Jérôme Barré

Problems in therapeutic drug monitoring: free drug level monitoring

Historically, it has been assumed that only free drug concentration is the pharmacologically active species. This article reviews the theoretical pharmacological and pharmacokinetic justifications for monitoring free drug levels. The determinants likely to influence plasma protein binding and the free concentrations of drugs are delineated. The different methods which can be used for determining free drug level are presented. Their advantages and drawbacks as well as their reliability and suitability for routine clinical practice are discussed. Currently, antiepileptic drugs such as valproic acid, phenytoin, carbamazepine and a few antiarrhythmic drugs meet the theoretical criteria justifying free drug level monitoring. Conditions causing alteration in free concentrations of these drugs are reported. But, for all these drugs, there is a considerable lack of data establishing the correlations between therapeutic or toxic response and free concentration. Presently, our capability to interpret correctly the free drug level data is still limited. In the future, much more effort must be devoted in order to provide sufficient information on the clinical relevance of free drug concentration.

Docetaxel serum protein binding with high affinity to alpha1-acid glycoprotein

SummaryThe binding of docetaxel to human plasma proteins was studied by ultrafiltration at 37°C and pH 7.4. Docetaxel was extensively (> 98%) plasma protein bound. At clinically relevant concentrations (1–5 μg/ml), the plasma binding was concentration-independent. Lipoproteins, alpha1-acid glycoprotein and albumin were the main carriers of docetaxel in plasma, and owing to the high interindividual variability of alpha1-acid glycoprotein plasma concentration, particularly in cancer, it was concluded that alpha1-acid glycoprotein should be the main determinant of docetaxel plasma binding variability. Drugs potentially coadministered with docetaxel (cisplatin, dexamethasone, doxorubicin, etoposide, vinblastine) did not modify the plasma binding of docetaxel. In blood, docetaxel was found to be mainly located in the plasma compartment (less than 15% associated to erythrocytes).

Disease-induced variations in plasma protein levels. Implications for drug dosage regimens (Part I).

SummaryMany diseases appear to lead to a decrease of drug plasma binding due either to hypoalbuminaemia or to a modification of albumin structure. In other diseases, the binding of a drug may increase due to elevated concentrations of α1-acid glycoprotein or lipoproteins. However that may be, the free fraction of a drug may vary in different pathologies. But an increase or decrease of the drug free fraction does not automatically mean an increase or decrease of the free drug concentration. Whatever the drug, a variation in the volume of distribution more or less proportional to the variation in the plasma free fraction can be expected. With respect to the clearance, the problem is much more complex and depends on the hepatic extraction ratio of drug. If the extraction is related to the free fraction (fu) of drug, a variation in fu will lead to a variation in the total drug concentration but no variation in the free drug concentration and no change in the pharmacological effect. If the extraction of a drug is dependent on hepatic flow, a variation in fu will lead to a change in the free drug concentration (with no change in the total drug concentration) and hence changes in the pharmacological effect.The aim of this article is to review the literature concerning disease-induced variations in plasma protein levels during the past 10 years. Finally, possible implications for drug dosage regimens are discussed generally from examples studied in the literature.

Expression of the genetic variants of human alpha-1-acid glycoprotein in cancer

OBJECTIVES We have investigated the AAG and its genetic variants concentrations in plasma samples of 61 patients suffering from different types of cancers. DESIGN AND METHODS The patients were shared out in three groups, breast, lung, and ovary cancers groups. AAG concentration was measured by an immunonephelometric method and the phenotype was determined, after desialylation of plasma by analytical isoelectric focusing. Detection of AAG variants was made by immunoblotting and their proportions were determined by laser densitometry analysis. A population of 74 healthy individuals served as controls. RESULTS The plasma concentrations of AAG in the breast and lung cancer groups were 2.5 times increased, while in the ovary cancer group, the concentrations were 1.6 times increased. AAG concentrations in the cancer population ranged between 0.45 and 2.85 g/L (mean value 1.12 +/- 0.51 g/L). The proportions of the ORM1 and ORM2 variants were similar to those in the healthy population. In breast and lung cancer groups, the relative concentrations of genetic variants were increased more than 2.5 fold, whereas a 1. 6-fold increase was observed in the ovary cancer group. CONCLUSIONS These results show that AAG plasma concentrations are increased in these types of cancers and that changes in the expression of the genetic variants of AAG could also occur according to the type of cancer.

Disease-Induced Variations in Plasma Protein Levels

SummaryPart I of this article, which appeared in the previous issue of the Journal, discussed the implications of variations in plasma protein levels in a number of diseases: hepatic and renal disease, acute myocardial infarction, burns, cancer, diabetes mellitus, hyperlipidaemia and inflammatory diseases. In Part II the authors continue their review with a further range of disease states, and consider their import for drug dosages.

Diclofenac binding to albumin and lipoproteins in human serum

The binding of diclofenac to human serum albumin (HSA) and to lipoproteins was studied in vitro by equilibrium dialysis. Binding to HSA is characterized by two classes of sites with one site each (K1 = 5 X 10(5) M-1 and K2 = 0.6 X 10(5) M-1). The binding to lipoproteins was shown to be saturable with a larger number of binding sites and low association constants. The evidence of two specific binding sites on HSA was confirmed by circular dichroism data. In addition, an identification of those sites was performed by displacement of fluorescent probes. The data show that the high affinity site (K1 = 5 X 10(5) M-1) is likely to be shared by benzodiazepines while the second one (K2 = 0.6 X 10(5) M-1) is common with the warfarin site.

Binding of indomethacin to human serum albumin. Its non displacement by various agents, influence of free fatty acids and the unexpected effect of indomethacin on warfarin binding

Abstract Binding of indomethacin to HSA was studied in vitro by equilibrium dialysis. Our results show that binding to HSA is 97 per cent at therapeutic levels. Binding is characterized by several saturable numbers of binding sites (n = 8) with a moderate association constant ( K = 1.2. 10 4 M −1 ) and by another nonsaturable phenomenon ( nK = 4200 M −1 ). Furthermore, acetylsalicylic acid (ASA), salicylic acid (SA), 1 anilinonaphtalene-8-sulfonic acid (ANS), chlorophenoxyisobutyrate (CPIB), digitoxin, tryptophan and warfarin do not alter indomethacin binding. On the other hand, free fatty acids (FFA) decrease indomethacin binding in human serum albumin (HSA). Moreover indomethacin markedly decreases warfarin binding.

Effects of plasma lipid levels on blood distribution and pharmacokinetics of cyclosporin A.

The present study attempted to characterize the distribution of cyclosporin A (CsA) among the lipoprotein fractions, very-low-, intermediate-, low-, and high-density (VLDL, IDL, LDL, and HDL, respectively) in the plasma of patients awaiting heart transplantation and the influence of plasma lipid constituents on the pharmacokinetics of CsA. Major fractions of a therapeutic concentration of CsA were found in HDL and in LDL. In addition, plasma lipid concentrations (total cholesterol, triglycerides, phospholipids, VLDL-cholesterol—TC, TG, PL, VLDLc, respectively) are positively correlated with the CsA distribution within the LDL fraction, and negatively correlated with the CsA distribution within the HDL fraction. Thus, the percentage of CsA in each type of lipoproteins was shown to vary with the lipid levels among individuals. A significant negative correlation was found between apparent distribution volume at steady state (Vss) in plasma and TC, PL, and LDLc and between the area under the curve measured in blood (AUCB) for whole blood and PL.

In vitro binding and partitioning of irinotecan (CPT-11) and its metabolite, SN-38, in human blood.

The binding of CPT-11 and SN-38 to human plasma proteinswas studied by ultrafiltration at 37°C and pH 7.4. In plasma,CPT-11 was 66–60% bound in the range 100–4000ng/ml and SN-38 was 94–96% bound in the range50–200 ng/ml. At these concentrations the plasma bindingof CPT-11 was slightly saturable, but the plasma binding of SN-38was concentration-independent. Albumin was the main carrier ofCPT-11 and SN-38 in plasma. In blood, the binding of CPT-11 wasmoderate (80%), mainly to plasma proteins (47%) anderythrocytes (33%). The binding of SN-38 was high(99%) and most of SN-38 in blood was located in bloodcells (approximately 66%) The simulation of a grade 3hematotoxicity (according to National Cancer Institutes CommonToxicity Criteria grading) on the SN-38 blood distributionyielded an increase in fu (free fraction of drug in plasma) from1.05 to 2.08 and a decrease in CBl/CP from1.66 to 1.14 (both resulting from a decreased cellbinding).

Evidence for a concentration-dependent polymerization of a commercial human serum albumin

Polymerization of a commercial human serum albumin (Sigma A-1887) was investigated by two different techniques, high-performance liquid chromatography and gel electrophoresis. The chromatographic technique was based on the frontal analysis principle using a column which excludes polymers but retains monomers. The results allowed the determination of the monomer--polymer affinity constant, X = 526 +/- 100. The electrophoresis technique was performed with a polyacrylamide gel containing sodium dodecyl sulphate in order to separate the different polymer species according to their molecular weights. The two techniques gave results in good accordance and showed a concentration-dependent aggregation. The higher the human serum albumin concentration, the more the monomer proportion decreases.