The Croatian BPC-157 problem: why one lab matters

If you've spent any time reading the BPC-157 literature, you've noticed something strange: the same names keep appearing on the author lists. Predrag Sikiric. Sven Seiwerth. Rudolf Rucman. Branko Lovric. Marko Sever. And one institutional address keeps repeating: the School of Medicine, University of Zagreb, Croatia. A handful of researchers in one city have produced the overwhelming majority of the BPC-157 evidence base over thirty years. This is not nothing. It is also not the same as the molecule having a broad, independently-replicated evidence base — and the difference matters more than the strength-peptide community usually acknowledges.

This post is for users who already use BPC-157 or are considering it, and want a clear-eyed look at the research landscape rather than the marketing version of it.

The basic facts

BPC-157 (Body Protection Compound 157) was first described in the early 1990s by Predrag Sikiric and colleagues at the University of Zagreb. The peptide is presented as a pentadecapeptide derived from a "body protection compound" found in human gastric juice — though that derivation claim itself has been questioned by independent peptide chemists, and the precise source-molecule mapping is less well-established than vendor marketing implies.

Over the past three decades, the Sikiric group and a small network of close collaborators have published well over 300 papers on BPC-157 — covering tendon, ligament, bone, muscle, gut, cardiovascular, nervous-system, and many other systems. The breadth is extraordinary. The depth, in terms of independent replication, is much narrower than the publication count suggests.

A literature search filtered for BPC-157 papers from authors not affiliated with the Sikiric network turns up:

• Some preclinical work, mostly building on Sikiric-group methods
• A few independent toxicology studies
• Some commentary and review papers
• Almost no independent confirmation of the central efficacy claims in original animal models replicated by labs with no connection to Zagreb

This is the "Croatian lab problem." Not malicious — Sikiric and colleagues have done genuine research, much of it published in peer-reviewed journals — but a structural feature of the evidence base that affects how confidently you should treat any given BPC-157 claim.

Why single-source evidence is a problem

In any scientific field, independent replication is the mechanism that turns interesting findings into established knowledge. When one lab reports an effect, you have one observation. When three independent labs report the same effect using different methods, you have a robust finding. This is not a bureaucratic preference; it's how science excludes bias, methodological artifacts, and (occasionally) outright fraud.

The risks specific to single-source evidence:

Selection bias in publication. A lab that has built its identity around a compound has an incentive to publish positive results and not publish (or de-emphasize) negative ones. This isn't unique to Zagreb — it's a structural pressure in every research program — but it's amplified when one group dominates a field.

Methodological correlation. When the same group runs all the studies, the same animal models, the same dosing approaches, the same readouts get used over and over. If any of those choices subtly bias toward positive findings, every study inherits that bias.

Publication outlet patterns. Much of the BPC-157 literature has been published in journals where the editorial standards are mixed — including journals with documented quality issues. This doesn't invalidate the findings, but it does mean the peer-review filter has been less rigorous than it would be in top-tier journals.

Self-citation feedback. Sikiric-group papers cite Sikiric-group papers extensively. This is natural — building on your own work is normal — but it creates a citation graph that looks deeper than the underlying independent foundation supports.

For the broader frame on research literacy in this space see reading a COA: worked example and vendor due diligence checklist.

What we actually know vs. what gets claimed

What we can say with reasonable confidence:

• BPC-157 has some effect on tendon healing in rodent models. This has been observed in enough independent contexts (Sikiric and others) to suggest the basic effect is real, though magnitudes and clinical translation remain unclear.
• The peptide is well-tolerated at typical doses in animal toxicology — no major safety signals.
• Mechanism of action is incompletely characterized. Various pathways have been proposed (angiogenesis, growth-factor modulation, nitric oxide system, etc.), and the literature can't yet point to a single, established mechanism.

What gets claimed but is not well-supported:

• Specific dose-response curves for human use
• Efficacy in specific human injuries (tendinopathies, joint conditions) at specific doses
• Mechanism specificity ("works through pathway X for condition Y")
• Specific timeline claims ("works in 2 weeks for tendinopathy")
• Comparative efficacy vs. established treatments

The gap between what the literature supports and what gets claimed in athlete forums is wider for BPC-157 than for most other strength peptides. Some of that is normal community optimism; some of it is the single-source evidence base allowing claims to inflate without independent pushback.

What independent research has shown

A few independent groups have weighed in, with mixed results:

• Some confirmation of basic effects. Independent labs have observed pro-angiogenic and wound-healing signals in their own preclinical models, generally consistent with Sikiric-group findings on the broad mechanism.
• Limited human data. What human data exists is small-trial or case-report level, and is not enough to nail down dose-response or efficacy claims.
• Skepticism from regulatory bodies. The FDA's 2023 decision to remove BPC-157 from the 503A compoundable list cited concerns about the quality of existing human safety and efficacy data, among other factors. See BPC-157 503A compounding status and the related FDA PCAC review.
• Recent commentary calling for replication. A growing number of methodology-focused reviews have called for independent, large-scale studies of BPC-157 in defined human populations. None have yet been funded at scale.

The recent NewBeauty piece by Dr. Steven Pearlman about BPC-157 nose-slimming (see the BPC-157 nose-slimming news post) explicitly cited the single-source issue: "nearly all research comes from a single lab in Croatia." Independent clinicians are starting to flag this more visibly than they used to.

What this means for users

Practical implications, not a "stop using BPC-157" argument:

Discount confidence in specific claims. When a forum post or vendor page says BPC-157 "heals tendinopathy in 3 weeks at 500 mcg/day," recognize that the specificity outruns the underlying evidence. The peptide does something useful in tendons in some contexts; the rest is extrapolation.

Expect higher individual variability than the literature suggests. Cleanly-controlled animal models compressed into a single research program produce more consistent effects than real-world human use. Your response may be larger, smaller, or absent compared to community expectations.

Lean on safety being the strongest signal. The most robust BPC-157 finding across the literature is that it's well-tolerated at typical doses. The efficacy questions are more contested than the safety questions.

Don't treat BPC-157 as a substitute for proven interventions. For specific injuries, evidence-based interventions (physical therapy, image-guided injections, surgical consults where appropriate) have stronger evidence bases. BPC-157 is reasonably positioned as an adjunct with mechanistic plausibility, not a primary treatment.

Watch for replication. If a major independent trial is funded — which seems increasingly possible given the commercial interest in this molecule — the evidence picture could change rapidly in either direction.

For the broader frame see BPC-157 research evidence and the BPC-157 pillar guide.

What independent confirmation would look like

The kind of evidence that would dramatically increase confidence in BPC-157:

1. A large randomized controlled trial in a defined human injury population, run by a group unaffiliated with the Sikiric network, with prospective registration and pre-specified endpoints
2. Independent mechanism-of-action work from a structural biology / molecular pharmacology group identifying the actual molecular target(s) BPC-157 binds
3. Large independent toxicology and pharmacokinetic studies characterizing absorption, distribution, metabolism, and excretion in humans in detail
4. Comparative effectiveness studies vs. existing standards of care for specific conditions

None of these currently exist. Whether they ever will depends on who's willing to fund expensive trials on a molecule that's already commercialized in the research-chemical market without their results.

The honest framing

BPC-157 might work for the things people use it for. The mechanism is plausible, the safety profile in the available data is clean, and the preclinical signals are not nothing. But the evidence is heavily concentrated in one research network, the human data is thin, and the specific claims that circulate in the community outrun what's been independently established.

This is a reasonable peptide to use with informed expectations and conservative dosing. It is not a reasonable peptide to recommend with the confidence that a more deeply replicated molecule would warrant. If that distinction feels uncomfortable, it should — and the discomfort is the point. Knowing where the evidence is solid and where it isn't is part of what it means to use these compounds responsibly.

Related reading

• BPC-157 research evidence
• BPC-157 pillar guide
• Plastic surgeon pushes back on TikTok BPC-157 nose-slimming trend
• FDA BPC-157 503A compounding status
• Vendor due diligence checklist