# Three Compounds, Three Clocks, One Missing Human Endpoint

> Compare BPC-157, TB-500 and Wolverine — Recovery & Tissue Repair Research Peptides — A side-by-side timeline comparison of three Recovery & Tissue Repair research peptides — BPC-157, TB-500 and the Wolverine blend — across half-life, follow-up length, species, dose-response shape and repair phase.

**RECOVERY & TISSUE REPAIR / MATRIX**

The same literature sorted by time rather than by claim — including the column most comparison tables leave out, which is the species the interval was measured in.

## The short version

This page lines up [BPC-157](/bpc-157), [TB-500](/tb-500) and the [Wolverine blend](/wolverine) on the dimensions that matter when a reader wants to know *how long* rather than *whether*.

The headline is uncomfortable but clean. BPC-157 has the most evidence and no measured recovery interval. TB-500 has the only dated recovery curve in this corpus, and it was measured in rats using a molecule more than five times larger than the one sold under the name. The Wolverine blend has no measurement at all — a systematic review that screened 36 BPC-157 studies did not encounter the combination once [13].

One fact links every row of the table below: not a single interval in it comes from a controlled human healing study. There is no such study for any of these three. What human data exist measure tolerability over days or weeks, in healthy volunteers, and were never designed to describe repair. Nothing on this page is advice, and no dose here is a recommendation.

## The timeline matrix

| Dimension | BPC-157 | TB-500 | Wolverine (BPC-157 + TB-500) |
| --- | --- | --- | --- |
| What it is | 15-amino-acid pentadecapeptide from a gastric-juice protein fragment | 7-amino-acid fragment (Ac-LKKTETQ) of the 43-residue protein thymosin beta-4 | Co-formulated pairing of the other two; not a single molecule |
| Measured half-life | Under 30 minutes; linear kinetics; IM bioavailability ~14-19% (rats), ~45-51% (dogs) [3] | Not measured for the fragment. Full-length parent protein: dose-proportional, half-life rising with dose [11] | Not measured |
| Species the clock was measured in | Rats and beagle dogs [3]; Wistar rats [5][6] | Male Wistar rats [9]; healthy human volunteers for the parent protein only [11] | No species — no combination study exists [13] |
| Longest dated follow-up in this corpus | None reported; endpoints scored without an onset or persistence window [5][6] | Day 14 through day 56, rat embolic stroke [9] | None [13] |
| Onset of effect | Not systematically measured in any species | Not detectable at early assessments; significant from day 14 in rats [9] | Never measured |
| Human evidence | Three pilot studies total; one enrolled two adults [1][2] | No completed controlled trial of the fragment; a 14-day Phase 1 of the parent protein in 40 volunteers [11] | None; a 36-study systematic review did not mention the combination [13] |
| Dose-response shape | Higher doses gave a larger ulcer-inhibition ratio in rats (45.7-65.6%) [5] | Non-monotonic: benefit at 2 and 12 mg/kg, none at 18 mg/kg, modelled optimum ~3.75 mg/kg [9] | Undefined; no ratio or dose established |
| Repair phase implicated | Proliferative: angiogenesis via VEGFR2, granulation tissue [4][5] | Earliest: platelet release at injury, actin buffering for cell migration, reduced scarring [10][12] | Both, in theory only; timing between them unstudied |
| Regulatory status | Not FDA-approved; WADA-prohibited at all times (S0); research use only | Not FDA-approved; WADA-prohibited; prescription medicine in some jurisdictions | Neither component approved; both WADA-prohibited |
| Single largest caution | Human evidence extremely thin and heavily single-group [2] | The evidence belongs to a molecule five times larger than the one sold [10] | The pairing has never been tested at all [13] |

Every interval in this table is quoted from the cited source, with the species it was measured in stated alongside it.

## Reading the species column first

The species column is placed high in the table deliberately, because it changes the meaning of every other row.

The tissue-repair peptide literature is overwhelmingly animal work. BPC-157's pharmacokinetics were characterised in rats and beagle dogs [3]. Its tendon result is a fully transected Achilles tendon in Wistar rats [6]. Its ulcer result is a rat gastric-ulcer model [5]. Its angiogenesis result spans chick chorioallantoic membrane, rat hindlimb ischaemia and human cells in culture [4]. The TB-500 recovery curve is a rat embolic-stroke model [9]. A 2025 systematic review of BPC-157 in orthopaedic sports medicine found 35 of its 36 included studies were preclinical [13].

Animal repair timelines do not transfer to people by rescaling. Rodents and humans differ in healing rate, in tissue architecture, in the load a tendon carries and in how an injury model is created — a surgically transected rat tendon is a controlled, clean, uniform injury, which is precisely what makes it a good experiment and a poor analogue for a human tear. A benefit measurable from day 14 in a rat is not a claim about day 14 in a person, and this desk does not present it as one.

When a timeline on this site was measured in an animal, the sentence says so. When it comes from a community account, the sentence says that instead, and marks it anecdotal. When neither applies, the sentence says the measurement does not exist.

## Where the three genuinely differ

**Evidence maturity.** BPC-157 has by a wide margin the largest body of work, and it is nearly all preclinical: a 2025 review put the human record at three pilot studies [2], and a 2025 systematic review found no clinical safety data across 36 included studies [13]. TB-500's fragment has no completed controlled human trial for any indication; its parent protein has a completed Phase 1 in 40 healthy volunteers [11]. The blend has nothing.

**Mechanistic resolution.** TB-500 is ahead here. Its parent protein's action is fixed by a 2 ångström crystal structure showing one-to-one sequestration of monomeric actin by capping both ends of the monomer [12] — an unusually exact anchor for this field. BPC-157's mechanism is inferred from functional experiments rather than structure, with VEGFR2 up-regulation and internalisation the best-characterised route [4].

**Dose-response behaviour.** This is the sharpest practical difference. In rats, BPC-157's ulcer-inhibition ratio rose with dose across the tested range [5]. Thymosin beta-4 in the rat stroke model did the opposite at the top end: 2 and 12 mg/kg improved neurological function, 18 mg/kg did not, and the modelled optimum was around 3.75 mg/kg [9]. A single non-monotonic result is not a general law, but it is the only formal dose-response curve in this corpus, and it points against the assumption that more produces more.

**Distance between the studied molecule and the sold one.** Zero for BPC-157, which is studied as itself. Large for TB-500, where the efficacy literature used the full-length protein [10]. Largest for the blend, which inherits TB-500's gap and adds an untested pairing on top [13].

## The shared blank space

What the three have in common is more consequential than what separates them.

None has a controlled human trial measuring time to healing. None has a characterised duration of effect after a single administration in any species. None has data on what happens when use stops. None has long-term human safety data. And all three sit outside regulatory approval, with two review-level verdicts saying so in similar terms: BPC-157 should be treated as investigational because rigorous large-scale trials are lacking [2], and unapproved musculoskeletal peptides as a class show animal-model promise alongside scarce human safety data, potential for serious harm and operation largely outside regulatory oversight [8].

That shared blank is the honest conclusion of a timeline comparison. The measured record describes what happened in rodents over days and weeks, and what happened to healthy human volunteers' safety markers over a fortnight [11]. It does not describe how long a person takes to heal on any of these compounds, because nobody has measured it. The [reference list](/references) gives every source behind this table in full.

---

A reference desk for the tissue-repair peptide literature kept on the record's own clock: it reports what was measured, when, and in which species, and it neither sells anything nor advises anyone.