Improving the care of Lassa fever patients: re-evaluating the use of Ribavirin and characterising cardiovascular function

Lassa fever is a viral haemorrhagic fever, endemic to several West African countries. Case fatality rates average 30% across the literature. Ribavirin, a broad-spectrum antiviral agent, has been used to treat Lassa fever for four decades. Ribavirin is a pro-drug, metabolised to its active molecules intracellularly. Ribavirin has multiple different mechanisms of action. Ribavirin also has immunomodulatory effects. These antiviral and immunomodulatory effects appear to be cell- and dose- dependent. Ribavirin has toxicity at higher doses, notably haemolytic anaemia. In this thesis, I investigated the evidence for ribavirin in Lassa fever, and present a path to gathering insights into its efficacy, or lack thereof, and potential for harm in Lassa fever patients. I performed a reconstruction and re-analysis of previously unpublished data from a United States Army/Centre for Disease Control cohort study in Sierra Leone (FDA investigational new drug 16666 (IND 16666)), a systematic review of the clinical evidence for ribavirin, a systematic review of the pre-clinical evidence for ribavirin, modelling simulations of ribavirin concentrations at currently used doses, simulations of hypothetical dosing regimens required to achieve concentrations needed to inhibit LASV, and, finally, validation work and development of a protocol to study ribavirin’s pharmacokinetics and pharmacodynamics in Lassa fever within the limitations of working with a viral haemorrhagic fever in West Africa. Separately, Lassa fever is characterised in the literature as a severe vascular leakage syndrome in some patients, one that leads to intravascular hypovolaemia, shock, pulmonary oedema, and death. The evidence for this mostly relies on clinical descriptions. There is almost no prospective data that has systematically investigated this. I therefore developed and implemented an observational study of cardiovascular function in Lassa fever patients to characterise endothelial function, vascular leakage, shock, and respiratory compromise. Reviewing the clinical evidence for ribavirin, including IND 16666, the evidence base for ribavirin is unreliable. Reconstruction and re-analysis of the previously unpublished data raised the possibility that ribavirin may be harmful in a subset of Lassa fever patients, though this data also suffers from limitations. There are plausible mechanisms by which ribavirin could be harmful, including anaemia and untargeted immunomodulation. The pre-clinical data for ribavirin in Lassa fever showed wide variation in the in vitro IC50/90 of ribavirin against Lassa virus. Interpreting these differences was not possible because of a lack of standardisation in the experimental protocols. Animal experiments showed conflicting data on the effect of ribavirin in vivo, but variations in ribavirin’s intracellular metabolism across species and cell types makes interpretation of the conflicting animal model data difficult. However, in NHP, at doses that are less than the doses currently in use in West Africa, ribavirin caused death due to anaemia. PK modelling showed that at the doses currently in use, ribavirin regimens are unlikely to exceed the mean IC50/90 derived from the literature for any significant amount of time. Simulations revealed that cumulative doses exceeding 900mg/kg would be needed to achieve > 80% of the time above the IC50/90 across 10 days of therapy. At these doses, anaemia would be severe. The continued used of ribavirin needs re-evaluating. However, use of a placebo or supportive care only arm is not feasible in West Africa. Indirect evidence for ribavirin’s effect, lack therefore, or harm, needs to be accrued. An observational pharmacokinetic/pharmacodynamic study could help provide insights. Implementing a pharmacokinetic study in a viral haemorrhagic fever poses logistical challenges, such as sampling at repeated intervals at all hours of the day, in the context of limited laboratory capacity. The validation work showed that Flinders Technology Associates drug metabolism and pharmacokinetic B cards (FTA DMPK-B), a type of dried blood spot card for pharmacokinetic sampling, inactive Lassa virus at concentrations described in patients, making a pharmacokinetic study feasible. The protocol developed, using FTA DMPK-B cards, presents a pragmatic approach to a pharmacokinetic and pharmacodynamic study in Lassa fever, which allows for a large sample size and increased power to detect associations between ribavirin pharmacokinetic parameters and pharmacodynamic endpoints. In the observational study of cardiovascular function, endothelial dysfunction was frequent in patients with Lassa fever and associated with severe extravascular fluid accumulation at admission but not later during admission. Hypotension (shock) was not associated with worse endothelial dysfunction or extravascular fluid accumulation, nor death. Hypotension appeared to be distributive, suggesting the possibility of a viral sepsis syndrome, albeit mild. Respiratory compromise was associated with worse endothelial dysfunction, renal failure, extravascular fluid accumulation, and death. Renal failure and fluid overload were probably the cause of respiratory compromise in these patients, rather than a severe vascular leakage syndrome, though this needs further research.