LAUSR

Background: Tafenoquine is a newly licensed antimalarial drug for the radical cure of Plasmodium vivax malaria. The mechanism of action and the optimal dosing - a balance between haemolytic risk and radical cure efficacy - are uncertain. Methods: We pooled individual data from 1,102 patients with acute vivax malaria and 72 healthy volunteers (all with >70% of the population median whole blood glucose-6-phosphate dehydrogenase [G6PD] activity) who were studied in the pre-registration trials of tafenoquine. Plasma tafenoquine concentrations were analysed under a population pharmacokinetic model. A series of Bayesian logistic and Emax dose-response models were fitted to the primary outcome of any P. vivax recurrence within 4 months. Acute vomiting and haemolysis were secondary outcomes. The key exposure variables were: the mg/kg tafenoquine dose; pharmacokinetic summaries of parent compound exposure (AUC[0,{infty}], Cmax) and metabolism (terminal elimination half-life); and day 7 methaemoglobin concentration (%) as a measure of oxidative activity. Results Tafenoquine dose (mg/kg) was a major determinant of recurrence (odds ratio [OR]: 0.70 per mg/kg increase, 95% credible interval [CI]: 0.65-0.75). After adjustment for dose, the tafenoquine terminal elimination half-life (OR: 1.15 per day increase, 95% CI: 1.06-1.25), and the day 7 methaemoglobin concentration (OR: 0.81 per absolute percentage point increase, 95% CI: 0.65 to 0.99), but not the parent compound exposure, were also associated with recurrence. Under the Emax model we estimate that the currently recommended 300mg dose in a 60kg adult (5mg/kg) results in approximately 70% of the maximal obtainable hypnozoiticidal effect. Increasing the dose to 7.5mg/kg (i.e. 450mg) would result in 90% reduction in the risk of P. vivax recurrence. Tafenoquine was well tolerated. No patients had severe haemolysis. Acute vomiting was not dose-related. In patients with normal G6PD enzyme concentrations, tafenoquine dose was associated with minor post-treatment haemoglobin reductions on days 2 or 3 (0.02 g/dL per mg/kg increase [95% CI: 0.04 to 0.00]). Conclusions and interpretation The currently recommended 300mg adult dose for radical cure of vivax malaria has suboptimal efficacy. Increasing to 450mg is predicted to increase radical cure rates substantially. The production of oxidative metabolites is central to tafenoquine's hypnozoiticidal efficacy. Clinical trials of higher tafenoquine doses are now needed to characterise efficacy, safety and tolerability.