Peptide pharmacokinetics (PK) exhibit unique characteristics: target-mediated drug disposition (TMDD), nonlinear clearance, and immunogenicity effects. This guide covers modeling approaches from empirical PK to physiologically-based models.
1. Peptide PK Characteristics
Section titled “1. Peptide PK Characteristics”Fundamental Properties
Section titled “Fundamental Properties”| Parameter | Typical Range | Factors |
|---|---|---|
| Half-life | 2–30 min (native), 4–24 hr (modified) | Enzymatic degradation, renal clearance |
| Bioavailability | 50–90% (SC), 100% (IV) | Injection site, formulation |
| Volume of distribution | 3–15 L | Tissue binding, hydrophilicity |
| Clearance | 5–20 mL/min/kg | Renal, hepatic, enzymatic |
Nonlinear PK Mechanisms
Section titled “Nonlinear PK Mechanisms”- TMDD: Receptor binding saturates at high doses
- Enzymatic degradation: Saturation of proteases
- Renal reabsorption: Carrier-mediated transport
- Immunogenicity: ADA formation alters PK
2. Compartmental Models
Section titled “2. Compartmental Models”One-Compartment Model
Section titled “One-Compartment Model”- A = amount in body
- CL = clearance
- V = volume of distribution
- Solution: , where
Two-Compartment Model
Section titled “Two-Compartment Model”- A₁ = central compartment (blood, rapidly equilibrating tissues)
- A₂ = peripheral compartment (slowly equilibrating tissues)
Three-Compartment Model
Section titled “Three-Compartment Model”For peptides with significant tissue distribution:
- Central compartment
- Rapidly equilibrating peripheral (muscle, organs)
- Slowly equilibrating peripheral (fat, bone)
3. Noncompartmental Analysis (NCA)
Section titled “3. Noncompartmental Analysis (NCA)”Key Parameters
Section titled “Key Parameters”| Parameter | Formula | Units |
|---|---|---|
| AUC₀₋∞ | Trapezoidal + extrapolation | ng·h/mL |
| C_max | Observed maximum | ng/mL |
| T_max | Time of C_max | h |
| t₁/₂ | ln(2)/λ_z | h |
| CL/F | Dose/AUC | mL/min/kg |
| Vd_ss | Dose·AUMC/AUC² | mL/kg |
| MRT | AUMC/AUC | h |
NCA Software
Section titled “NCA Software”- WinNonlin (Certara): Industry standard
- Phoenix: WinNonlin successor
- PKNCA (R): Open-source alternative
4. Population PK (PopPK)
Section titled “4. Population PK (PopPK)”PopPK Framework
Section titled “PopPK Framework”Advantages:
- Handles sparse sampling
- Quantifies inter-individual variability (IIV)
- Identifies covariates
- Supports dose optimization
Structural Models
Section titled “Structural Models”Base model: Two-compartment with first-order absorption
Covariate Model
Section titled “Covariate Model”Where:
- θ = fixed effect parameter
- WT = body weight
- η = random effect (IIV)
Covariate Selection
Section titled “Covariate Selection”| Covariate | Effect on PK | Clinical Relevance |
|---|---|---|
| Body weight | ↑ CL, ↑ Vd | Dose adjustment |
| Renal function (eGFR) | ↑ CL if renal elimination | Dose adjustment |
| Age | Variable | Pediatric/geriatric |
| Sex | Minimal for most peptides | Usually no adjustment |
| ADA status | ↑ CL, ↓ efficacy | Monitoring |
Software
Section titled “Software”- NONMEM: Gold standard for PopPK
- Monolix: Alternative, SAEM algorithm
- nlmixr2: R-based, free
- PsN: Perl-speaks-NONMEM
5. Target-Mediated Drug Disposition (TMDD)
Section titled “5. Target-Mediated Drug Disposition (TMDD)”TMDD Concept
Section titled “TMDD Concept”For peptides with high-affinity receptor binding:
- At low doses, most drug is bound to target → TMDD kinetics
- At high doses, target is saturated → linear PK
Full TMDD Model
Section titled “Full TMDD Model”- C = free drug concentration
- R = free receptor concentration
- RC = drug-receptor complex
- k_on, k_off = binding kinetics
- k_int = internalization rate
Quasi-Steady State (QSS) Approximation
Section titled “Quasi-Steady State (QSS) Approximation”When k_off >> k_on·C + k_int:
Where
Michaelis-Menten Approximation
Section titled “Michaelis-Menten Approximation”When receptor recycling is fast:
Where and
6. Physiologically-Based PK (PBPK)
Section titled “6. Physiologically-Based PK (PBPK)”PBPK Framework
Section titled “PBPK Framework”Advantages:
- Predicts tissue concentrations
- Supports allometric scaling
- Handles drug interactions
- Supports special populations
Compartment Structure
Section titled “Compartment Structure”| Compartment | Volume | Blood Flow | Peptide Considerations |
|---|---|---|---|
| Gut lumen | Variable | — | Oral absorption |
| Liver | 1.5 L | 1.5 L/min | Metabolism, extraction |
| Kidney | 0.3 L | 1.2 L/min | Filtration, reabsorption |
| Muscle | 30 L | 0.7 L/min | Peripheral distribution |
| Fat | 15 L | 0.3 L/min | Lipophilic peptides |
| Brain | 1.4 L | 0.55 L/min | BBB transport |
PBPK Software
Section titled “PBPK Software”- Simcyp: Industry standard
- PK-Sim: Open-source
- GastroPlus: Oral absorption focus
7. Immunogenicity Effects on PK
Section titled “7. Immunogenicity Effects on PK”ADA Impact
Section titled “ADA Impact”| ADA Effect | Mechanism | PK Consequence |
|---|---|---|
| Neutralization | Blocks target binding | ↓ efficacy, ↓ PD effect |
| Clearance enhancement | ADA-drug complex formation | ↑ CL, ↓ t₁/₂ |
| Altered distribution | Tissue deposition | ↑ Vd |
| Anaphylaxis | Immune complex formation | Safety concern |
Modeling Approach
Section titled “Modeling Approach”Time-varying ADA model:
- ADA formation follows immune response kinetics
- ADA binds drug, forming immune complexes
- Complexes cleared faster than free drug
- Effect on PK increases with ADA titer
8. Simulation Approaches
Section titled “8. Simulation Approaches”Clinical Trial Simulation (CTS)
Section titled “Clinical Trial Simulation (CTS)”Steps:
- Define trial design (population, doses, sampling)
- Simulate PK using PopPK model
- Apply PD model for efficacy
- Add variability and error
- Analyze simulated data as planned
- Evaluate operating characteristics
Sample Size Re-Estimation
Section titled “Sample Size Re-Estimation”- Use interim data to refine variance estimates
- Re-estimate sample size for target power
- Control Type I error
9. Model Validation
Section titled “9. Model Validation”Internal Validation
Section titled “Internal Validation”| Method | Purpose |
|---|---|
| Bootstrap | Parameter uncertainty |
| Visual predictive check (VPC) | Model fit |
| Simulation-based calibration | Model performance |
External Validation
Section titled “External Validation”| Method | Purpose |
|---|---|
| External dataset | Model transportability |
| Prospective validation | Prediction accuracy |
| Cross-validation | Model robustness |
Goodness-of-Fit Criteria
Section titled “Goodness-of-Fit Criteria”| Criterion | Acceptable |
|---|---|
| Objective function value (OFV) | Lower is better |
| AIC, BIC | Lower is better |
| Parameter precision (RSE) | <30% |
| Correlation of random effects | <0.7 |
10. Case Study: Semaglutide PopPK
Section titled “10. Case Study: Semaglutide PopPK”Model Structure
Section titled “Model Structure”- Two-compartment base model
- TMDD via QSS approximation
- SC absorption with lag time
- Allometric scaling on CL, V
Key Parameters
Section titled “Key Parameters”| Parameter | Value | Units |
|---|---|---|
| CL | 0.05 | L/h |
| Vd | 12 | L |
| k_a | 0.1 | h⁻¹ |
| t₁/₂ | 165 | h |
| K_D | 0.1 | nM |
Covariates
Section titled “Covariates”- Body weight: CL ↑, V ↑
- eGFR: No significant effect
- ADA: ↑ CL by 30–50%
References
Section titled “References”- Gibaldi, M., Perrier, D. Pharmacokinetics. 2nd ed. Marcel Dekker, 1982.
- Mould, D.R., Upton, R.N. “Basic concepts in population modeling, simulation, and model-based drug development.” CPT: Pharmacometrics & Systems Pharmacology 1 (2012): 1–13.
- Dua, P., et al. “Target-mediated drug disposition model for peptides.” Journal of Pharmacokinetics and Pharmacodynamics 42 (2015): 447–462.
- Rowland, M., Tozer, T.N. Clinical Pharmacokinetics and Pharmacodynamics: Concepts and Applications. 4th ed. Lippincott Williams & Wilkins, 2011.
Further Reading
Section titled “Further Reading”- Peptide Pharmacokinetics — PK overview
- Peptide Bioanalysis — Bioanalytical methods
- Clinical Trial Design — Trial design
- Pharmacology — PK/PD concepts