INTRODUCTION
Pharmacokinetic/pharmacodynamic (PK/PD) studies during epidemics pose substantial logistical and safety challenges. However, the data generated can be used in drug dose optimisation1 and delineating toxicology thresholds (TT). It is possible to draw conclusions from a relatively small number of subjects. PK can alter substantially in the disease state, making information from patients invaluable. This is particularly true for the haemodynamic destruction caused by Ebola virus disease (EVD), culminating in multiorgan failure.2 For example, isolated PK measurements were used to evaluate favipiravir concentrations in EVD patients, and an embedded PK/PD study of Tekmira (TKM-130803) (Arbutus Biopharma, Warminster, Pennsylvania, USA) in EVD patients yielded sufficient information to develop an in silico model, presented below.3,4
METHODS
TKM-130803 is a specific, anti-EVD therapeutic comprised of two small, interfering RNA (siRNA) siLpol-2 and siVP35-2. During the clinical trial in Sierra Leone in 2015, patients were given an intravenous infusion of 0.3 mg/kg of TKM-130803 over 2 hours daily for up to 7 days.2,4 The trial was discontinued having reached a predefined statistical endpoint, which indicated a low probability of demonstrating overall therapeutic benefit compared to historic controls.2 Plasma concentration of siRNA was compared to survival at 14 days. PK data were fitted to two-compartment models, after which Monte Carlo-simulated PK profiles were compared to efficacy thresholds (Cmax: 0.04–0.57 ng/mL; mean concentration: 1.43 ng/mL), and TT (3,000 ng/mL).
RESULTS
siRNA was in quantitative excess of virus genomes throughout treatment: a level considered needed for efficacy, but the 95th percentile exceeded TT. The maximum area under the curve (AUC) for which the 95th percentile remained under TT was a continuous infusion of 0.15 mg/kg per day. Plasma concentration of both types of siRNA were higher in subjects who died, compared with subjects who survived (p<0.025 for both siRNA).
CONCLUSION
Subjects who died exhibited impaired drug clearance, justifying caution in dosing strategies for such patients. This analysis is the first PK model derived from patients with EVD and indicates that such studies are possible, though challenging. It has given a useful insight into the PK of the siRNA in the disease state and illustrates the value of designing PK/PD studies into future clinical trials in epidemic situations.5