The Drug Concentration Conundrum

Making progress in treating diseases with pharmacotherapeutic agents demands that we study the interaction between drugs and living organisms at the molecular level. Measuring drug concentrations in biological samples to understand the relationships between drug dose, concentration, and effect is an important part of such efforts. To find meaningful relationships that are applicable for all individuals, we really need to know the drug concentration at the site of action; unfortunately, measuring this concentration can be exceedingly complicated due to the high compartmentalization of living organisms. Furthermore, drug concentrations are usually very low and can change rapidly over time.

For decades, the method of choice has been to collect a small part of the investigated organism followed by sample analysis in the laboratory. And the biological sample of choice is blood plasma because it is in close contact with all tissues. Once the sample gets to the laboratory, analytical scientists – for whom the overall purpose of the study is generally unknown – treat the sample from a purely analytical point of view, finding the concentration of the target analyte in the sample. The sample is effectively separated into fractions and the total drug concentration is measured in the portion that exhaustively contains the analyte.

While the established approach is very reproducible and has become the gold standard in most pharmacokinetic-pharmacodynamic studies, it does not contribute much to the overall purpose of the study: finding a relationship between dose, concentration, and effect. This is because the drug in plasma is several biological membranes away from the site of action – the surroundings of the biological receptor. Plasma composition varies between individuals and contains numerous components – especially proteins – that interact with most drug molecules, affecting the freely diffusible drug concentration and making it very difficult to predict drug effects in specific individuals based on total concentrations alone.

In stark contrast, methods developed for minimally invasive in-vivo sampling and analysis – such as microdialysis, microextraction, ultrafiltration, and biosensors – directly measure the freely diffusible drug concentration (1). These methods are based on partial extraction of the target analyte, usually through a semipermeable membrane. Although these methods can also be used to analyze sample aliquots, they are rarely applied in this way as they are sensitive to changes in temperature, pH, and even dissolved gas content. Moreover, these partial extraction methods can be time consuming and difficult to calibrate, and they tend to have lower sensitivity, accuracy, and precision than methods based on measuring total concentrations in sample aliquots. Accordingly, they have not gained widespread acceptance in clinical practice and are used mostly for research purposes.

One of the main impediments to finding good correlations between drug concentration and effect at population levels is the high inter-individual variability in drug distribution between body components and target receptors. The total concentration does not compensate for this variability, while the free concentrations does. This has been demonstrated for antibiotics, antiepileptics, immunosuppressants, and even endogenous hormones, such as vitamin D metabolites and testosterone. On one hand, we have the highly accurate and reproducible methods based on measuring total concentrations in sample aliquots that are poorly correlated with therapeutic effects; and, on the other hand, we have the less accurate but better correlated methods for measuring free concentrations. The result? Progress in finding good population-level correlations between drug concentrations and effects has been slow.

One solution to avoid concentration conundrum is to determine an effect-normalized concentration based on free or total concentration and the composition of the investigated organism (2). The normalized concentration would be the total drug concentration that produces the same therapeutic effect in an organism with average chemical composition.

As the sensitivity of analytical methods continues to improve and sampling is performed with minimal interference close to the site of drug action, analytical scientists will be able to provide highly accurate data that is much more closely related to therapeutic effects. However, for this to happen, the specialists in measuring drug concentrations – analytical scientists – have to be supported in their fundamental research efforts and asked for input in interdisciplinary drug development teams.