Every other category on this site is sorted by biology — by receptor, by tier of an axis, by hallmark of aging. The blends shelf is sorted by something else: these are the products that contain more than one active compound in a single vial.
That is a manufacturing fact, not a mechanistic one, and it introduces an attribute no single-compound product has. A blend is not just its components. It is its components in a specific proportion — and that proportion is the one thing on the label routine documentation does not confirm.
A previous foundations piece covered the rationale for combining compounds and the blend-versus-stack distinction. This is the companion piece about the shelf itself.
What Is Actually on the Shelf
Five entries, in three groups.
One metabolic co-formulation. CagriSema pairs cagrilintide, a long-acting amylin analog, with semaglutide, a GLP-1 receptor agonist. Two separate receptor systems, two separate molecules, one vial.
One growth-hormone pairing. CJC-1295 with ipamorelin — a GHRH analog alongside a ghrelin-mimetic secretagogue. This is the textbook two-receptor case: GHRH-R signalling through Gs/cAMP and GHS-R1a through Gq/PLC, converging on the same pituitary output. The mechanistic argument is laid out in the GH secretagogue axis explainer.
Three nested repair blends. BPC-157 + TB-500 is the base pairing. Adding GHK-Cu gives the three-component version, and KPV the four. These are a series, each built by appending one compound to the last — additive by construction, not the result of any experiment comparing the three against each other.
Sorted by evidence maturity the spread is wide. One member has a large registered clinical program. The others rest on single-agent extrapolation.
The Ratio Is the Product
Read the label strengths as proportions rather than amounts and the shelf's real specification becomes visible. The two-component blends are 1:1. The three-component repair blend is 1:1:5, with GHK-Cu at five times the proportion of each of the other two. The four-component version is 1:1:5:1.
Those ratios are the product definition. Change them and you have made a different product under the same name.
Is that a pharmacological variable or a packaging detail? There is now a clean piece of evidence that it can be the former. In JACS Au 2026;6(3):1517-1528 (doi 10.1021/jacsau.5c01071), a group at the University of Tokyo's Institute of Industrial Science examined what happens when two antimicrobial peptides occupy the same solution. One of them is a catalog compound: LL-37, the human cathelicidin. The other is human α-defensin-1, which is not.
Their finding is that the two peptides heteroaggregate — they form mixed assemblies rather than each behaving as it would alone. Explicitly drawing on the cross-seeding described for α-synuclein and β-amyloid, the authors report that reduction of cytotoxicity toward MDCK-I cells occurred only at a stoichiometric ratio of α-defensin-1 to LL-37 of approximately 0.025. Outside that narrow window, the effect was not observed.
Two things are worth holding onto. First, the outcome was ratio-dependent to a degree far finer than any label on this shelf specifies. Second, the direction was mutual antagonism, not synergy: α-defensin-1 suppressed LL-37-induced necrosis while LL-37 reduced α-defensin-1-induced apoptosis. Each blunted the other.
This is a cell-culture study, and it is about a pair of compounds that appear in no product here. It predicts nothing about the actual blends. What it establishes is narrower and more useful: two peptides in one solution can form a species that is neither of them, and the proportion can decide the result. That is a physical property of mixtures. It is not addressed by knowing what each component does alone.
One Purity Number, Two to Four Actives
Every blend on this shelf carries a single purity specification: 99% or greater by HPLC.
Run the arithmetic. A correctly manufactured 1:1 co-formulation of two pure peptides, injected onto a reversed-phase column, produces two well-separated peaks of broadly comparable area. Neither is 99% of the total. The chromatogram of a correct blend, read as a purity figure, looks like a failed purity test.
So the number cannot be describing the mixture. At best it is a specification carried over from the individual components before they were combined — which may be perfectly honest, and which is a claim about different material than what is in the vial. A single area-percent figure has no clean referent for a deliberate multi-component product.
This is a different problem from the one on the HMG and Thymalin entries, where purity has no referent because the material is an undefined biological fraction. Here the components are fully defined molecules. The number is not meaningless in principle — it is just not a property of the blend.
What would answer the question is mass spectrometry against stated theoretical masses for every component, and that matters more here than anywhere else on the site, because two blend entries carry an unresolved identity before analysis even starts:
- TB-500 is a name, not a sequence. As covered in the BPC-157 vs TB-500 comparison, the name is applied both to full-length thymosin β4 (~4960 Da) and to the Ac-LKKTETQ heptapeptide (~889 Da). Three of the five blends contain "TB-500." Intact mass resolves which one trivially — a roughly 4000 Da difference is not subtle — but only if the expected mass is written down.
- The GH blend does not state whether its CJC-1295 component is the DAC or No-DAC form. Those differ by an albumin-binding linker and, as covered in that comparison, by a half-life running from roughly half an hour to several days. Same target, very different duration.
There is also a plain chromatographic risk that grows with component count: two species can co-elute. On a multi-active product, an unresolved pair reads as one peak, and no purity figure flags it.
What Else Shares the Vial
Combining compounds means their non-target properties meet as well.
Charge. BPC-157 is a net-anionic sequence. KPV is net-cationic. GHK is cationic as the free peptide. The charge and solubility primer covers why this governs solubility, retention and surface adsorption for a single peptide; a co-formulation puts oppositely charged species in the same solution, where electrostatic association is a possibility rather than a certainty. The honest statement is that it is an unexamined question, and the Tokyo heteroaggregation work is a reminder that such questions have measurable answers.
Metal speciation. As covered in the GHK-Cu deep dive, the copper in GHK-Cu is an equilibrium rather than a covalent component. Lau and Sarkar's potentiometric and spectrophotometric work (Biochem J 1982;199:649-656) mapped multiple species across pH and showed GHK forming ternary complexes with an additional ligand. A blend introduces other peptides — other potential donor atoms — into that equilibrium. Copper content is measured by ICP-MS or atomic absorption, a separate instrument answering a separate question, and its absence from a blend's documentation is a real gap.
The cake is a mixture. A blend is co-lyophilized, so the solid is a multi-component amorphous glass whose glass transition temperature belongs to the formulation rather than to any single ingredient. Work on co-lyophilized insulin systems in Eur J Pharm Sci 2026;218:107450 reports glass transition and onset glass transition humidity as formulation-level parameters determined experimentally for each composition. That study concerns excipients rather than peptide-peptide blends, but the principle transfers: a mixture's solid-state behaviour cannot be inferred from its components separately. Shelf life is set by whichever component fails first, and these stability profiles are not alike — BPC-157 is aspartate-rich, GHK-Cu carries a redox-active metal, and aggregation is the one degradation pathway that is not a covalent reaction and so cannot be anticipated from synthesis-time purity.
CagriSema Is the Shelf's Control Case
One member shows what testing a combination actually costs.
The CagriSema clinical program runs to dozens of registered trials. Two are worth naming because of what they were built to answer.
NCT06716307 was a Phase 1 randomized crossover in 18 participants, completed March 2025. Its primary outcomes were the area under the curve and maximum concentration of cagrilintide and semaglutide, measured separately, across two different presentations of CagriSema at the same dose. A sponsor ran a dedicated pharmacokinetic trial simply to compare two versions of one combination. That is not routine caution. It is an acknowledgement that how two peptides are presented together is capable of changing what happens to each.
NCT06409130 was a Phase 2 quadruple-masked randomized trial in 270 participants. Its arm structure included CagriSema, cagrilintide with semaglutide placebo, and double placebo. That design exists so that an effect can be attributed — to the combination, to one component, or to neither. Without those arms, a combination result is uninterpretable, a point this site has now made about factorial designs three times in different contexts.
No blend on this shelf other than CagriSema has anything of this kind, and CagriSema itself remains an investigational combination. Two further registered studies comparing CagriSema presentations were withdrawn before enrolling anyone, which is its own reminder that a registration is an intention, not a result.
How to Read This Shelf
A blend answers a convenience question and creates an evidence question. Whatever is observed cannot be attributed to any one component — the confound is built into the product.
Three questions do most of the documentation work. Which molecule is each component, stated by sequence and expected mass rather than by trade name. What ratio is claimed, and on what evidence. And which failure modes this specific combination can produce that neither component produces alone, since a test only means something if it can detect a failure the material is capable of.
More on documentation is at /quality/; the compound reference is at /library/.
Frequently Asked Questions
Does a blend save anything analytically? No — it costs. One document now covers two to four molecules, purity has no single clean referent, identity requires a stated theoretical mass per component, and stability is governed by the least stable member. A blend is one product and several characterization problems.
Is the LL-37 finding a reason to avoid blends? It is not about any product on this shelf, and it should not be read as one. It is evidence for a general claim: peptide-peptide interactions in a shared solution are real, measurable, ratio-sensitive, and not always in the direction the combining rationale predicts. The relevant experiments have not been run on these combinations.
Why does CagriSema sit on the same shelf as the repair blends? Because the shelf is defined by co-formulation, not by evidence. That is exactly the reading error this site's category primers exist to prevent: the label describes what is in the vial, not how much anyone knows about it.
This article is educational and for the laboratory research community. Trulogic Labs products are sold for laboratory and research use only and are not for human consumption.