Evaluation of in vitro in vivo correlations for dry powder inhaler delivery using artificial neural networks

The aim of these experiments was to investigate the use of artificial neural networks (ANNs) for generating models able to predict the relative lung bioavailability and clinical effect of salbutamol when delivered to healthy volunteers and asthmatic patients from dry powder inhalers (DPIs). ANN software was used to model in vitro, demographic and in vivo data from human subjects for four different DPI formulations containing salbutamol sulfate. In 12 volunteers, a model linking the in vitro aerodynamic characteristics of the emitted dose and volunteer body surface area with the urinary excretion of drug and its metabolite in the 24 h period after inhalation was established. In 11 mild asthmatics, a predictive model correlating in vitro data, baseline lung function, body surface area and age with post-treatment improvements in forced expiratory volume in 1 s (FEV1) was also generated. Models validated using unseen data from individual subjects receiving the different DPI formulations were shown to give predictions of in vivo performance. The squared correlation coefficients (R2) for plots comparing predicted and observed in vivo outcomes were 0.83 and 0.84 for urinary excretion and lung function data, respectively. It can therefore be concluded that ANN models have the potential to predict the in vivo performance of DPIs in individual subjects.