Compound education · interactions · what is actually known
Educational reference
Educational reference compiled from published research and community reports — not medical advice, not an endorsement of any compound or combination.
No doses, no protocols, no rankings · every claim carries its evidence tier · anything touching prescription medication, pregnancy, cancer history or an upcoming procedure belongs with a prescriber
Compound reference
All seven tracked compounds — what each is studied for, and how much is actually known. The evidence tier carries the same visual weight as the name on purpose.
Tirzepatide
approved drug
A dual GIP/GLP-1 receptor agonist approved for type 2 diabetes and for chronic weight management, with an additional approved indication in obstructive sleep apnea with obesity. Community tracking centers on a fixed weekly injection day and on the symptom window that follows each dose change.
Flags
Boxed warning: thyroid C-cell tumours in rodents — contraindicated with a personal or family history of medullary thyroid carcinoma or MEN2 (label).
Labelled warnings include pancreatitis, gallbladder disease, and acute kidney injury from vomiting/diarrhoea-driven dehydration (label).
Labelling notes that delayed gastric emptying may reduce oral contraceptive exposure around initiation and dose escalation (label).
Delayed gastric emptying is a periprocedural question — tell an anaesthetist or endoscopist before any procedure (guideline consensus).
Hypoglycaemia risk rises when combined with insulin or insulin secretagogues (label).
Compounded and grey-market vials have been the subject of FDA dosing-error and adverse-event warnings (regulatory).
Semaglutide
approved drug
A GLP-1 receptor agonist approved for type 2 diabetes and for weight management, with an approved cardiovascular risk-reduction indication in specified populations. The longest community history of any compound here, so most maintenance and post-cessation discussion attaches to it.
Flags
Boxed warning: thyroid C-cell tumours in rodents — contraindicated with a personal or family history of medullary thyroid carcinoma or MEN2 (label).
Labelled warnings include pancreatitis, gallbladder disease, dehydration-related acute kidney injury, and diabetic retinopathy monitoring (label).
Delayed gastric emptying can alter absorption of co-administered oral drugs — most relevant for narrow-therapeutic-index medications (label / mechanistic).
Delayed gastric emptying is a periprocedural question — flag it before surgery, endoscopy or anaesthesia (guideline consensus).
A meaningful fraction of weight lost on GLP-1 therapy is lean tissue (human trials) — a monitoring fact, not a reason to add anything.
Retatrutide
human trials
An investigational triple agonist at the GIP, GLP-1 and glucagon receptors, studied in a phase 3 obesity and type 2 diabetes programme with positive topline readouts reported. It is not approved anywhere, so anything sold under the name today sits entirely outside a dispensing pharmacy.
Flags
Investigational only — no label, no pharmacovigilance, no verified identity or purity outside a trial (regulatory).
Glucagon-receptor agonism is the differentiator: trial programmes monitor heart-rate increases and glycemic parameters alongside dose-dependent GI events (human trials).
Long-term safety, carcinogenicity and drug-interaction data are not public in label form; class-level thyroid C-cell and pancreatitis questions should be treated as unresolved rather than absent (mechanistic).
Grey-market supply carries identity, purity, endotoxin and sterility risk on top of the pharmacology (regulatory).
BPC-157
animal studies only
A synthetic pentadecapeptide from a gastric protein sequence, studied in rodent models of gut, tendon, ligament, muscle and nerve injury, largely via a VEGFR2/Akt/eNOS angiogenic axis. There is no completed published human efficacy trial and no human safety data of any tier.
Flags
Not an approved drug anywhere; no human toxicology, carcinogenicity, immunogenicity or reproductive-safety data exists (regulatory).
Theoretical oncologic concern: the same angiogenic and cell-migration signalling that drives repair is exploited by tumours. Never studied in tumour-initiation, progression or cancer-patient settings — plausible, unquantified, untested, not an observed human harm (animal / mechanistic).
Personal or family cancer history, active malignancy, undiagnosed masses or proliferative retinopathy are prescriber conversations, not app questions (mechanistic).
Prohibited at all times for tested athletes under anti-doping rules (regulatory).
Reported symptom relief is not evidence a tissue injury has resolved — returning to load on a masked injury is a commonly reported regret (community report).
US compounding status has moved repeatedly (Category 2 listing in 2023, procedural removal in 2026 following withdrawn nominations, further committee review scheduled) — a procedural change is not a safety finding (regulatory).
TB-500
animal studies only
A synthetic acetylated fragment corresponding to residues 17–23 of thymosin beta-4 — not the full-length molecule that most published trial data refers to. Preclinical work describes actin binding, cell migration, anti-inflammatory and pro-angiogenic effects; the fragment itself has no comparable human programme.
Flags
Not an approved drug; no human efficacy trial, toxicology, carcinogenicity or long-term data for the fragment (regulatory).
Sources that equate TB-500 with full-length thymosin beta-4 are overstating the evidence — the Phase 1 healthy-volunteer data belongs to the full-length peptide (animal / human trials).
Theoretical oncologic concern reads stronger on paper than BPC-157's: thymosin beta-4 is associated with increased metastatic potential in several tumour models, alongside pro-angiogenic action. No clinical evidence demonstrates this in humans and none excludes it (animal / mechanistic).
Named on the WADA prohibited list under growth factors — prohibited at all times for tested athletes (regulatory).
Rarely run alone in community reports, so its individual effects are almost never cleanly observed by anyone (anecdotal).
NAD+
anecdotal only
A coenzyme central to redox and sirtuin/PARP biology. Most published human NAD-biology work uses oral precursors (NR/NMN) rather than injection, and those precursor trials are largely null on glucose and lipid endpoints; subjective effects reported for injectable NAD+ — energy, clarity, recovery — rest on self-report.
Flags
Infusion reactions are the dominant documented flag: chest tightness, flushing, nausea, vomiting, cramping, headache, lightheadedness and raised heart rate, all rate-dependent (human observational).
Chest tightness during an infusion is symptomatically indistinguishable from a cardiac presentation — that ambiguity is the clinical reason it matters, not a branding quirk.
No FDA-approved indication for injectable NAD+; sterility and identity risk applies to anything obtained outside a licensed pharmacy (regulatory).
Precursor data are mixed to null: nicotinamide riboside showed no change in insulin sensitivity or glucose disposal in obese men, and NMN meta-analyses show no significant effect on fasting glucose, insulin, HbA1c or lipids (human trials).
Nicotinic-acid (niacin) formulations are a different molecule with flushing, hepatotoxicity at sustained-release doses and glucose-raising effects — a real flag alongside any glucose-lowering agent (human trials).
Usually run continuously in community reports, which leaves no off-period to attribute anything to (anecdotal).
No human lifespan or healthspan endpoint has been demonstrated.
GHK-Cu
human trials
A copper-binding tripeptide with the strongest human record in this set — but almost entirely topical, in small controlled trials measuring skin density, fine lines and scar appearance. Injectable/systemic GHK-Cu has no published clinical safety study, so topical results do not transfer to the injected route.
Flags
Human trial evidence is topical and short-duration with small samples; there is no Phase III programme and no injectable safety study (human trials / regulatory).
Community reports for the injected route describe flushing within minutes, injection-site reactions, and sparse reports of discoloration at repeatedly used sites (anecdotal).
Injection bypasses gastrointestinal copper regulation — a flag for anyone with a copper-handling disorder such as Wilson's disease, or stacking multiple copper-containing products (mechanistic).
Pro-angiogenic remodeling activity means the same cancer-history caution applied to BPC-157 and TB-500 applies here, at lower intensity for topical use (mechanistic).
US compounding status changed procedurally in 2026 with a further committee review scheduled — not a favourable safety finding (regulatory).
Commonly co-tracked stacks
What people report running together, with the evidence stated as it actually stands. Listed in no particular order — inclusion is a description of community behaviour, not a recommendation.
Healing stack ("Wolverine")
BPC-157TB-500
Reported goal
Soft-tissue, tendon and ligament recovery after injury or surgery. Rationale is complementary rodent mechanisms; the combination itself has never been studied in humans (animal / anecdotal).
Scheduling
Typically run concurrently as a time-boxed block tied to an injury rather than continuously, on two different cadences — BPC-157 on a frequent one, TB-500 on a less frequent one because of its longer reported duration. That mismatch is the most common source of tracking confusion, which is why each protocol carries its own schedule here.
What to watch
Additive angiogenic signalling is the standing caution — a prescriber conversation for anyone with cancer history. There is no published stability data for combining reconstituted peptides in one syringe. Both are prohibited for tested athletes, and reported pain relief is not evidence a tissue has healed.
Two goals get conflated here — easing GI discomfort during titration, and supporting training recovery while in a calorie deficit. Zero published trials, observational studies or case series exist for BPC-157 with any GLP-1 (anecdotal).
Scheduling
The GLP-1 keeps its fixed weekly injection day; BPC-157 runs on its own frequent cadence. Community write-ups commonly describe running the GLP-1 alone first and adding anything else later, which is also the only way to attribute a new symptom.
What to watch
Escalating GI symptoms are a clinician question, not a stacking question — pancreatitis and gallbladder disease are labelled GLP-1 concerns and present as upper-abdominal pain. Using a healing peptide to push through them can delay evaluation.
Facial volume loss, laxity and skin texture during rapid weight loss. The phenomenon itself is increasingly characterised in the literature; GHK-Cu's human evidence is topical photoaging work, not post-weight-loss laxity (human trials / anecdotal).
Scheduling
GHK-Cu is usually started later, once visible change appears, rather than at GLP-1 initiation, and it runs independently of the weekly injection day — topical daily, or injected on a cycled basis in community reports.
What to watch
Route matters more than anything else in this pairing: the human safety and efficacy data is topical, and injectable GHK-Cu has no published clinical safety study. Injection also bypasses gastrointestinal copper regulation.
Skin quality, collagen, hair and general tissue remodeling — framed in community write-ups as surface plus systemic plus deep-tissue. No study of the triple exists; component evidence is topical-human for GHK-Cu and rodent for the other two (anecdotal / animal).
Scheduling
Typically run concurrently as a cycled block with three different cadences. It is also commonly started all at once, which is the classic attribution problem — if something changes, nothing identifies which compound did it.
What to watch
Three simultaneous new variables, and all three are pro-angiogenic to some degree, so the cancer-history caution stacks rather than sits parallel. Community consensus puts the practical ceiling at two to three concurrent compounds before injection count and timing constraints outrun any claimed benefit.
Blunting lean-mass loss during rapid weight loss. The lean-mass fraction of weight lost is trial-supported; the mitigation claim is mechanistic reasoning only — no adequately powered trial tests a peptide stack for this (human trials for the problem, theoretical for the fix).
Scheduling
Weekly GLP-1 plus two independent healing-peptide cadences. Community versions of this stack usually add a GH secretagogue with its own evening and fasting-window constraints — those compounds are not modelled in PepCraft yet, so a stack built here will be missing that leg.
What to watch
Every credible source in this space notes that resistance training and adequate protein have substantially stronger evidence than any peptide addition. GH secretagogues also affect glucose handling, which makes combining them with a glucose-lowering agent a monitoring question rather than a neutral addition.
Energy, cognition, 'cellular aging' and recovery. NAD+ repletion is biologically well-motivated, but oral precursors carry more human evidence for raising NAD+ than injection does, and a 2024 systematic review found insufficient high-quality evidence for routine subcutaneous NAD+ (human trials for precursors, anecdotal for the injected route).
Scheduling
NAD+ is usually run continuously rather than cycled, which leaves no off-period to attribute anything to; GHK-Cu is more often cycled. Some biohacker versions add a GLP-1 at low dose framed as a metabolic-health intervention — that is off-label and outside any approved indication.
What to watch
Infusion-rate-dependent effects are the documented flag: flushing, nausea, chest tightness, transient hypotension or tachycardia, more intense with rapid IV than subcutaneous. Chest tightness is not distinguishable from a cardiac presentation. No human lifespan or healthspan endpoint has been demonstrated.
Moving between incretin agents to escape a plateau or reduce side effects. Semaglutide and tirzepatide are approved individually; retatrutide is investigational. No trial supports concurrent use of two incretin agents (label / human trials).
Scheduling
Intended as sequential — one stops, the next starts, with spacing that accounts for the outgoing drug's half-life. Community threads routinely describe overlapping the tail of one with the start of another, which is exactly what the labels advise against.
What to watch
This is the highest-salience safety flag in the whole reference. Overlap risks additive GI toxicity, dehydration and hypoglycaemia — especially alongside insulin or a sulfonylurea. If two incretins show up as active protocols here, the flag report will say so.
Not a stack at all — the dominant tracking behaviour in the GLP-1 community, and the pattern most often recommended by users who have made attribution mistakes: one compound, observed for a stretch before anything else is introduced.
Scheduling
A fixed weekly shot day acts as the calendar anchor; the other tracked field is days since the last dose change, which correlates with symptom windows better than absolute dose does.
What to watch
GI symptoms are the most commonly reported issue and cluster in the days after each dose change rather than at random. Recording them against a single compound is the only version of this data that can ever answer a question.
Coming later — recognised in these threads but not modelled by PepCraft yet: Ipamorelin · CJC-1295 · Tesamorelin · Sermorelin · MOTS-c · AOD-9604 · NR / NMN (oral NAD+ precursors) · Cagrilintide / CagriSema · Survodutide · Mazdutide · Melanotan II · Semax / Selank · Epitalon · DSIP · KPV · LL-37 · PT-141.
Pairs with no record
No published interaction data — that means untested, not proven safe.
PepCraft flags a combination only when there is something published to flag. Silence in the table means nobody has looked, and it is never rendered as an all-clear.