The early clinical development of Ebola virus treatments

KEYWORDS:

• Ebola virus
• Marburg virus
• filovirus
• treatment
• antibody
• passive transfer
• small molecules
• antisense

1. Introduction

Filoviruses are highly pathogenic negative-stranded RNA hemorrhagic fever viruses. First discovered in 1967, filovirus infections in humans have increased, culminating with the recent Ebola virus epidemic in 2013–2016 that infected over 28,000 people. There are three genera of filoviruses. Ebolavirus consists of five species: Ebola virus (EBOV), which was first discovered in 1976 and was responsible for the 2013–2016 West African epidemic; Sudan virus (SUDV), Tai Forest virus (TAFV), Bundibugyo virus (BDBV), and Reston virus (RESTV). Marburgvirus (Marburg, MARV, and Ravn, RAVV) is the founding member of the filoviruses and just as lethal as the ebolaviruses. The newest filovirus genera, cuevavirus, is composed of a single species (Lloviu, LLOV). The recent West African EBOV epidemic has brought the need for filovirus therapeutics to the forefront. Presented here is a brief discussion of the different categories of experimental therapeutics for filovirus infection (Figure 1), with an emphasis on those that have been used in EBOV or MARV infection in humans or are in advanced preclinical development.

Figure 1. Therapeutics for Ebola virus and other filoviruses consist of four main categories: (a) host-directed approaches, which target the coagulation cascade, cytokine responses, or other dysregulated host responses in order to mitigate pathogenesis; (b) antibody therapy, both monoclonal and polyclonal, targeting viral GP1,2 in order to inhibit viral entry; (c) small molecules, aimed at reducing viral replication or entry; and (d) antisense, which inhibits viral transcription or translation.

2. Antibodies

2.1. Polyclonal antibodies

Polyclonal antibodies have been used in experimental treatment of filovirus infections since the first MARV outbreak in 1967. These novel infections were treated aggressively, including the transfer of sera from convalescent individuals; all four patients receiving this therapy survived [1], with the overall lethality rate of the outbreak reaching ~25%. In several EBOV outbreaks, transfer of whole blood or plasma from convalescent donors were tried in a limited number of patients, with anecdotal but ultimately inconclusive results. In 2012, a seminal study demonstrated that purified IgG polyclonal antibodies isolated from immune sera of macaques protected naïve animals when given 48 h after EBOV or MARV infection [2]. In the 2013–2016 EBOV epidemic in West Africa, one study did not show a survival benefit after treatment with convalescent plasma [3]. However, the anti-EBOV antibody titer and neutralization activity of the convalescent plasma was not reported, so no definitive conclusions on the efficacy of these treatments can be drawn from this study.

2.2. Monoclonal antibodies

A number of neutralizing monoclonal antibodies protect rodents from EBOV infection, but early studies found that these regimens failed to protect nonhuman primates. For years, many posited that rodent studies were largely irrelevant when it came to translation of results to nonhuman primates or humans. However, beginning in 2012, a series of publications demonstrated partial to complete efficacy of monoclonal antibody cocktails, consisting of 1–3 antibodies against EBOV, in nonhuman primates. The ZMapp™ antibody therapy is a cocktail of three monoclonal antibodies against EBOV [4], and is protective in nonhuman primates when given 5 days after infection. Thirty-six human EBOV patients have received ZMapp™ in the PREVAIL II study, but there are not sufficient data at this point to determine if it is effective in humans.

3. Small molecules

The adenosine nucleoside analog BCX4430 is protective when given 48 h post-exposure in nonhuman primates after MARV infection [5] and is currently in phase I safety testing. BCX4430 is also effective against EBOV in mouse models, and if similar results are seen in nonhuman primate studies, it could be used in future human clinical trials.

GS-5734, an adenosine analogue prodrug, protects nonhuman primates from EBOV infection when given 72 h post-infection [6]. Importantly, metabolites of GS-5734 are found in eye, brain, and testes of nonhuman primates after injection. This is an important feature of the drug, since there is evidence of long-lived EBOV persistence in so-called immune privileged sites. Indeed, GS-5734 was given to a human patient who relapsed with EBOV disease, when the virus re-emerged after 9 months of convalescence and was subsequently found in the cerebrospinal fluid. This patient survived, but it is not known if the treatment was effective [7].

Favipiravir, also known as T-705, is an RNA polymerase inhibitor originally discovered as an anti-influenza drug. Favipiravir was found to protect mice from EBOV when given after infection, and was tested in West African EBOV patients in a non-randomized trial. Results suggested that it was not effective in EBOV patients with high viral titers, but those patients with low or intermediate viral titers could possibly benefit from favipiravir treatment, although the data were not conclusive [8].

Brincidofovir is an experimental drug being tested in clinical trials against adenovirus and herpesvirus infections, and has been shown to be effective against EBOV in vitro. Due to its known safety in early human clinical trials, it was used in combination with other treatments in a handful of EBOV patients, but it is not known if the drug is efficacious.

4. Antisense

4.1. siRNA

TKM-100802 is an anti-EBOV siRNA-based antisense therapeutic that demonstrated post-exposure efficacy against EBOV in nonhuman primates, and was used in combination with other therapies in five human EBOV patients with inconclusive results. A modified treatment (TKM-130803) with nucleotide changes to match the West African EBOV strain paired with a different lipid formulation was effective when begun 72 h after infection in nonhuman primates [9]. However, in a phase II trial TKM-130803 treatment did not protect humans from EBOV, with 9/12 patients succumbing to infection [10]. Since these patients had high viral load at time of treatment, and the lipid formulation was based on the TKM-100802 drug, additional studies should be completed before firm conclusions can be drawn.

4.2. Phosphorodiamidate morpholino oligomer (PMO)

Another approach to using antisense is the use of PMOs, which are synthetic single-stranded RNA molecules. AVI-7537 against EBOV [11] was largely protective in nonhuman primates when treatment started shortly after infection. AVI-7288 targeting MARV [12] was protective in nonhuman primates when given up to 4 days after infection, and a phase I trial demonstrated safety and favorable pharmacokinetics in humans.

5. Host-directed therapeutics

5.1. Clotting cascade

In nonhuman primates, recombinant nematode anticoagulant protein c2 or the anticoagulant recombinant activated protein C had mild-to-moderate beneficial effects after EBOV and/or MARV infection. Conversely, some patients in the 1967 MARV outbreak were treated with pro-coagulation therapies, including platelets, vitamin K, and fibrinogen [1]. In this outbreak, the lethality rate was much lower than later, larger MARV outbreaks; however, it is unclear if this was due to the myriad treatments applied to these patients. Unfortunately, the use of pro-coagulant therapies in filovirus animal models has not been published.

FX06, also known as Bß15-42, is an experimental drug targeting vascular leak syndrome. Since late-stage filoviral infection can present with shock-like symptoms, including vascular leakage, this drug has been tested in two EBOV patients [13]. However, this treatment has not been tried in animal models of filovirus infection.

5.2. Type I interferon

Type I interferons (IFN) comprise an important early antiviral response by inhibiting viral gene expression and recruiting and activating immune cells. The use of a ‘hybrid’ IFN-alpha (interferon alfacon-1) delays time-to-death in nonhuman primates but has not yet been shown to confer protection. However, type I IFN used as an adjunct therapy with other treatments have shown efficacy in nonhuman primates. Type I IFN administration has been used sporadically in human EBOV patients, but it is not known if it is effective.

6. Expert opinion

6.1. If at first you do not succeed…

One lesson from filovirus research is that failure of a given therapeutic in advanced testing should not doom that strategy. After much success in rodents with monoclonal and polyclonal antibody experiments, failure of these regimens in nonhuman primate models led many to doubt this approach as a whole. However, additional testing and modification of these approaches led to dramatically effective results in nonhuman primates, leading to human clinical trials. Early studies using inactivated whole EBOV as a vaccine failed in nonhuman primate studies after showing promise in rodent models; however, attenuated EBOV inactivated with alternate approaches produced a vaccine that completely protects nonhuman primates against infection [14].

6.2. Animal models

Several experimental therapies that work post-infection in nonhuman primates have been tested in humans, with either unclear or unsuccessful results. In human infection, the incubation period, infectious dose, lethality rate, and disease course varies over a broad range, as opposed to the more rapid and uniform disease in even outbred animal populations. There are many animal models that each offer significant advantages, with nonhuman primate models largely recapitulating human disease, and all have been used to develop the experimental therapies addressed here. Additional research is needed to understand how results in animal models translate into human studies.

6.3. Host-directed therapies

An understudied aspect of filovirus therapeutics is host-directed therapies to minimize filovirus disease while allowing the immune system to clear infection. Indeed, this was attempted in the first filovirus outbreak in 1967, by administration of pro-coagulants in MARV infections, but that strategy has not been tested since. Other experiments, such as limiting vascular leakage, could have positive consequences and should be explored.

6.4. Containing filovirus escape mutants

EBOV has spontaneously mutated in untreated patients at many target sites of experimental therapeutics, and infected nonhuman primates treated with monoclonal antibody cocktails can generate EBOV escape mutants that evade antibody therapy [15]. In addition, therapeutic small molecules may drive mutant filoviruses with widespread use. This is not a criticism of the use of anti-viral treatments, but a reminder that the toolbox for treating the next generation of filovirus outbreaks needs to be expanded.

6.5. Therapeutics for non-EBOV filoviruses

Although EBOV is the most prevalent filovirus in humans to date, there are at least five other human pathogenic filoviruses. However, most therapeutics are directed toward EBOV. Advanced research into other filoviruses, or the continued development of pan-filovirus therapeutics, would be beneficial, since it is unknown which filovirus will cause the next large-scale outbreak.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Acknowledgments

The author apologizes for the many impactful publications that could not be referenced due to space constraints.

Additional information

Funding

This manuscript was funded in part by the Defense Threat Reduction Agency (HDTRA1-15-1-0061)