Pharmacokinetics analysis: An overview
Pharmacokinetics (PK) analysis is the study of the time course of drug concentration. It summarises the movement of drugs through the body and what the body does to a given drug over time.
The four properties of drugs relevant to PK analysis are:
- Absorption: How the drug is absorbed after administration.
- Distribution: How the body distributes the drug into different bodily compartments or tissues.
- Metabolism: How the body metabolises or degrades the drug.
- Excretion: How the body excretes or gets rid of the drug.
Metabolism and excretion are often grouped together under “elimination”.
Measuring the amounts or concentrations of the drug in blood, urine, cerebrospinal fluid, or other fluids or tissues, and the time of extraction relative to drug administration are required for analysis. Multiple samples across similar time points are often required and should be timed so as to reliably estimate both when the concentration is at the highest (Tmax) and the change in concentration over time. The analysis and timepoints chosen often depends on whether drugs are administered intra- or extra-vascularly, as this affects bioavailability.
Analytic choices for PK analysis should depend on your study objectives and design. This will be discussed with you and your team by our experienced Datapharm statisticians. These analytic methods are either non-compartmental or compartmental and will be discussed in more detail in future blogs.
Relevant patient characteristics
Some patient characteristics that may affect PK results include:
- Age
- Disease state
- Sex
- Concomitant medications
- Renal function
- Hepatic function
- Smoking status
If the sample size is sufficiently large, it may be worthwhile stratifying the analyses by one or more of these variables or including them as model parameters.
Below limit of quantification
If the drug concentration is below the limit of quantification (BLQ), the exact value for the concentration is not clear. Rules for handling such data should be pre-specified. A typical set of rules include:
- If a BLQ value occurs before the first quantifiable concentration, it should be assigned a value of zero
- If a BLQ value occurs after a quantifiable concentration, it should be treated as missing.
Differences between non-compartmental and compartmental analysis
|
Non-compartmental analysis |
Compartmental analysis |
|
Simpler |
Complex |
|
Quicker to calculate |
Longer to calculate |
|
Can be used with small sample sizes |
Requires large sample sizes |
|
No assumptions required about model or body compartments |
Assumptions about model or body compartments required |
|
Requires intensive sampling to capture change in concentration across time profile |
Less samples required per subject |
|
Can only use existing data or (if sufficient data) extrapolation |
Can create predictions or simulations |
|
Can only analyse one dose at a time |
Can analyse multiple doses simultaneously |
|
Limited scope in terms of methodology for both study design and statistical analysis |
Much larger scope for different methodologies |
|
Cannot pool studies together |
Can pool studies together |
|
Linear modelling of data only (ignoring transformations) |
Both linear or non-linear modelling feasible |
If you need assistance with any aspect of your clinical trial analysis, get in touch with us today!
