Pharmacokinetics: how long a peptide stays in the body, and why

• Bioavailability: the share of a dose that reaches the bloodstream unchanged. It is measured against an injection straight into a vein, which counts as 100%.¹
• First-pass metabolism: breakdown before a medicine reaches the general circulation, for example in the gut wall or liver after swallowing.¹
• Clearance: the volume of plasma (the liquid part of the blood) that is cleared of the medicine per unit of time. Total clearance adds up what the liver, the kidneys and other routes remove.¹
• Volume of distribution: a theoretical measure of how far a medicine has moved from the blood into tissues. Binding to blood proteins and tissues affects how a medicine moves through the body.¹
• Elimination covers both breakdown and excretion; excretion is only the physical removal.¹

Semaglutide shows how these fit together. More than 99% of it in the blood is bound to albumin. It is broken down by enzymes that cut its peptide backbone (one of them is called neutral endopeptidase) and by breaking down its fatty side chain, and only about 3% leaves in the urine unchanged.⁵

Each half-life closes half the gap

When a medicine is used repeatedly, it builds up until the amount removed between doses equals the amount given. This balance is called steady state.¹ Because the body removes a fixed share in each half-life, each half-life closes half of the remaining gap to that balance. The same arithmetic runs in reverse after the last dose.

This is why the EU label for semaglutide tablets states that, with a half-life of about one week, steady state is reached after four to five weeks of daily use.⁵ It is also why semaglutide stays in the circulation for about five to seven weeks after the last dose.^8,5

Why a long half-life cuts both ways

• Fewer injections and steadier levels. A long half-life allows weekly use, and daily tablets with a long half-life give fewer ups and downs from day to day.^5,8
• Side effects last long too. After an overdose of semaglutide, the EU label warns that a prolonged period of observation and treatment may be needed because of its one-week half-life. There is no specific antidote.⁸
• Slow to start, slow to stop. It takes weeks to reach steady levels, and weeks for them to fade. Because of the long half-life, the EU label says to stop semaglutide at least two months before a planned pregnancy.⁵

Enzymes everywhere

The amide bonds that link amino acids are easily cut by enzymes in the body when nothing protects them.¹⁰ Proteases in blood, liver and kidney each prefer particular amino acids, so where they cut can be predicted from the sequence.⁶ In lab tests with human kidney preparations and liver fractions, the synthetic peptide TB-500 was broken down actively, by cleavage of all its peptide bonds Cells(Evidence level: Shown in cells or tissue in the laboratory).¹¹

The kidney filter

The kidney’s filter holds back molecules by size and charge, letting only small amounts of albumin through.¹² Peptides are generally filtered freely, and filtration plus breakdown in the kidney remove many peptide medicines.³ Exenatide (Byetta, authorised in the EU until 1 January 2026) is an example: in animal studies it was cleared mainly by kidney filtration followed by breakdown, and its half-life in people was 2.4 hours.¹³

Where the BPC-157 numbers come from

The half-lives above come from one study in which rats and dogs received a single injection into a vein. The FDA notes that BPC-157 resisted gastric juice in a lab test, so its short half-life in the body is probably due to enzymes in the blood or tissues.^9,14 Animal half-lives do not translate directly to people, and no human measurement has been published (see the last section of this page).

• Albumin binding through a fatty acid. Albumin is a large blood protein that the kidneys hardly filter. A fatty-acid chain makes the peptide cling to it; bound peptide escapes kidney filtration and is protected from breakdown.^12,5,10 Liraglutide adds a second trick: its molecules cluster under the skin and are absorbed slowly.¹⁶
• Fusion to an antibody part (Fc). Dulaglutide joins two GLP-1-like chains to part of a human antibody. Its large size slows absorption and reduces kidney clearance.⁷
• PEG chains. Attaching chains of polyethylene glycol (PEGylation) increases a molecule’s size and shields it from protein-cutting enzymes.¹⁸
• A reactive hook for albumin. CJC-1295, an experimental growth-hormone-releasing peptide, carries a reactive chemical group that bonds to albumin after injection. In the developer’s own study in rats (numbers not given in the abstract), it was found bound to albumin 15 minutes after injection and was still in the blood beyond 72 hours Animals(Evidence level: Shown in animal studies).¹⁹ In two randomised, placebo-controlled trials in healthy adults its estimated half-life was 5.8 to 8.1 days.²⁰

Every trick changes the molecule. A longer half-life also means that a side effect, or an error in the amount used, lasts longer.⁸

• Size. Large molecules are absorbed more slowly. The EU label for dulaglutide says its large size slows absorption.⁷ The sheep study above found more lymphatic absorption with every step up in molecular weight, from 246 to 19,000.²
• Clustering. Liraglutide molecules cluster together under the skin, which releases them slowly.¹⁶
• Site. For semaglutide, the label reports similar exposure after injection in the abdomen, thigh or upper arm.⁵

Bioavailability after an injection under the skin is a property of each product, measured in trials. It cannot be assumed for a molecule that has never been studied in people.

What the regulators found

In 2026 the FDA evaluated several unapproved peptides that had been put forward for use in US pharmacy compounding, including these three. For BPC-157 it found no human pharmacokinetic data after use by mouth, under the skin, in the nose or through the skin. It noted that in two human studies of BPC-157 given as an enema, the authors reported that it was not detected in the blood (one abstract said “most” levels were too low to measure).⁹ For TB-500 and KPV it found no human studies of pharmacokinetics by any route.^21,22 Because BPC-157’s absolute bioavailability by the proposed routes was not known even in rats and dogs, the FDA could not use the animal safety data to estimate no-harm levels for those routes.⁹

What follows from it

• Animal half-lives do not transfer. A peptide that lasts minutes in a rat may behave differently in a person; that is the subject of the translation gap.
• The product adds its own uncertainty. A vial from outside the legal supply chain may hold a different amount, a different salt form or a different substance from its label, which changes everything above (see product risk).
• A doctor can still help. Even without data on the substance, a doctor can check for the effects that matter, such as blood sugar or liver tests, and record what happened. See monitoring.

This gap has a cost in both directions. People who use these products cannot know how long an effect lasts, and researchers cannot tell whether a promising animal result could ever work in people until someone measures it.