Peptide Half-Life: How It's Measured and Why It Matters

The short version

Terminal elimination half-life (t1/2) is the time it takes for the concentration of a drug in plasma to fall by half during the elimination phase. It determines how often a drug needs to be dosed, how long the body takes to reach a stable concentration (steady state), and how quickly concentrations fluctuate between doses.

For peptides, half-life is particularly consequential because native peptides typically have very short half-lives. Native GLP-1 is degraded by the enzyme DPP-4 within approximately 2 minutes of entering the circulation.^[1]A drug with a 2-minute half-life cannot be given once a week; almost nothing would remain by the next morning. The history of GLP-1 receptor agonist development is largely a story of engineering around that constraint, from liraglutide's once-daily dosing (half-life approximately 13 hours) to semaglutide's once-weekly dosing (half-life approximately 165 hours) and tirzepatide's once-weekly profile (half-life approximately 5 days, or 116.7 hours).^[2],[6],[7]

What terminal elimination half-life means

After a drug is absorbed into the systemic circulation, its plasma concentration rises to a peak (Cmax) and then falls as the body distributes, metabolizes, and excretes it. The terminal phase of that fall, once distribution is complete and only elimination is occurring, follows an approximately exponential decline. The half-life is the time constant of that decline: if plasma concentration is 100 ng/mL at time zero, it will be 50 ng/mL one half-life later, 25 ng/mL two half-lives later, and so on.^[8]

Two practical rules flow from this:

• Time to steady state is approximately five half-lives. With repeated dosing at a fixed interval, plasma concentrations accumulate until the rate of elimination equals the rate of input. This equilibrium (steady state) is reached after approximately five half-lives regardless of dose or dosing interval.
• Drug is largely eliminated after five half-lives. After five half-lives of a single dose, approximately 97% of the drug has been cleared. This matters for wash-out periods between treatments and for understanding how long an effect can persist after discontinuation.

How half-life is measured

Pharmacokinetic studies collect serial blood samples after dosing and measure drug concentration over time. The resulting concentration-time profile is then analyzed by one of two approaches:

Non-compartmental analysis (NCA)

NCA estimates key parameters (area under the curve, Cmax, t1/2) directly from the observed data without assuming a specific mathematical model of drug distribution. It is the most common approach for regulatory submissions and produces reliable estimates when you have enough time points in the terminal phase to characterize the slope accurately.

Compartmental modeling

Compartmental models represent the body as one or more mathematically described compartments (central, peripheral) and fit rate constants to the observed data. A one-compartment model assumes the drug distributes instantly throughout the body; a two-compartment model captures an initial rapid distribution phase followed by a slower terminal elimination phase. Compartmental analysis is more informative for drugs with complex distribution kinetics and allows simulation of dosing regimens.^[8]

Both methods require the terminal phase to be well characterized. For a peptide with a 5-day half-life like tirzepatide, this means collecting samples over several weeks. Assay sensitivity matters too: the peptide must be measurable at low concentrations for long enough to define the terminal slope accurately.

Why half-life matters for dosing

Dosing interval

The dosing interval is chosen to maintain plasma concentrations within a therapeutic window between doses. A drug with a half-life of 2 minutes requires continuous infusion or injection every few minutes to maintain any meaningful exposure. A drug with a half-life of 165 hours can be dosed once weekly and still maintain concentrations well above the minimum effective level at trough.

Time to steady state

For semaglutide with a half-life of approximately 165 hours (6.9 days), steady state is reached after approximately 5 x 6.9 = 34.5 days of once-weekly dosing.^[6]This is why dose escalation protocols for GLP-1 receptor agonists typically take several weeks per step: the body has not fully equilibrated to any given dose until 4-5 weeks have passed at that dose level. Clinical improvements that appear to plateau early may actually represent ongoing progress toward a new steady state that has not yet been reached.

Peak-to-trough variation

The ratio of peak concentration (Cmax) to trough concentration (Cmin) at steady state determines how much concentrations fluctuate within a dosing interval. A longer half-life relative to the dosing interval produces smaller fluctuations. For once-weekly semaglutide, the peak-to-trough ratio is modest because the half-life is close to the dosing interval. This contributes to a smoother pharmacodynamic profile compared with shorter-acting agents.

Half-life versus duration of action

Half-life and duration of action are related but not the same. Duration of action is how long a pharmacological effect persists; half-life is a measure of plasma concentration kinetics. The two diverge when:

• Target binding is irreversible or very slow to reverse. A drug that covalently binds its target can produce effects that outlast its plasma half-life by hours or days, because the drug-receptor complex must be replaced by newly synthesized receptors. Aspirin is the classic small-molecule example (platelet function inhibition for the platelet's lifetime despite the drug being cleared in hours).
• The active species is a metabolite. If the metabolite is pharmacologically active and has a longer half-life than the parent compound, the duration of action will exceed what the parent half-life predicts.
• Receptor internalization or resensitization is rate-limiting. After a GPCR agonist occupies its receptor, the receptor may internalize and become temporarily unavailable. The pharmacodynamic effect then depends on receptor recycling kinetics, not just plasma drug concentration.
• There is a downstream amplification cascade. A peptide might activate a signaling pathway that triggers transcriptional changes; the downstream effects could persist long after the peptide is cleared from plasma.

For most GLP-1 receptor agonists, plasma half-life and pharmacodynamic duration track reasonably closely because the receptor interaction is reversible and concentration-driven. But this should be confirmed for each compound rather than assumed.

How peptide half-life is extended

Modern peptide drug design employs several strategies to extend half-life beyond what a native sequence would have:

DPP-4 protection via Aib substitution

DPP-4 (dipeptidyl peptidase-4) cleaves the two N-terminal amino acids from many peptides, including native GLP-1, in seconds to minutes.^[1]Substituting the amino acid at position 2 with aminoisobutyric acid (Aib, alpha-methylalanine) creates steric bulk that prevents DPP-4 from engaging. Tirzepatide uses Aib at positions 2 and 13 of its sequence, and this is a primary reason it survives much longer in circulation than native GIP or GLP-1.^[7]

Albumin binding via fatty acid acylation

Serum albumin is the most abundant plasma protein and is itself recycled by the neonatal Fc receptor (FcRn), giving it a half-life of approximately 19 days. Conjugating a fatty acid chain to a peptide allows it to bind non-covalently to albumin, dramatically reducing renal clearance (albumin-bound peptides are too large to filter freely at the glomerulus) and protecting from proteolysis.^[4]Liraglutide uses a C16 fatty acid chain attached via a glutamic acid linker. Semaglutide uses a C18 fatty diacid chain via a more complex linker, giving it stronger albumin binding and a longer half-life. Tirzepatide uses a C20 fatty diacid linked to lysine at position 20, achieving its approximately 5-day half-life through the same mechanism.^[5],[7]

PEGylation

Attaching polyethylene glycol (PEG) chains to a peptide increases its hydrodynamic radius, reducing renal filtration and slowing proteolytic access. PEGylation is widely used in protein biologics (pegylated interferons, certolizumab pegol) and has been applied to peptides. The strategy adds molecular weight without adding pharmacophoric groups.^[3]

Depot formulation

Depot formulations slow the release of a drug from the injection site rather than slowing its elimination from plasma. Microsphere-based depots (such as those used for leuprolide acetate and some somatostatin analogs) embed the peptide in a slowly degrading polymer matrix. The apparent half-life measured in plasma then reflects release from the depot rather than true elimination half-life. Some long-acting somatostatin analogs achieve monthly dosing intervals through this approach.

Worked examples: native GLP-1 to tirzepatide

The progression from native GLP-1 to approved weekly GLP-1 receptor agonists illustrates how each engineering intervention compounds on the previous one:

Half-life values are population means from clinical pharmacology studies.^{[1],[2],[6],[7]}

Note that tirzepatide's half-life (approximately 5 days) is shorter than semaglutide's (approximately 7 days) despite both supporting once-weekly dosing. The slightly shorter half-life of tirzepatide is sufficient because its trough concentrations remain above the therapeutically relevant range throughout the week.^[7]

Summary

Terminal elimination half-life tells you how quickly a drug is cleared from plasma. Five half-lives to steady state and five half-lives to near-complete elimination are practical rules that apply to any compound following first-order kinetics. For peptides, the native half-life is often too short for clinical use; engineering strategies including DPP-4 resistance, albumin binding via fatty acid acylation, PEGylation, and depot formulation extend half-life by orders of magnitude.

Half-life and duration of action are related but distinct concepts. Target binding kinetics, metabolite activity, receptor trafficking, and downstream signaling can each cause the pharmacodynamic effect to outlast or fall short of what plasma concentrations alone would predict. Interpreting the clinical pharmacology of any peptide requires considering both dimensions.