ABSTRACT
BACKGROUND: Methotrexate (MTX) infusions may induce severe side-effects, and alkaline hydration along with folinic acid rescue is a common way to reduce such toxic risks. The purpose of this study was to develop an adaptive rescue strategy based upon the early detection of patients with impaired MTX elimination. METHODS AND RESULTS: In this study, we propose a simple population-based Bayesian approach for predicting MTX plasma concentration from a limited number of samples, so as to adapt both duration and dosage of the rescue agent to be used next. Ten kinetic profiles obtained after 10 courses of MTX (1.5 g/m2) in seven patients with inflammatory breast cancer were used to establish the population pharmacokinetic parameters (Cl, 8.16 L/h; t1/2, 12.7 h). This population was next involved in the Bayesian estimation of MTX individual pharmacokinetic parameters from only two blood samples (T24 and T36 h), thus allowing one to forecast the elimination of this drug by predicting MTX levels at 48 h. According to the MTX concentrations predicted during the elimination phase, folinic acid rescue was then prolonged in patients likely to be overexposed. CONCLUSION: The Bayesian estimation presented in this study was an easy and convenient method to efficiently detect patients with impaired MTX elimination in routine clinical practice. This information enabled the introduction of strategies for minimizing the risk of severe drug toxicity.