Isavuconazole Concentration in Real-World Practice: Consistency with Results from Clinical Trials

ABSTRACT

Clinical use of voriconazole, posaconazole, and itraconazole revealed the need for therapeutic drug monitoring (TDM) of plasma concentrations of these antifungal agents. This need for TDM was not evident from clinical trials. In order to establish whether this requirement also applies to isavuconazole, we examined the plasma concentrations of 283 samples from patients receiving isavuconazole in clinical practice and compared the values with those from clinical trials. The concentration distributions from real-world use and clinical trials were nearly identical (>1 μg/ml in 90% of patients). These findings suggest that routine TDM may not be necessary for isavuconazole in most instances.

TEXT

Therapeutic drug monitoring (TDM) is now frequently used to optimize the safety and efficacy of antifungal agents (1–15). The need and utility of TDM are based on several widely accepted factors, including the lack of reliable dose-concentration relationships, relatively large interpatient pharmacokinetic variability, demonstrated relationships between plasma drug concentrations and efficacy or toxicity, and availability of a validated drug assays. These criteria have been fulfilled for the three antifungal triazoles itraconazole, voriconazole, and posaconazole and for flucytosine. The importance of this management strategy for these antifungal agents has led to exploration of the need for TDM for new agents early in their clinical development and use.

Isavuconazole, a broad-spectrum triazole antifungal agent, is the most recently approved antifungal agent for treatment of systemic fungal infections (16, 17). Several pharmacokinetic features of isavuconazole identified during clinical development suggested little need for routine TDM (18–20). Specifically, isavuconazole pharmacokinetics are dose proportional over a wide dose range (40 to 600 mg), interpatient pharmacokinetic variability among patients is modest compared with that of other antifungal triazoles (coefficient of variation [CV], ∼60%), and no clear efficacy or toxicity concentration thresholds have been identified.

Pharmacokinetic factors indicative of the need for TDM are often not identified until data collected during real-world clinical use are available. In particular, most clinical-use pharmacokinetic studies detect pharmacokinetic variability that is in excess of that observed in healthy volunteers or defined patient populations from clinical trials (2, 10). The goal of this report was to explore isavuconazole concentrations from patients treated during real-world clinical use and compare the data with those from samples collected during clinical trials of drug development.

Astellas aided the development of a validated plasma-concentration assay for isavuconazole by the clinical diagnostic laboratory Viracor (Lee's Summit, MO) (18, 21). The assay utilized reversed-phase ultraperformance liquid chromatography with tandem mass spectrometry (UPLC-MS/MS) to measure isavuconazole concentrations in serum or plasma. From August 2015 to August 2016, following regulatory approval of isavuconazole, 283 plasma samples were collected at the request of clinicians as part of routine care and shipped to Viracor for assay. Findings from these samples were compared with plasma concentrations from >2,459 samples from 551 patients in three phase 3 trials with isavuconazole.

Figure 1 shows the distribution of serum and plasma concentrations in data sets from real-world clinical use and clinical trials. In general, the concentration distributions were similar for the two populations, as were the pharmacokinetic parameters and drug exposures (Fig. 2). For the current clinical-use population of 283 samples, the isavuconazole plasma concentration mean, median, and standard deviation values were 2.98, 2.6, and 1.91 μg/ml, respectively; the coefficient of variation was 64%. The clinical-trial concentration mean, median, and standard deviation values were 3.30, 3.02, and 2.18 μg/ml, respectively; the coefficient of variation was 66%. Mean concentrations in clinical use were statistically lower than those in trial patients (Mann-Whitney rank-sum test, P = 0.014).

FIG 1.

Histogram of clinical-use (A) and trial (B) isavuconazole concentrations.

FIG 2.

Box and whisker plot of clinical-use (A) and trial (B) isavuconazole concentrations.

Concentration thresholds associated with efficacy have been reasonably defined for itraconazole, voriconazole, and posaconazole (∼1 μg/ml) (3). Similarly, concentrations of voriconazole and itraconazole commonly linked to toxicity have been defined (6, 7). Limited data are available to assess concentration thresholds for isavuconazole. Recent analysis of concentration data from the aspergillosis and other mold trials did not identify a threshold linked to outcome (21). Analysis of the trial and clinical-use data sets identified ∼10% of patients with concentrations below 1 μg/ml. This observation is quite different from those in clinical-use studies describing concentration monitoring with voriconazole and posaconazole, in which as many as 60% of patient values fell below this threshold (2, 10). Even with the newer posaconazole tablet formulation, the rate of suggested subtherapeutic concentrations can approach 30% (19, 20).

There are limitations to the current data set, primarily related to the lack of patient-level data. Because of this, we were unable to assess outcome relative to infection or the impact of isavuconazole concentrations on safety profiles. In addition, we do not have information regarding dosage or timing of concentration monitoring relative to drug administration. We speculate that approved dosing and trough concentrations were utilized in most patients, given the absence of recommendations for dose escalation in the literature and the convention of trough monitoring with other triazoles. Despite these limitations, these results can reassure clinicians that exposures in clinical use are very similar to those observed in clinical trials. Additionally, 90% of patients receiving isavuconazole would be anticipated to achieve a putative therapeutic concentration (>1 μg/ml). Whether additional studies will identify a different threshold of efficacy in mold infections or a toxicodynamic relationship remains to be seen as additional clinical-use cohorts are analyzed.

In the absence of additional data and lack of a defined threshold in clinical-trial analyses, the findings in this study support the premise that routine TDM is not required for patients receiving isavuconazole. However, there are circumstances where TDM may be warranted. These include patients who are receiving isavuconazole and who are displaying therapeutic failure or unexplained hepatotoxicity, as well as those who are noncompliant, obese, receiving concomitant medications that are predicted to reduce isavuconazole concentrations, age <18 years, or suffering from moderate hepatic failure (22).