In vitro vs in vivo tells you where a study happened. In vitro means testing outside a living body, often in cells or tissue. In vivo means testing in a living organism, such as a mouse or a person.

A study of human cells isn't a study of patients. And “in vivo” doesn't automatically mean tested in people. A peptide is a short chain of protein building blocks. These words locate the experiment; they don't tell us whether it helps someone.

In vitro vs in vivo: the official meanings

The National Cancer Institute defines in vitro as “In the laboratory (outside the body).” It defines in vivo as “In the body.”

Two related terms describe other settings. Ex vivo means testing removed tissue outside the body, often while preserving its original structure. In silico means analysis or modeling done with computers.

Term Where the work happens Example
In vitro Outside a living body Gut cells grown in a laboratory dish
In vivo In a living organism A mouse experiment or a study in people
Ex vivo In tissue removed from the body Testing a donated organ slice
In silico In a computer Analyzing existing cell data or simulating a drug's behavior

FDA's broader term, nonclinical, covers cell tests, computer models and nonhuman animal studies. These can happen before or during human trials. Section 3209 of Public Law 117–328 sets out that usage. Human cells can supply nonclinical evidence without any patient receiving treatment.

Why the distinction matters

Cells help researchers examine one process closely. A living body adds digestion, blood flow, organs and immune responses. Neither setting answers every question, and a human-cell model may reveal something an animal model misses.

Moving a finding into patient care takes more evidence. A 2024 review in PLOS Biology covered 122 reviews and 367 therapies. Among 281 therapies first studied in animals more than 10 years earlier, about one in 20—5%—had reached regulatory approval. That describes a selected collection, rather than every peptide's chances.

The BPC-157 tendon literature includes a 2003 study of rats with surgically cut heel tendons and separate experiments with tendon cells grown in a laboratory. Those are animal and laboratory findings; they don't establish a treatment result for someone's tendon pain.

How these terms show up with peptide blends

A single peptide paper can contain several kinds of evidence. Reading the methods, the section explaining what researchers did, helps separate them.

KPV research includes human gut and immune cells grown in a laboratory, plus mice with bowel inflammation. The human origin of those cells doesn't turn the experiment into a human treatment trial.

The GHK-Cu “4,000 genes” claim comes from computer analysis of GHK responses in cancer-cell cultures. GHK is the peptide portion of GHK-Cu, a peptide bound to copper. Gene expression means how actively cells read their genetic instructions. A changed reading isn't a rewritten gene or a measured change in someone's age.

KLOW combines BPC-157, TB-500, GHK-Cu and KPV. A finding about one ingredient doesn't test that complete blend. In a prescription discussion, the question becomes which findings concern an ingredient and which actually concern the combination.

Four mistakes that change what a study means

Reading a cell concentration as a blood level. A cell-culture concentration is the amount added to fluid around cells. It doesn't establish what medication amount would produce that level in someone's blood. Absorption, movement through tissues and breakdown all need separate study.

Converting an animal dose by body weight alone. FDA's July 2005 starting-dose guidance discusses body-surface-area scaling, a method that considers body size and shape. It also considers the relevant species and a safety margin. That helps researchers plan an initial trial dose in healthy adults; it doesn't supply a routine prescription dose.

Treating “studied” as “shown in people.” A study may measure cell movement, a blood marker or an animal's performance. Those findings don't establish better symptoms or function in patients. The measured outcome matters as much as the setting.

Assuming no observed harm means no human risk. An experiment can be too short or too small to reveal uncommon harms. FDA's guidance adds a safety margin because people may be more sensitive than the animal model suggests. An absence of reported reactions isn't a guarantee about longer use.

Why FDA is changing the words

The change commonly called the FDA Modernization Act 2.0 was enacted December 29, 2022, within Public Law 117–328. It broadened testing options and kept animal tests among them.

FDA released its Roadmap to Reducing Animal Testing in Preclinical Safety Studies on April 10, 2025. It describes using validated alternatives, meaning methods checked for the question they need to answer. Examples include computer models and miniature human-organ systems grown in laboratories.

As of October 1, 2026, the day this article was written, FDA had published its Nonclinical Testing Terminology rule on September 22. The September 21 announcement says it preserves existing evidence standards and keeps animal studies available. The rule lists February 4, 2027 as its effective date, subject to withdrawal if significant adverse comments arrive. Clearer wording doesn't turn laboratory findings into treatment results in people.

Animal studies vs human trials: the MOTS-c example

MOTS-c was studied in cells and mice in Lee and colleagues' 2015 Cell Metabolism paper. Researchers examined metabolism, meaning how the body processes and uses fuel.

Later human work can mean observation. A 2021 Nature Communications study measured the body's own MOTS-c in 10 men around exercise. It didn't give MOTS-c to them; treatment experiments in that paper involved mice. Measuring a naturally occurring signal doesn't establish what adding it as medication would do.

A prescribed preparation adds another question: does the evidence apply to that exact medication? Compounding means a pharmacy prepares medication for a prescription. FDA's compounding explanation distinguishes that process from review of a finished product.

Promise's compounded medications are different from FDA-approved products: the preparations offered here are not FDA-approved. A licensed provider may still prescribe — that decision is between you and your doctor.

Preclinical usually describes laboratory and animal work before human testing. Nonclinical is broader because that testing can continue alongside human studies.

Clinical trial phases describe stages of testing in people. An IND, or Investigational New Drug application, supports a proposed human investigation. Relative risk compares how often an event occurs in two groups. Bioavailability describes how much drug becomes available at its site of action and how quickly.

Together, these terms help answer a concrete question: was this measured in cells, animals or people, and what was actually measured?

At Promise, a licensed provider reviews every request, and not everyone qualifies.