Comparing Dried Blood Spots and Plasma Concentrations for Busulfan Therapeutic Drug Monitoring in Children.

BACKGROUND: Busulfan (Bu) is one of the conditioning regimen components for pediatric hematopoietic stem cell transplantation. Bu therapeutic drug monitoring (TDM) is essential for a successful treatment outcome and toxicity evasion. Dried blood spot (DBS) sampling is a rapid and simple method for Bu TDM, compared with conventional plasma sampling. This study evaluated the feasibility of using the DBS method for Bu TDM. The hematocrit (Hct) and conditioning day were also examined for their impact on the DBS method's performance. METHODS: Venous blood collected from 6 healthy volunteers was diluted, using their plasma into 4 samples of varying Hct values. Each sample was spiked with Bu calibrators (300, 600, and 1400 ng/mL), prepared using DBS and dried plasma spot (DPS) sampling and analyzed using a validated liquid-chromatography tandem-mass spectrometry method. Clinical blood samples (n = 153) from pediatric patients (n = 15) treated with Bu (mainly from doses 1, 2, 5, and 9) were used to prepare paired volumetric DBS and DPS samples. A Bland-Altman plot and Deming regression were used to define the agreement between the paired DBS and DPS measurements. Passing-Bablok regression analyses investigated the effects of Hct and conditioning day on the linearity between both methods. RESULTS: In vitro analyses showed good agreement between DBS and DPS measurements, with a mean difference of -5.4% and a 95% confidence interval on the limits of agreement of -15.3% to 4.6%. Clinical samples showed good correlation (Pearson correlation coefficient = 0.96; slope = 1.00) between the DBS and DPS methods. The DBS method met the clinical acceptance limits for clinical samples, with a bias <±20%. Bland-Altman plots showed good agreement, with only 5.8% of paired measurements exceeding the limits of agreement (±1.96 SD), although within its 95% confidence interval. Hct observations ranged from 21.7% to 34.7% and did not affect Bu concentrations measured from DBS in either the in vitro or in vivo studies. CONCLUSIONS: These results show that DBS is a useful method for Bu TDM, provided samples are analyzed on the collection day. DBS sampling offers advantages over traditional plasma sampling in infants and younger children because only small volumes of blood are required.

GSTA1 Genetic Variants and Conditioning Regimen: Missing Key Factors in Dosing Guidelines of Busulfan in Pediatric Hematopoietic Stem Cell Transplantation.

Busulfan (Bu) is a key component of conditioning regimens used before hematopoietic stem cell transplantation (SCT) in children. Different predictive methods have been used to calculate the first dose of Bu. To evaluate the necessity of further improvements, we retrospectively analyzed the currently available weight- and age-based guidelines to calculate the first doses in 101 children who underwent allogenic SCT in CHU Sainte-Justine, Montreal, after an intravenous Bu-containing conditioning regimen according to genetic and clinical factors. The measured areas under the curve (AUCs) were within target (900 to 1500 µM/min) in 38.7% of patients after the administration of the first dose calculated based on age and weight, as locally recommended. GSTA1 diplotypes linked to poor Bu metabolism (G3) and fludarabine-containing regimens were the only factors associated with AUC within target (OR, 4.7 [95% CI, 1.1 to 19.8, P = .04]; and OR, 9.9 [95% CI, 1.6 to 61.7, P = .01], respectively). From the 11 methods selected for dose calculation, the percentage of AUCs within the target varied between 16% and 74%. In some models G3 was associated with AUCs within the therapeutic and the toxic range, whereas rapid metabolizers (G1) were correlated with subtherapeutic AUCs when different methods were used. These associations were confirmed by clearance-prediction analysis, in which GSTA1 diplotypes consistently influenced the prediction errors of the methods. These findings suggest that these factors should be considered in Bu dose prediction in addition to the anthropometric data from patients. Furthermore, our data indicated that GSTA1 diplotypes was a factor that should be included in future population pharmacokinetic models, including similar conditioning regiments, to improve the prediction of Bu exposure after its initial dose.