Voriconazole (VCZ) is a triazole antifungal agent widely used in immunocompromised patients with suspected or proved invasive fungal infection. The achievement of therapeutic range (1–5 mg/L) is clinically important to maximise VCZ efficacy. Many factors are known to influence its pharmacokinetics characterised by a wide inter and intra-individual variability in trough concentrations (age, genotype, comedications.). VCZ is metabolised primarily by CYP2C19 producing the main metabolite, N-oxide VCZ, pharmacologically inactive but potentially responsible for the occurrence of toxicities. Polymorphisms on 2 C19 gene induce different phenotypes impacting the VCZ plasma levels. Our objectives are to evaluate the variability of voriconazole trough concentrations and identify practical issues to optimise and interpret such monitoring data in children and adolescents with oncohaematological disease.Methods
Children (<18 years old) with oncohaematological disease treated with VCZ for documented or suspected invasive fungal infection who had samples drawn for VCZ monitoring from January 2014 to December 2016 were included in the study. Demographic data (age, sex, weight, initial disease and biological parameters) were collected for each patient from medical prescription or informatic extraction. Indication of VCZ, detailed treatment and monitoring data were also collected. The statistical analyses were performed using SPSS v24.0. For some analysis, patients were divided in two group based on prescription recommendations: Group 1 (G1) included children<2 years, 2–12 years and 12–14 years<50 kg and Group 2 (G2) included adolescents>12–14 years>50 kg (instead of 40 kg) and adolescents>14 years>40 kg.Results
A total of 380 trough concentrations at steady state (C0,ss) were identified in 79 patients (46 girls and 33 boys), 126 concentrations had to be excluded (66 were not at steady state, 22 were not trough levels, 38 had incomplete medical or biological information). Median age at the initiation of VCZ was 9 (1–16.5) years. The majority of patients had acute leukaemia (60.8%) and received VCZ after allogenic hematopoietic stem cell transplantation. Median oral doses in G1 was 8.1 (2.5–13.8) mg/kg (n=38) vs 3.9 (2.4–6.0) mg/kg (n=13) in G2. Median intravenous doses in G1 was 7.8 (4.2–13.3) mg/kg (n=22) vs 3.7 (2.8–4.6) mg/kg (n=6) in G2. In the global cohort, 45.6% attain therapeutic range at first monitoring, 46.8% had C0,ss below 1 mg/L and 7.6% had C0,ss over 5 mg/L. Forty-one patients were treated with recommended doses but only 53% of them reach therapeutic range. There was no impact of age, sex, biological parameters, indication of VCZ on C0,ss values. The number of C0,ss in the therapeutic range increases with the number of monitoring per pa-tient following dosage adaptation.Conclusion
There is a wide variability in VCZ trough con-centrations and our data shows that conditions should be precisely defined. for optimal monitoring. It is necessary to identify factors which contribute to this variability to individualise treatment for each patient and to take them into consideration for establishing a standardised thera-peutic drug monitoring.