Does body weight matter?

It would appear, not very much. After all drug doses are seldom adjusted for body weight. Only in situations where the therapeutic index is very low, such as for oncologicals, is the drug dose adjusted for body weight. Clearance as a pharmacokinetic parameter, is seldom denominated by body weight.

Often when discussing differences in PK between populations, you can almost hear the sigh of relief when the differences in AUC can be discounted by body weight. Suddenly it's like the differences should not matter any more.

So should body weight matter?

The answer to my mind is, yes.

Pharmacokinetically, body weight does not matter as much to clearance as it does to distribution. For this reason Vd is often denominated by body weight. But not so clearance. The general reluctance to consider body weight as a major determinant of drug effect related to an older line of reasoning where drug response is seen primarily as a function of AUC, steady state concentrations and consequently clearance, rather than Vd. But as pointed out in previous posts, Vd changes can have considerable effects on drug PK, particularly Cmax and trough concentrations. These are less considered mainly because of their relative instability compared to steady state concentrations. It is however possible that variability in drug responses may in fact be more sensitive to changes in Cmax and troughs rather than steady state concentrations. If so, Vd effects may be potentially more profound than have been previously thought.

Why does this reality bother us?

Principally because one of the immediately noticeable differences between our Asian population and Western population is the difference in body weights. Caucasian males may have an average body weight of 85 kg as compared to age matched Chinese males of 65 kg. Chinese females would have an average of about 55 kg. If one recognizes that drug response may be affected by body weight, it would make us serious reconsider if drug dosage regiments developed from studies involving healthy Caucasian males, may be easily applied to Chinese females without adjustments to the average of 35% lesser body weight.

And this is not yet even considering differences in lean body mass, especially since Chinese/Asians have much less lean body mass for a given body weight, when compared to Caucasians.

Further thoughts about distributions.....

For the most part, the volume of distribution Vd is conceptualized as little more than a proportioning factor between the administered dose of a drug and the initial plasma concentrations. While it was important in determining the C0, it had little impact on drug exposure, the AUC or steady state concentrations. These were squarely in the domain of clearance mechanisms.

We expected that while protein binding was inversely related to the Vd, the free drug concentrations would eventually equilibrate across all tissue compartments, regardless of tissue binding. What was observed in the plasma of the central compartment would represent what was happening in all tissues. Furthermore if free drug clearance remained unchanged, free concentrations would remain unchanged, regardless of protein or tissue binding.

According to these ideas, central compartment pharmacokinetics was of prime importance in understanding drug efficacy, or lack of it.

In recent years, the increasing appreciation that few molecules actually permeate across membranes with the involvement of transporter processes, have led many to question the wisdom of an approach that assumed drug molecules existed in equilibrium across membranes and consequently, tissue compartments. Depending on the expression and function of transporters at various membranes, drug concentrations can vary independently of central compartment concentrations. Hence, in some individuals, the plasma concentrations may mirror concentrations in any tissue compartments if the transporters involved are ubiquitous.

Conversely, in some individuals, the concentrations in tissues 'compartments' may be very different if specific transporters are differentially expressed, leading to widely different efficacy-toxicity profiles even though central compartment concentrations appear invariate.

Do changes in Vd have any impact on clinical drug effects?

This will depend entirely on which drug concentration measurements have the greatest impact on drug effects. And here is where our relative ignorance of this particular aspect of drug pharmacology constrains our ability to use PK as a predictor of drug effects.

If for example our understanding of what best predicts drug effects is limited to considerations of 'average' or 'steady-state' drug concentrations, then we are compelled to expect that Vd changes will have little or negligible impact on clinical drug effects. This is because Vd changes do not alter the area-under-the-curve (AUC) of a drug's pharmacokinetics (since AUC is determined primarily by clearance and bioavailability). Remember (F x Dose)/AUC = Clearance?

This line of reasoning has dominated and shaped our thinking for the last few decades as evidenced by the large number of studies looking at clearance or AUC changes as predictors of drug effects. Correspondingly, we have tended to minimize the effect of distributional changes as a predictor.

Is this line of reasoning correct?

Only to a limited extent. We have seen in the previous post, that the immediate and most obvious effect of Vd changes is on the C0. Consequently if a drug's effect is dependent on peak concentrations, Vd changes will have inversely related effects on peak concentrations of the drug profile, and by extension, any clinical effect, be it efficacy or toxicity, associated with the peak.

We see this also with multiple dose regiments.

With a multiple dose regiment, even though changes in Vd are not expected to change the AUC or steady state concentrations, you can see that the fluctuations over the dosage interval are greater if the Vd is smaller (in the above case 115L compared to 230L). With the increased fluctuations, one should note that the peak and trough concentration actually move in different directions, i.e. with a smaller Vd, the peak increases but the trough decreases, while the AUC remains unchanged.

Going back to the original question....do Vd changes have an impact on clinical drug effect? It will depend on whether drug effects relate best to steady-state concentration, the AUC, peak or trough concentrations. And this is poorly understood at the moment.

For the moment however, being over-focused on the AUC therefore limits our ability to see potential causes of variability in drug response.

Even more problematic ideas of distribution.....

The simplest idea of the volume of distribution (Vd) of a drug, is to think of it as the volume that a drug distributes into when that drug is administered into the body, but before any elimination has taken place. The simplest approach therefore is to divide the dose of the drug (intravenously administered) by the observed plasma concentration at zero time, C0, i.e. before any elimination has had a chance to occur.

Vd = Dose/C0

Not all of any drug, however, is uniformly distributed throughout the body; and what is seen in the plasma only represents a fraction of all of the drug molecules distributed throughout the body. If more is distributed outside of the plasma 'compartment', the C0 will be smaller for a given dose of drug, and the Vd will appear correspondingly higher. The converse is also true, that if less is distributed outside of the plasma compartment, the Vd will appear lower.

The original ideas of the process of distribution, was that drug molecules were mostly freely permeable entities, and found a 'distributional equilibrium' across cell and tissue membranes. Binding to proteins or other large molecules prevented their effective permeation across membranes, and therefore 'trapped' these drug molecules into various 'compartments'.

Conceptually therefore, the Vd of a drug may be seen to be 'governed' by an expression relating body volumes (presumably determined by body weight), plasma protein binding and tissue binding:

Vd = Volume of central compartment + Volume of peripheral tissue x (fu/fut),

where central compartment referred to plasma volume, tissue referred to undefined number of tissues outside of plasma, fu is unbound or free fraction of drug in plasma, and fut is the unbound fraction of drug in the tissues.

By this expression one can see that the Vd can be reasonably expected to be related directly to the extent of tissue binding, and inversely related to the extent of plasma protein binding.

The Vd of a drug can also be seen to have direct and immediate effects on the starting concentrations of any drug administered, particularly if administered intravenously. And if the starting concentrations were responsible for efficacy or toxicity, the Vd may be expected to be a major determinant of efficacy or toxicity.

Knowing the Vd also helps us estimate the starting dose of a drug if we have a target drug concentration in mind. For this reason, dosage regiments of drugs with large Vds often incorporate a loading dose regiment before settling into a lower maintenance dose regiment.

The enigmatic AUC (area-under-the-curve)

The area under the plasma concentration-time curve (AUC) is an easily measurable pharmacokinetic parameter. It is used extensively in clinical pharmacokinetic studies, but students very often have a poor idea of what to make of the AUC.

Mathematically, the AUC is obtained by integrating the mathematical function that describes the plasma concentration-time profile. Practically however, it is estimated by summing all the small trapezoids that can be constructed under the concentration-time plot, using what is well known as the "trapezoidal rule".

Since it is mathematically the sum of all the plasma concentrations over the dose interval, it is often taken to represent clinical drug 'exposure'.

Pharmacokinetically however, the AUC is used to estimate drug clearance,

Clearance = Dose/AUC ...................(1)

After an oral dose, the equation is,

Clearance = (Bioavailability x Dose)/AUC ................(2)

A corollary of the above statements is that, since AUC is assumed to represent clinical drug exposure, and since AUC is determined principally by Bioavailability and Clearance, clinical drug exposure can be assumed to be determined primarily by clearance and bioavailability.

This concept has shaped our thinking for many decades, and it has closed our minds to distribution being perhaps an equal if not more important determinant of drug effects (more about this later!).

One particular area of confusion for students is that they too readily associate the AUC with bioavailability. When asked why the AUC changes for a particular drug, their first response is often that the bioavailability has changed. This is only half right.....since the AUC is determined by both clearance and bioavailability. In a situation where there are no bioavailability issues, AUC is determined primarily by clearance. AUC is only reflective of bioavailability when the clearance remains stable.

Experimentally, bioavailability is determined by measuring the AUC under oral and intravenous administrations. The clearance of the drug, measured under intravenous administration allows calculation of the clearance, which can then be used to estimate the bioavailability from the oral experiment.