not as a dosing guide.

* **Pre-existing conditions:** Diabetic, irradiated, injured and oestrogen-deficient animals all received standard doses in their respective studies; no dose adjustment for any condition has been proposed or tested.
  
## Discontinuation & Cycling

* **Intended duration unknown:** No human treatment course exists. Animal work has run from five days to 38 weeks, with no study designed to establish whether continuous or finite use would be appropriate.

* **Benefit persisted after stopping in rodents:** A 30-day course given a year after brain injury was followed by cognitive recovery that [persisted for months after dosing ended](https://pubmed.ncbi.nlm.nih.gov/33087571/), suggesting repair rather than symptom suppression.

* **No withdrawal effects documented:** No study has looked for rebound or withdrawal phenomena after stopping. Animals returned to baseline husbandry without reported deterioration, but this was never a measured endpoint.

* **No tapering protocol:** Every published study stopped dosing abruptly. No tapering schedule has been designed or tested in any species, and none is implied by the proposed mechanism.

* **Cycling not studied:** No intermittent or cyclical schedule has been tested. Whether the target enzyme downregulates under sustained activation — the usual reason to cycle — is unknown.
  
## Sourcing and Quality

* **No human-grade product exists:** P7C3 is sold only as a research chemical. The main supplier's page for it ([P7C3](https://www.caymanchem.com/product/16682/p7c3)) states "This product is not for human or veterinary use" and lists it as a biochemical for neuroscience research.

* **No pharmacopeial standard:** There is no monograph, no approved manufacturing standard and no regulatory oversight of composition. Catalogue material is specified for laboratory purity (at least 98%) rather than for human exposure.

* **No independent consumer testing:** No consumer testing programme covers this compound, because no consumer product exists to test. Certificates of analysis are batch documents written for laboratories, not evidence of suitability for human use.

* **Mirror-image composition is the critical specification:** Activity sits in one mirror-image form, so racemic material — an even mixture of both forms — is roughly half inactive; the single-form analogue required a dedicated synthesis to reach 99% purity.

* **Identity checkpoints:** The parent compound carries registry number 301353-96-8 (Chemical Abstracts Service), formula C₂₁H₁₈Br₂N₂O, molecular weight 474.2, formal name 3,6-dibromo-α-[(phenylamino)methyl]-9H-carbazole-9-ethanol. Analogues sold under the same family name differ chemically and in potency.

* **No compounding route:** In the United States, pharmacy compounding may use only bulk substances meeting a pharmacopeial monograph, contained in an approved drug, or on the [agency's permitted list](https://www.fda.gov/drugs/human-drug-compounding/bulk-drug-substances-used-compounding-under-section-503a-fdc-act) — P7C3 meets none.
  
## Practical Considerations

* **Time to effect:** Unknown in humans. In rodents, new-neuron survival changed over 5–15 days of dosing; functional recovery after brain injury required about 30 days of treatment, and bone changes took weeks to months.

* **Common pitfall — scaling animal doses:** Treating 10 mg/kg in a mouse or monkey as a human dose ignores species differences in clearance and the absence of human pharmacokinetic data. No valid conversion can be made from the published record.

* **Common pitfall — assuming precursor results transfer:** Oral coenzyme precursors are often treated as interchangeable with direct enzyme activation. They act at a different step, and the human precursor results have not been favourable.

* **Common pitfall — mistaking a research chemical for a supplement:** Material bought from chemical catalogues has no human safety testing, no sterility or excipient control, and supplier documentation that explicitly excludes human use.

* **Regulatory status:** Not approved by any regulator, not a dietary supplement, and not a prescription drug — a [DailyMed](https://dailymed.nlm.nih.gov/dailymed/search.cfm?labeltype=all&query=P7C3) search returns no label. In the United States it would be an unapproved new drug.

* **Cost and accessibility:** Research quantities cost roughly tens to hundreds of euros per 5–50 mg, but price is not the barrier: there is no clinical supply chain and no lawful route of access outside research.

* **Structural funding asymmetry:** A patented molecule attracts development capital that inexpensive, unpatentable vitamin B3 precursors do not, which shapes which arm of this mechanism gets tested in humans and which gets tested only in animals.
  
## Interaction with Foundational Habits

* **Sleep:** Indirect, direction unknown. The target enzyme oscillates with the daily clock ([Khaidizar et al.](https://pubmed.ncbi.nlm.nih.gov/33918226/)), so disrupted sleep-wake cycles would be expected to alter the pathway the compound acts on. No study has measured sleep in treated animals or dosed at different circadian times.

* **Nutrition:** Directly potentiating. The enzyme's substrate is vitamin B3, and co-administered nicotinamide [raised a subthreshold dose to full effect](https://pubmed.ncbi.nlm.nih.gov/29125463/) in rats. Practical consequence: substrate supply is part of the intervention, and animal dosing used oil-based vehicles reflecting poor water solubility.

* **Exercise:** Overlapping mechanism, untested in combination. Exercise raises this coenzyme and its salvage enzyme in muscle through the same fuel-gauge signalling the compound engages, and treated diabetic mice ran [fourteen times further on a wheel](https://pubmed.ncbi.nlm.nih.gov/35060352/). Whether the two are additive or redundant is unknown.

* **Stress management:** Indirect, protective direction in animals. Treatment blunted the behavioural consequences of [chronic social defeat](https://pubmed.ncbi.nlm.nih.gov/24751964/) in mice — an effect lost when hippocampal stem cells were ablated — and, when dosed to pregnant dams, of [gestational stress](https://pubmed.ncbi.nlm.nih.gov/33501899/) in their offspring. No human stress or cortisol data exist.
  
## Monitoring Protocol & Defining Success

No human monitoring protocol exists for P7C3, because no registered trial has given it to a person; what follows is the monitoring logic the preclinical record implies for a first-in-human setting, not a protocol anyone currently follows. Baseline work in such a setting would establish liver enzymes, a complete blood count, fasting glucose, and a blood measurement of the cell's energy-carrying coenzyme and its breakdown products — the systems the animal and in-vitro data actually flag: hepatic clearance, the blood-forming marrow examined in the primate toxicology study, glucose handling, and the pathway the compound is meant to engage. Ongoing testing would follow the cadence typical of first-in-human small molecules: at 1 week, 4 weeks, 12 weeks, then every 3–6 months, with any new symptom prompting earlier testing rather than waiting for the next scheduled visit.

| Biomarker | Optimal Functional Range | Why Measure It? | Context/Notes |
|---|---|---|---|
| Blood NAD⁺ metabolite panel | No established target; track change from the individual's own baseline | Marker the compound is expected to change — it is the direct output of the pathway | NAD⁺ = nicotinamide adenine dinucleotide, the coenzyme cells use for energy transfer and repair. Mass-spectrometry research assay, not a routine clinical test; no treated-human data exist |
| ALT | 7–56 U/L (standard reference range) | Safety check — clearance is hepatic, and a rise would pause or stop use | ALT = alanine aminotransferase, an enzyme released when liver cells are injured. No fasting needed; pair with AST (aspartate aminotransferase) and bilirubin to localise the signal |
| Complete blood count with differential | Haemoglobin 13.5–17.5 g/dL (men), 12.0–15.5 g/dL (women); neutrophils 1.5–8.0 ×10⁹/L (standard reference ranges) | Safety check — a falling count would stop use | Bone marrow was among the 35 tissues examined without findings in the primate study, and marrow is the dose-limiting tissue when this same enzyme is blocked pharmacologically |
| Fasting glucose | 70–99 mg/dL (standard reference range) | Marker expected to change, and a safety check for glucose falling too low | Requires an 8-hour fast. Diabetic mice showed lower fasting glucose, and the prototype also activates the glucagon-like peptide-1 receptor, target of drugs such as semaglutide |
| Plasma p-tau217 | No established target; track change from the individual's own baseline | Marker expected to change where brain injury or Alzheimer-type pathology is present | p-tau217 = a phosphorylated form of the tau protein measurable in blood. It fell in treated Alzheimer-model mice; no treated-human data exist |

Qualitative markers would carry unusual weight here, since no validated response biomarker exists:

* Memory and word-finding in daily tasks, recorded in a diary rather than from recollection.
* Mood and motivation, given the antidepressant-like signal in rodent stress models.
* Physical function — grip, stair climbing, walking distance — reflecting the muscle findings in animals.
* Any new numbness, tingling or burning in hands and feet, which would be the first human sign of the nerve toxicity seen in culture.
  
## Emerging Research

* **No registered human trial of P7C3 exists:** A ClinicalTrials.gov search returned zero records for "P7C3" and no interventional trial of any direct NAMPT activator. Nothing is recruiting, nothing is publicly planned, and no first-in-human study has been announced by the patent holders.

* **NOPARK — completed, results published, negative:** [NCT03568968](https://clinicaltrials.gov/study/NCT03568968) completed in June 2025 with 410 participants; no results are posted on the registry. The [publication](https://pubmed.ncbi.nlm.nih.gov/42848371/) reports that raising this coenzyme with an oral precursor left patients worse than placebo on the standard Parkinson rating scale (adjusted mean difference 2.72 points, 95% confidence interval 0.47–4.98).

* **What NOPARK changes here:** It does not test P7C3, but it is the first well-powered human test of the premise that raising this coenzyme slows nerve loss, and it weakens that premise considerably. Any reading of the P7C3 animal literature now has to account for it.

* **NADAPT — ongoing in atypical parkinsonism:** [NCT06162013](https://clinicaltrials.gov/study/NCT06162013), phase 2, 330 participants, recruiting, completion December 2028, with disease-rating scales at 78 weeks as primary endpoints. A positive result would restore the mechanism's human plausibility; a null result would further undercut this review's benefit section.

* **NO-ALS — ongoing in motor neuron disease:** [NCT04562831](https://clinicaltrials.gov/study/NCT04562831), 380 participants, active and not recruiting, completion October 2027, with a functional rating scale as primary endpoint. A positive result would support the pathway where P7C3 showed rodent efficacy; a null result would leave that claim resting on mice.

* **Direct activators versus precursors:** [SBI-797812](https://pubmed.ncbi.nlm.nih.gov/31324777/) (Gardell et al., 2019) acts on the same enzyme through different chemistry and raised liver coenzyme levels in mice. Human data on any direct activator would be the cleanest test of whether P7C3's route is the problem or the pathway is.

* **Target identity still open:** Work on [phosphoglycerate kinase 1](https://pubmed.ncbi.nlm.nih.gov/34221965/) (Chen et al., 2021) and the [failure to reproduce enzyme-mediated protection in culture](https://pubmed.ncbi.nlm.nih.gov/30334859/) (Hill et al., 2018) leave the mechanism unsettled. Resolving it would determine which human indication, if any, is worth testing.

* **Tau acetylation as an alternative route:** The discovery group's [brain-injury work](https://pubmed.ncbi.nlm.nih.gov/33852912/) (Shin et al., 2021) identified rapid tau acetylation as a driver of nerve fibre loss, with blood acetylated tau as a candidate biomarker — a line that could make a human trial measurable rather than speculative.

* **Reversal claims await replication:** The [2026 report of reversing advanced disease](https://pubmed.ncbi.nlm.nih.gov/41435831/) (Chaubey et al., 2026) in two Alzheimer mouse strains is the strongest preclinical claim in this literature and comes from the patent-holding group. Independent replication is the obvious next step.
  
## Conclusion

P7C3 is a laboratory-made molecule, not a supplement and not a licensed medicine, and the most important fact about it is that no person has knowingly received it in a published study. Everything in this review comes from cells, rodents and one small group of monkeys. Within that world the findings are unusually consistent: across many separate laboratories and many kinds of injury, treated animals kept more nerve cells alive and did better on the tests used, and animal work also points to effects on the heart, bone, muscle, blood sugar handling, liver fat and egg cells.

Consistency in animals is not benefit in people. The work rests heavily on the laboratories that discovered the compound and hold the patents covering it, which is a reason for care rather than dismissal, and independent groups have both extended the findings and contradicted part of the proposed explanation for how it works. One cell study found the compound harmful to nerve fibres above a narrow concentration. The broader idea behind it — that topping up the cell's energy-carrying molecule slows nerve loss — was recently tested in people using a widely sold nutrient, and those taking it fared worse than those on placebo.

For someone focused on long-term health, P7C3 is an open research story rather than an available option: no human dose, no safety record in people, no approved product, and no validated way to tell whether it is doing anything.

**[Top](#top) - [Benefits](#expected-benefits) - [Risks](#potential-risks--side-effects) - [Protocol](#therapeutic-protocol)**