In addition to considerations of which algorithms to use, there are a variety of modeling processes that may be followed in fitting PK and PD models. We will consider examples of each of these approaches in this chapter. For more complex PK or PD models, the general ADVANs (i.e., ADVAN5–9 or ADVAN13) may be used. If the PK model can be described by one of the specific ADVANs (i.e., ADVAN1–4 and 10–12), and the PD model is simple enough to be expressed algebraically, a specific ADVAN may be used with the PD model expressed in the $ERROR block. Alternatively, the PD model may be expressed using PREDPP in a couple of ways. Direct linear or nonlinear regression can be performed between dependent and independent variables using $PRED. The particular approach to implementation of PD models depends on the characteristics of the data and the questions to be answered. As an introduction to PK/PD modeling with NONMEM, this chapter will focus primarily on typical PD models of continuous data.ġ0.2 Implementation of PD Models in NONMEM These considerations make PD models inherently more complex than most PK models. In addition, PD data may be continuous or discrete (e.g., binary or categorical). Frequently, PD models must consider issues of baseline response, relationship to PK, disease progression, development of tolerance, and many other factors. Disease processes and drug action are complex in nature and require significant skill to model quantitatively. Pharmacometricians must have excellent modeling skills as well. The pharmacometrician should compose the presentation of modeling results in a way that directly supports decision making by the project team. A clinician does not need to understand the mathematical details of a complex system of differential equations in order to appreciate the behavior of a model and the extent to which this model answers interesting questions about drug effects. Graphical representation of models and their implications can be of great benefit to that communication. Perhaps of equal or greater importance to the value of their contribution, the pharmacometrician must be able to communicate the findings of modeling results to the team in clear terms that inform the decision-making process. The pharmacometrician must translate these questions of interest into quantitative expressions and ensure that sufficient data are collected to allow the successful use of pharmacostatistical models to address these important questions. Crafting the target product profile encourages disciplined thought regarding the questions of interest and the data needed to address those questions. Defining the primary questions to be addressed in the development program is often enabled by preparing a target product profile early in the development process. Considerations of pharmacology, chemistry, pharmaceutics, regulatory, pharmacoeconomics, and other disciplines contribute valuable input to the team in constructing the questions of primary importance for the development of the product.
The team must understand the clinical context of the disease to be treated and of currently available drug therapy to effectively assess the potential for a new drug. Defining these and determining the questions of primary interest during drug development require significant input from a multidisciplinary team including clinicians and thought leaders in the field. PK questions are generally not primary, though they are an essential piece of the landscape over which the clinical decisions must be made.ĭefinitions of acceptable safety and adequate efficacy are specific to each drug and indication. Though these are specifically kinetic concepts of extent and rate, the clinician is essentially asking questions of safety and efficacy, which when considered quantitatively we call pharmacodynamics (PD). Primary questions of interest to the physician and pharmacist, once a particular drug has been selected for use, are how much and how often should the drug be administered to a particular patient.
However, PK considerations in the development and clinical use of drugs generally have a supportive role. Most of the discussion up to this point in our text has been in regard to pharmacokinetics (PK) alone.
0 kommentar(er)
