The same compound produces three different kinds of result depending on whether the system was a dish, an animal, or a piece of tissue kept alive on a bench — and the three do not rank in a hierarchy. In vitro shows what a molecule does to cells with everything else removed. In vivo shows what happens in a whole organism with everything else present. Ex vivo sits between them, and each answers a question the others cannot.
Three systems, three questions
| In vitro | Ex vivo | In vivo | |
|---|---|---|---|
| System | Cells or purified components in a dish | Tissue removed from an organism, studied alive | A living organism |
| Can establish | Mechanism, receptor engagement, potency | Tissue-level response with native architecture | Whole-system effect, distribution, clearance |
| Cannot establish | Whether the compound reaches the target in a body | Systemic effects; what other organs do | Which cell or receptor produced the effect |
| Concentration | Known and controlled | Known at the bath; unknown inside the tissue | Unknown at the target; inferred from blood |
| Confounds | Adsorption, degradation in medium, cell line identity | Tissue viability declining over hours | Everything: metabolism, feedback, behaviour |
Why is in vitro not simply the weakest evidence?
Because it is the only system in which the concentration at the target is actually known. An EC50 measured in a dish is a number about the receptor. The same compound in an animal produces a response at an administered amount, and the concentration at the receptor that produced it is inferred from blood levels, tissue penetration and clearance — none of which are measured directly in most studies. For any question about mechanism, in vitro is the stronger system, not the weaker one. It becomes weak only when its result is read as a prediction about a body, which it was never designed to be. A concentration–response curve is the canonical in-vitro result, and its strength is precision about a narrow question.
What does in vivo add that a dish cannot?
Whether the molecule survives, arrives and acts in a system that is trying to remove it. A peptide can be potent at a receptor in a dish and inert in an animal because it is cleared in minutes, never crosses a barrier, or is bound by a protein the dish did not contain. It can also be inert in a dish and active in an animal, because the effect runs through a tissue or a feedback loop the dish lacked. In vivo is the system in which the compound's fate and the organism's response are both present, and the cost of that completeness is that the result cannot be attributed to any one mechanism. Which organism then becomes the next question.
Where ex vivo fits
A tissue slice, an isolated organ, a strip of muscle in a bath: alive, with its native cell types and architecture, but removed from circulation, hormones and the rest of the body. The compound is applied at a known bath concentration, and the tissue responds as tissue rather than as a monolayer of one cell type.
It is the system for questions about how cells behave together — contraction, secretion from an intact gland, signalling between cell types — and it is limited by the clock. Tissue kept alive on a bench declines over hours, and a response measured late is a response from tissue that is dying. Most ex vivo results are honest snapshots; few are honest time courses.
Why do the three so often disagree?
Because they are measuring different things, and agreement was never guaranteed. A compound potent in vitro and inactive in vivo has usually failed at delivery, not at mechanism. A compound active in vivo and inert in vitro has usually acted through a system the dish lacked. A compound active ex vivo but not in vitro has usually needed a cell type the culture did not include. Each disagreement is informative about where the effect lives, and the error is to treat one system as having falsified another. A potentiator that needs its ligand present is a clean example: inert in a dish without HGF, and the dish result says nothing about the animal.
How should the concentration be reported in each?
In vitro, as molarity at the cells, with the medium's serum content stated, since degradation and albumin binding change what fraction is present. Ex vivo, as bath concentration, with the acknowledgement that tissue concentration is lower and unknown. In vivo, as the amount administered per unit of body mass together with the route, and, wherever it was measured, the resulting blood concentration — because the amount given and the concentration reached are different numbers, and only the second one connects to the in-vitro result. Papers that report the first and compare it to an EC50 have compared unlike quantities.
Which system does most of the research peptide literature use?
In vivo in rodents, followed by in vitro in cell lines, with ex vivo a distant third. That distribution shapes what is known: whole-animal effects in one species are well documented for many compounds, mechanisms are established for some, and tissue-level behaviour is often assumed from one or the other. For a compound with a named receptor, the in-vitro literature is where its identity is pinned down. For one without — most of the repair peptides, the bioregulators — the evidence is almost entirely in vivo, which means the effect is documented and its mechanism is not.
What does a well-designed research programme look like across the three?
It starts in vitro to establish that the compound engages its target and at what concentration, moves ex vivo where a tissue-level question exists, and goes in vivo only with a hypothesis about mechanism already in hand — so that the whole-animal result can be interpreted rather than merely observed. Running the sequence backwards, from an animal effect toward a mechanism, is how most of the peptide literature was actually built, and it is why so much of it consists of effects in search of an explanation. Neither order is wrong; the second simply leaves more unanswered, and reading it means holding that in view.
All products referenced here are supplied for laboratory and research use only. They are not drugs, foods, supplements or cosmetics, and are not for human or veterinary use.
