Half-life is the central engineering problem of peptide therapeutics. A peptide is a short chain of amino acids, and the body is built to dismantle it: proteases cleave the backbone within minutes, and the kidneys filter small chains out of circulation. The native hormones that peptide drugs imitate are frequently gone from plasma faster than any practical dosing interval. Much of the second half of peptide drug development — from liraglutide to CJC-1295 to octreotide LAR — is a set of chemical and formulation techniques for buying time. This explainer sets out the main strategies, what each one changes pharmacologically, and why the difference between exposure and efficacy matters when you read an evidence grade.
01. Why native peptides are fragile
Two clearance mechanisms dominate. First, proteolysis: exopeptidases and endopeptidases in plasma and tissue cleave peptide bonds. Second, renal filtration: peptides below roughly 30–50 kDa are filtered by the glomerulus and degraded in the kidney. Native glucagon-like peptide-1 (GLP-1) has a plasma half-life of about 1.5–2 minutes, largely because dipeptidyl peptidase-4 (DPP-4) removes its first two residues [ref8]. Native somatostatin has a half-life of less than three minutes [ref6]. Native growth hormone is cleared with a half-life of roughly 15–20 minutes. A molecule with that profile cannot be a practical therapeutic without modification.
02. Lipidation and fatty-acid acylation
The most commercially successful strategy attaches a fatty acid chain to the peptide. Liraglutide carries a C16 fatty acid via a glutamic-acid spacer; semaglutide carries a C18 fatty diacid. The lipid tail binds serum albumin, and because albumin is a large, long-lived protein, the peptide is shielded from both DPP-4 and renal filtration while it rides along [ref1][ref2]. The result is a half-life measured in hours to days rather than minutes, which is what turns a native incretin into a once-daily or once-weekly medicine. The backbone is also modified at position 8 to resist DPP-4 directly — a reminder that these strategies are layered, not exclusive.
03. Albumin binding by other routes
Lipidation is not the only way to recruit albumin. CJC-1295 (also known as DAC:GRF) carries a maleimidopropionic acid group that forms a covalent bond with the cysteine-34 residue of circulating albumin, extending the action of a growth-hormone-releasing hormone analogue from minutes to days [ref3]. In the licensed-drug world the same principle appears as Fc-fusion (for example dulaglutide) and direct albumin fusion (for example albiglutide), both of which exploit the neonatal Fc receptor recycling pathway to extend circulation time.
04. PEGylation
Attaching one or more polyethylene glycol (PEG) chains increases hydrodynamic size, shields the peptide from proteases and antibodies, and pushes it above the renal filtration threshold. PEGylation has been applied to peptides and proteins such as PEG-MGF and to a range of licensed biologics [ref4]. The trade-offs are real: PEGylation can reduce receptor affinity and has itself become a regulatory and immunogenicity topic, which is one reason newer programmes often prefer lipidation or fusion.
05. D-amino-acid substitution and cyclisation
If proteases recognise specific stereochemistry, flipping it can slow cleavage. Introducing D-amino acids — the mirror-image forms of the standard L-amino acids — improves proteolytic stability and is used across research and licensed peptides [ref5]. Cyclisation (head-to-tail or side-chain bridges) rigidifies the backbone, which can improve both stability and receptor selectivity. These approaches change degradation kinetics and molecular shape rather than simply adding bulk.
06. Depot and sustained-release formulations
A different route to a long duration of action is the formulation rather than the molecule. Octreotide LAR encapsulates the peptide in slowly degrading polymer microspheres, releasing it over weeks; native octreotide has a plasma half-life of roughly 90–120 minutes, so the depot is what converts it into a monthly injection [ref6]. Leuprolide and other GnRH analogues are likewise supplied as depot implants and microsphere suspensions that sustain release over one to several months [ref7]. Here the peptide is unchanged; the delivery system controls exposure.
07. Exposure is not efficacy
A recurring problem in the grey-market literature is the conflation of a long half-life with a stronger effect. Engineering that extends plasma exposure changes how long a concentration is maintained, not whether the molecule does anything useful at the target. A long-acting analogue with no credible human endpoint remains an analogue with no credible human endpoint. This is the distinction Peptide Data's four-tier evidence grading is designed to preserve: pharmacokinetic convenience is not evidence of clinical benefit, and it should never lift a compound's grade on its own.
08. UK regulatory framing
In the UK, long-acting peptide analogues that are used as medicines — semaglutide, liraglutide, octreotide, lanreotide, leuprolide and similar — are licensed prescription-only medicines (POM) regulated under the Human Medicines Regulations 2012, with the Medicines Act 1968 providing the underlying framework [ref9]. Depot and sustained-release products are medicines, not research reagents. Where a peptide is sold "for research use only", the MHRA applies a case-by-case test of whether the product is in fact being placed on the market as a medicine; the label alone does not change its legal status [ref9]. Nothing in this explainer describes or endorses human use of these compounds — it is a summary of chemistry and pharmacology reported in the published literature, provided for research and educational purposes only.
This article is AI-researched and editorially reviewed. It is provided for research and educational purposes only and is not medical advice. Research peptides are not licensed for human consumption in the UK.