[Therapeutic monitoring: analytic, pharmacokinetic and clinical aspects].

This paper gives an overview of present aspects and future prospects of therapeutic drug monitoring (TDM). The main aims of TDM are to avoid therapeutic failures due to bad compliance or too low dose of a given drug, as well as adverse or toxic effects due to an excessive dose. The therapeutic drugs frequently monitored depend on the country, but are generally few. For some of these drugs or for others, only patients at risk or belonging to particular sub-populations for a given drug, need TDM. A pre-analytical management is necessary, comprising a correct information of the physician, concerning the nature of the sample to collect and the clinical data necessary to the interpretation, as well as their recording; the control of the sample routing and storing conditions. Nowadays, drug analyses are essentially performed using immunochemical techniques, rapid and easy to operate but limited to a small number of drugs, and chromatographic methods, more specific and adaptable to almost any therapeutic drug and financially and technically more and more accessible. The interpretation of analytical results is a most important part of TDM, which requires knowledge of clinical data, precise collection time, co-administered treatments, and to dispose of a previously defined therapeutic range or target concentration, adapted to the population to which the patient belongs; the limitations of the analytical technique used must also be considered. Clinical pharmacokinetics is a further step in the use of analytical results, allowing the prediction of an efficient dose and administration schedule in one step, using a limited number of blood samples and generally a Bayesian estimation algorithm, readily available through commercial software dedicated to a few drugs in different reference populations. The pharmacokinetic characteristics of different populations and the validation of bayesian estimation have also been published for a number of drugs, sometimes by pharmaceutical companies following phase I and II clinical trials, even taking into account various physiopathological co-variables, but mostly by independent researchers using smaller populations. The efficiency and cost of routine TDM are questionable when it is prescribed with no clinical information or even no indication of administration and sampling times. On the contrary, several studies reported that clinical pharmacokinetics significantly improved patient outcome and were cost-saving, particularly in terms of duration of hospitalisation. The author's opinion is that TDM, in the near future, will be mainly dedicated to drugs used to treat life-threatening diseases, such as anti-HIV, anticancer and immunosuppressive drugs, and maybe also biotechnological peptides or proteins, because of cost considerations. TDM will probably also be used preferentially in target populations, characterised by higher risk or pharmacokinetic variability. Very sensitive, specific and partly automated separative techniques, such as liquid chromatography-tandem mass spectrometry, might become more common than immunochemical methods, owing to a higher flexibility and improved sample throughout. Clinical pharmacokinetics may spread to a larger number of drugs and patients, due to larger reference populations available, taking into account a number of co-variables, computerised data collection and simplified modelisation. Therefore, TDM will mainly be performed in hospitals, with an essentially clinical role for the pharmacists or pharmacologists involved and routine use of recent and efficient technologies for the TDM laboratory technical staff.