It is generally accepted that only the unbound fraction of a drug is pharmacologically active. Posaconazole is an antifungal agent with a protein binding of 98 to 99%. Taking into account the degree of protein binding, plasma levels in patients, and MIC levels of susceptible strains, it can be assumed that the free concentration of posaconazole sometimes will be too low to exert the expected antifungal effect. The aim was therefore to test the activity of posaconazole in serum in comparison with that of the calculated unbound concentrations in protein-free media. Significant differences (P < 0.05) from the serum control were found at serum concentrations of posaconazole of 1.0 and 0.10 mg/liter, with calculated free concentrations corresponding to 1؋ MIC and 0.1؋ MIC, respectively, against one Candida lusitaniae strain selected for proof of principle. In RPMI 1640, the corresponding calculated unbound concentration of 0.015 mg/liter resulted in a significant effect, whereas that of 0.0015 mg/liter did not. Also, against seven additional Candida strains tested, there was an effect of the low posaconazole concentration in serum, in contrast to the results in RPMI 1640. Fluconazole, a low-gradeprotein-bound antifungal, was used for comparison at corresponding concentrations in serum and RPMI 1640. No effect was observed at the serum concentration, resulting in a calculated unbound concentration of 0.1؋ MIC. In summary, there was a substantially greater pharmacodynamic effect of posaconazole in human serum than could be predicted by the non-protein-bound serum concentration. A flux from serum protein-bound to fungal lanosterol 14␣-demethylase-bound posaconazole is suggested.The goal of antimicrobial chemotherapy is to achieve active drug concentrations at the site of infection in order to kill or inhibit the growth of a causative microorganism. Binding of antimicrobial agents to serum proteins is described as one of the major determinants for drug activity (1, 7, 13). According to the free drug hypothesis, it is generally accepted that only the unbound fraction of a drug in serum is pharmacologically active. Posaconazole, a broad-spectrum antifungal agent, has a protein binding of 98 to 99% (3, 6). In large studies of the in vitro activity of posaconazole against Candida spp. and Aspergillus spp., 90% of the strains tested had MICs of 0.5 mg/liter or less, with median MICs being 0.03 and 0.12 mg/liter, respectively (8, 17). Plasma levels in patients treated with posaconazole are mostly in the range of 0.1 to 1.0 mg/liter (20). When the degree of protein binding is taken into account, it can be assumed that in many patients the free concentration of posaconazole will be too low to exert the expected antifungal effect and to obtain a favorable outcome. Yet, posaconazole has shown excellent results in clinical trials studying treatment or prevention of invasive fungal infections (5,19,20). The reason for this discrepancy is unknown, but one hypothesis might be that the protein-bound fraction can act as a reservoir and ...

Voriconazole inhibited the fungicidal effect of sequentially administered amphotericin B in a concentration- and time-dependent manner; the clinical significance of this needs further investigation.

The aim of this study was to develop and validate a new in-vitro kinetic model for the combination of two drugs with different half-lives, and to use this model for the study of the pharmacodynamic effects of amphotericin B and voriconazole, alone or in combination, against a strain of Candida albicans. Bolus doses of voriconazole and amphotericin B were administered to a starting inoculum of C. albicans. Antifungal-containing medium was eliminated and replaced by fresh medium using a peristaltic pump, with the flow-rate adjusted to obtain the desired half-life of the drug with the shorter half-life. A computer-controlled dosing pump compensated for the agent with the longer half-life. Voriconazole and amphotericin B half-lives were set to 6 and 24 h, respectively. Pharmacokinetic parameters were close to target values when both single doses and sequential doses were simulated. Voriconazole and amphotericin B administered alone demonstrated fungistatic and fungicidal activity, respectively. Simultaneous administration resulted in fungicidal activity, whereas pre-exposure of C. albicans to voriconazole, followed by amphotericin at 8 and 32 h, resulted in fungistatic activity similar to that observed with voriconazole alone. Using this model, which allowed a combination of antifungal agents with different half-lives, it was possible to demonstrate an antagonistic effect of voriconazole on the fungicidal activity of amphotericin B. The characteristics and clinical relevance of this interaction require further investigation.

The result indicates that whole body utilisation of the unactivated protein C was low. Endothelial impairment of protein C activation does not seem to be restricted to the skin vessels only.

An antagonistic effect of voriconazole on the fungicidal activity of sequential doses of amphotericin B has previously been demonstrated in Candida albicans strains susceptible to voriconazole. Because treatment failure and the need to switch to other antifungals are expected to occur more often in infections that are caused by resistant strains, it was of interest to study whether the antagonistic effect was still seen in Candida strains with reduced susceptibility to voriconazole. With the hypothesis that antagonism will not occur in voriconazole-resistant strains, C. albicans strains with characterized mechanisms of resistance against voriconazole, as well as Candida glabrata and Candida krusei strains with differences in their degrees of susceptibility to voriconazole were exposed to voriconazole or amphotericin B alone, to both drugs simultaneously, or to voriconazole followed by amphotericin B in an in vitro kinetic model. Amphotericin B administered alone or simultaneously with voriconazole resulted in fungicidal activity. When amphotericin B was administered after voriconazole, its activity was reduced (median reduction, 61%; range, 9 to 94%). Levels of voriconazoledependent inhibition of amphotericin B activity differed significantly among the strains but were not correlated with the MIC values (correlation coefficient, ؊0.19; P ‫؍‬ 0.65). Inhibition was found in C. albicans strains with increases in CDR1 and CDR2 expression but not in the strain with an increase in MDR1 expression. In summary, decreased susceptibility to voriconazole does not abolish voriconazole-dependent inhibition of the fungicidal activity of amphotericin B in voriconazole-resistant Candida strains. The degree of interaction could not be predicted by the MIC value alone.Combination antifungal therapy may be one method to improve outcome in invasive Candida infections. Because animal model pharmacodynamic studies may have difficulties in simulating human pharmacokinetics, in vitro pharmacodynamic studies with simulated human pharmacokinetics represent an additional tool in the study of antifungal combinations in that they can yield data on drug-specific antifungal activities that will be of interest before initiating clinical trials.Recently, an in vitro kinetic model for the study of combination treatment with drugs having different elimination rates has been presented and validated (12). This model has been used to demonstrate an antagonistic effect of voriconazole on the fungicidal activity of sequential doses of amphotericin B against Candida albicans (12, 13). These studies investigated C. albicans strains fully susceptible to voriconazole. The postulated mechanism is that inhibition of ergosterol synthesis leads to decrease of ergosterol content and thus diminishes the effect of amphotericin B, which uses this sterol to mediate its inhibitory effect. Consequently, it is reasonable to hypothesize that voriconazole-induced inhibition of the fungicidal activity of amphotericin B will not be observed in voriconazole-resistant stra...