Ebola Virus Infection: Overview and Update on Prevention and Treatment

Prior to 2000, antigen detection methods [e.g., enzyme-linked immunosorbent assay (ELISA)] were the gold standard for EBOV detection in some outbreaks [62]. In the acute phase of EVD, ELISA has a relatively high sensitivity (93%), but EBOV antigen levels decline as disease progresses, rendering lower sensitivity for antigen detection 1–2 weeks following symptom onset [41, 63]. Several other antigen detections tests are currently under evaluation and may be deployed in the near future to complement RT-PCR testing [60]. ELISA testing has been largely replaced by RT-PCR, which permits more rapid detection and can now be deployed in mobile (portable) testing platforms in outbreak settings [64].

Detection of IgM antibodies against EBOV is performed by ELISA in the first week after the onset of symptoms with a peak of IgM levels occurring in the 2nd week of illness [41, 48, 62]. IgM antibodies are cleared at variable rates from 1 to 6 months after illness onset [41]. Data showed that serology can be highly specific for the EVD diagnosis but less sensitive in the intensive care unit setting. Hence, antibody testing may be less clinically useful in the diagnosis and management of critically ill EVD patients [49]. Although IgG antibodies appear soon after the IgM and may persist for years [41], a substantial number of EVD patients have died before they develop an IgG antibody response [48].

One formulation (i.e., TKM-Ebola) of small interfering RNAs (siRNAs) that target EBOV is encapsulated in lipid nanoparticles to facilitate cellular delivery. SiRNAs cause cleavage in the messenger RNAs, which subsequently prevent EBOV production of three key viral proteins. Early animal studies have demonstrated that TKM-Ebola prevents infection in animals challenged with a lethal dose of EBOV [74, 75]. TKM-Ebola was administered by intramuscular injection to two groups of macaques 30 min following receipt of a lethal dose of EBOV. One group was treated with TKM-Ebola on days 1, 3 and 5 post-exposure, and the other group was treated post-exposure every day for 6 consecutive days. The first regimen provided 66% protection, and the second gave 100% protection [74]. Although the drug was tested on quite a few patients with EVD in Europe and the US with most of them surviving the disease, but because these patients received other experimental therapies including hyperimmunoglobulin serum and proper supportive care in medically advanced facilities, clear evidence of effectiveness and safety in humans is lacking [76]. In 2014, TKM-Ebola entered phase I clinical trials to evaluate the safety and pharmacokinetics among volunteer participants. However, clinical manifestations of inflammatory mediator (cytokine) appeared in participants who were treated with TKM-Ebola [77, 78]. Given the observed adverse events, the US Food and Drug Administration (FDA) placed a partial clinical hold on the trial. Since dose modifications were introduced in the TKM-Ebola trial, the FDA has allowed continuation of the study for patients with EVD. Currently, one TKM-Ebola phase I trial is active and being undertaken in San Antonio, Texas, and another TKM-Ebola phase I trial has been terminated by Tekmira, Inc., aiming to reformulate the investigational therapeutic (Table 1). Additionally, Tekmira, Inc., started a phase II trial on TKM-Ebola in Guinea. However, on 19 June 2015, Tekmira, Inc., released a letter stating that the phase II trial closed enrollment prior to completion. Preliminary data from the incomplete phase II trial indicated no therapeutic benefit was achieved from the use of TKM-Ebola. A full report from this trial is pending [79].

Other siRNA-based agents are in development, including phosphorodiamidate morpholino oligomers [80‐82]. These agents include AVI-6002 and AVI-6003, which are composed of multiple oligomers with positively charged piperazine residues located along the oligomeric backbone. In 2011, a phase I randomized, double-blind, placebo-controlled, single-dose, dose-escalation trial to assess the safety, tolerability and pharmacokinetics of AVI-6002 in healthy adult subjects was completed (ClinicalTrials.gov Identifier: NCT01353027). In a similar trial, the same group of investigators evaluated the use of AVI-6003 against Marburg virus as a post-exposure therapy (ClinicalTrials.gov Identifier: NCT01353040). In these trials, both AVI-6002 and AVI-6003 were well tolerated by healthy human volunteers [80].

Favipirarvir is a nucleotide analog and viral RNA polymerase inhibitor with a wide range of antiviral effects against numerous negative- or positive-strand RNA viruses [83‐88]. Initially, favipiravir was developed to treat influenza viruses, and a phase III clinical trial was completed in which favipiravir was tested on several thousands of people and proven to be safe and effective [84]. Recently, favipiravir has also shown efficacy against EBOV in vitro and in vivo in a mouse model [89]. Two independent animal studies have demonstrated that treatment with favipiravir resulted in rapid viral clearance and led to survival of all animals infected with EBOV through intranasal inoculation [89, 90]. A phase II clinical trial to evaluate the efficacy of favipiravir against EBOV in 225 patients with an EVD trial has been completed in Guinea [91, 92]. The investigators of this trial are in the midst of data analysis with results forthcoming in the near future. The French drug safety agency approved compassionate use of favipiravir in patients with EVD [93], and the drug was used to treat a French nurse who contracted EBOV while a volunteer with Médecins Sans Frontières (MSF) in Liberia [94].

BCX4430 is a novel adenosine analog that inhibits viral RNA polymerase activity, which results in termination of RNA synthesis. This compound has shown promising results and confers protection to EBOV-challenged mice and monkeys, even when administered following challenge with filoviruses such as Marburg and Ravn viruses [95]. Furthermore, BCX4430 has demonstrated broad-spectrum antiviral activity against many viruses, including bunyaviruses, arenaviruses, paramyxoviruses, coronaviruses and flaviviruses [95]. From a safety perspective, it is worth noting that BCX4430 may have an acceptable side effect profile as it does not incorporate into human RNA or DNA. In vitro activity against EBOV has been shown for BCX4430, but no data in humans have been obtained to date. Currently, the timing of treatment for drugs such as BCX4430 has not been established, although early treatment in high-risk or potentially EBOV-exposed individuals may be an option [96]. Oral administration of BCX4430 may be feasible, although the pharmacokinetic data suggest that the intramuscular route may provide more favorable therapeutic levels [95].

Brincidofovir is a prodrug of cidofovir and a fairly recent oral nucleotide analog that prevents viral replication by inhibiting DNA polymerase [97]. Brincidofovir has shown broad- spectrum antiviral activity against DNA viruses such as herpes viruses and adenovirus and is currently in a phase III clinical trial against cytomegalovirus and adenovirus [98, 99]. Although the exact mechanism of action for brincidofovir in EVD is not yet well understood, brincidofovir may interfere with RNA polymerase of EBOV. The US FDA has put brincidofovir on fast-track approval for treatment of EVD based on in vitro data alone [99]. A phase II open-label multicenter study to assess the safety and efficacy of brincidofovir against EBOV in humans has been withdrawn prior recruitment (ClinicalTrials.gov Identifier: NCT02271347). As a result of the dramatic decline in the number of new cases in Liberia in the month of January 2015 where the trial was first initiated, Chimerix, Inc., decided to discontinue the trial with no further discussions [100].

There are a number of known agents and newly identified compounds that have shown anti-EBOV activity. For example, Compound 7 is a benzodiazepine derivative that also prevents EBOV entry into the host cells [101]. Preliminary analysis suggests that Compound 7 binds near a hydrophobic pocket near the EBOV GP1 and GP2 interface. Analysis of this compound suggests that it may have activity against several filoviruses including EBOV [101]. No animal or human trials have been reported to date.

NSC 62914 is a small molecule, which has an antioxidant property, and was found to inhibit many viruses, including EBOV, Lassa virus, Venezuelan equine encephalitis virus and Rift Valley fever virus [102]. This compound, a reactive oxygen species, has shown in vitro activity against EBOV as well as some evidence in the EBOV mouse model for protection against lethal EBOV infection at a treatment dose of 2 mg/kg/injection (higher doses did not improve survival in the mouse model). Further studies of this or related compounds in mouse and other animal models may be warranted to elucidate their role in the treatment of EBOV and other filovirus infections.

Of additional interest are compounds that have been proven to be effective in preventing the entry of filoviruses, including EBOV, into host cells [103]. LJ-001 binds to lipid membranes and prevents virus-cell fusion across a wide range of viruses. Additional research will be needed to understand how such compounds can be optimally formulated to maximize both the safety and pharmacologic activity in vivo. FGI-103, FGI-104, and FGI-106 are a group of broad-spectrum antiviral agents that inhibit viral replication in a dose-dependent manner among multiple and genetically distinct viruses including EBOV, bunyaviruses, dengue virus, HIV and hepatitis C virus [104]. Using a mouse model, Aman et al. found that FGI-106 yields a protection after a challenge with a lethal dose of EBOV. Aman and colleagues showed that protection was also found when FGI-106 was administered in a prophylactic fashion. In related studies, FGI-103 and FGI-104 are also small molecules that inhibit filovirus infection and are found to protect EBOV- or Marburg-infected mice, although their mechanism of actions are unclear [105, 106].

In 2011, a novel finding by Carette et al. showed that EBOV cell entry requires the endo/lysosomal Niemann-Pick C1 cholesterol transporter protein (NPC1) [107]. NPC1 is a fundamental component to facilitate EBOV glycoprotein host membrane fusion. Cells that are defective in the NPC1 are resistant to infection by EBOV. In addition, inhibition of NPC1 can also disrupt the release of EBOV from the intracellular vesicular compartment. Benzylpiperazine adamantane diamides are NPC1 inhibitors that have been found to interfere with EBOV binding to NPC1 on the host cells and subsequently inhibit EBOV entry and infection [108]. Clomiphene and toremifene, which are selective estrogen receptor modulators, have recently been discovered to have NPC1 inhibitor activity and act as potential inhibitors of EBOV [109, 110]. Some anti-arrhythmic therapeutics such as amiodarone and verapamil also have been identified as potent NPC1 inhibitors of the entry of the EBOV [109, 111]. Their effectiveness has been proven in cell culture, and some trial work on amiodarone is under preparation now in West Africa [112].

Historically, convalescent whole blood (CWB) and plasma (sometimes referred to as convalescent sera) and hyperimmune serum (HS) have been employed for diseases that may be severe or result in serious consequences and for which there is no safe and effective drug or vaccine. In 1995, passive immunotherapy or convalescent immune plasma was used clinically to treat patients with EVD in the outbreak of Kikwit, DRC [113]. In this study, eight patients were given the blood of convalescent patients where 7/8 (87.5%) of patients survived the disease. To date, however, there are no definitive human clinical trial data showing that CWB or CP are effective in reducing either the severity or duration of EVD [114]. In animal studies reported in 2007, Jahrling et al. used whole blood from non-human primates (NHPs) surviving EVOV infection to treat other animals, which challenged with lethal dose of EBOV [115]. Researchers in this study concluded that there was no beneficial effect following the administration of convalescent-phase whole blood to EBOV-infected NHPs.

In contrast, Dye and colleagues used multiple doses of concentrated polyclonal IgG antibody from NHPs that had survived filovirus infection to treat Marburg virus-infected NHPs [116]. Results of this study demonstrated very effective post-exposure IgG treatment where none of the experimental NHPs showed signs of the disease or detectable viremia. Furthermore, surviving NHPs were re-challenged with Marburg virus, and all survived the re-challenge, indicating that they were able to produce Marburg virus-specific IgM antibodies to fight virus replication. Subsequent evaluations were conducted to demonstrate Marburg virus or EBOV-infected NHPs could survive the disease even if IgG treatment was delayed 2 days post infection. Successfully, both Marburg- and EBOV-infected animals survived the disease, though one out of three infected animals showed mild disease yet fully recovered afterwards. Promising results of these studies demonstrate the effectiveness of post-exposure antibody treatments and represent an option for EVD therapies for human use.

In the current West Africa Ebola outbreak, convalescent-phase plasma administration for treatment of EVD is undergoing an open label, phase II/III nonrandomized trial among 130–200 patients in Guinea [117]. For use of CP or CWB, the WHO has provided guidance to improve the safety of product development as well as safety for patients who receive these products [118‐120]. Convalescent sera-based therapy may cause some toxicity related problems, such as transmission of undetected pathogen(s) and transfusion reactions. A recent case report from treating a Spanish nurse who had contracted EBOV during her care to a patient with EVD in Spain shed some light on the issue of using experimental therapeutics including CP [121]. The infected nurse had received convalescent sera from two survivors of EVD, high-dose favipiravir and other supportive care treatment. On the 10th day of clinical disease, the nurse developed clinical signs and symptoms suggestive of post-transfusion acute lung injury, which was managed conservatively without the need of supportive mechanical ventilation. Although purified IgG can lower these risks, lot-to-lot variation remains a potential problem. Previous experience also highlights the risk of antibody-dependent enhancement of EBOV infection [122]. To address challenges with convalescent sera-based therapies, researchers have developed highly purified polyclonal antibodies or monoclonal antibodies that specifically target neutralizing glycoprotein epitopes of the EBOV envelope [123]. The long-term prospects for use of CP also require clinical laboratory infrastructure for the collection and testing of CWB or CP from recovered Ebola patients in order to ensure administration of safe blood products in the context of an EVD outbreak.

Natural infection with EBOV induces antibodies directed against the EBOV envelope transmembrane glycoprotein, which is essential to virus attachment, fusion and entry into host cells. In the past several years, a number of studies have developed and characterized monoclonal antibodies directed against epitopes of EBOV antigens. Within the genus Ebolavirus, there are five antigenically distinct viruses that generate antibodies that may or may not cross-react with other Ebola species. In 2015, Hernandez and colleagues used a panel of mouse monoclonal antibodies to extensively study cross-reactivity, avidity and viral GP epitope binding [124]. This research team identified four monoclonal cross-reactive antibodies that bind to GPs of five Ebolavirus species. These findings suggest that monoclonal antibodies directed against the five species of Ebolavirus may be useful for both clinical diagnosis and therapy.

In earlier efforts, investigators developed a cocktail of antibodies from two groupings (MB-003 and ZMab), each containing three unique monoclonal antibodies. MB-003 was composed of three humanized chimeric monoclonal antibodies (c13C6, c6D8 and h13F6) [123], while ZMAb was composed of three other different monoclonal antibodies (m1H3, m2G4 and m4G7) [125]. By direct and specific reaction with EBOV GP, these antibodies are thought to provide passive immunity against the virus by directly reacting with the EBOV envelope glycoprotein [126‐128]. In one study, ZMAb demonstrated 100% efficacy in NHPs where four of four cynomolgus macaques survived EBOV infection when given the first dose of ZMAb 24-h after exposure followed by two additional doses 3 days apart. However, when the first dose was delayed and given at 48 h after a lethal dose of EBOV exposure, 50% of the monkeys fully recovered [129]. A potent humoral and cell-mediated immune response was mounted in all animals recovered from the lethal challenge of EBOV, and all survived monkeys fully recovered from a subsequent EBOV reintroduction [130]. Recent research showed that treatment with ZMAb can be delayed as late as 72 h post EBOV exposure when ZMAb is given in combination with adenovirus vectored human interferon-α [131]. In 2012, another group of researchers demonstrated a high protection rate against EBOV infection among rhesus macaques when MB-003 monoclonal antibodies were first administered at 48 h post-exposure followed by two additional doses [125]. In this study, MB-003 monoclonal antibodies protected four of six animals (67%) against EBOV infection.

In an effort to optimize the monoclonal antibody cocktail composition to contain the most potent epitopes for neutralizing antibodies that may effectively inhibit EBOV replication and further prolong the post-exposure treatment window, a group of researchers conducted an animal study to test different combinations of monoclonal antibodies from MB-003 and ZMAb [132]. After several animal experimental trials, investigators of this study selected ZMapp as the best monoclonal antibody formulation with the best protection effect. ZMapp was composed of one monoclonal antibody from MB-003 (c13C6) and two monoclonal antibodies from ZMAb (2G4 and 4G7). With this selected combination of monoclonal antibodies, rhesus macaques were treated with ZMapp at 5, 8 and 11 days after challenge with EBOV at a lethal dose of 50 mg/kg per dose. Although animals of this study exhibited signs and symptoms of EVD and viral load was detected at 5 days post-challenge before treatment with ZMapp was initiated, all animals survived. A follow-up at 3 weeks post EBOV-exposure demonstrated an undetectable viral load in all animals [132].

This is strong evidence that ZMapp could be a potential therapeutic modality in humans even when signs and symptoms of EVD have manifested. In July 2014, two US healthcare workers were brought from Liberia to the USA after being diagnosed with EVD and were treated with aggressive fluid replacement and electrolyte correction. Both were given ZMapp and showed a decline in the level of Ebola plasma viral load in correlation with clinical improvement. Although they improved shortly after receiving ZMapp, their clinical course cannot be exclusively attributed to the monoclonal antibodies, as other measures of care could have and probably did play a major role to the clinical improvement [133].

Clinical use of ZMapp maybe practically challenging given the limited supplies of the drug since the three monoclonal antibodies of ZMapp are currently extracted from the plant Nicotiana benthamiana [132]. A new scalable transient expression technology (magnifection) could overcome future ZMapp shortages. Using magnifection, 0.5 g of ZMapp can be extracted and purified from 1 kg N. benthamiana leaf biomass [134]. While this product holds promise, a major hurdle in its future utility is to manufacture large quantities of each monoclonal antibody from plants in a way that ensures sustained high yield of monoclonal antibodies at reasonable cost [135]. To overcome potential rate-limiting steps in large-scale production, ZMapp can be manufactured using Chinese hamster ovary (CHO) cells.

In February 2015, a partnership between Liberia and the US National Institute of Allergy and Infectious Diseases (NIAID) initiated clinical trials to evaluate the safety and efficacy of ZMapp in the treatment of EVD. The original study is designed to have two-arm comparison with each arm including 100 people; one arm will receive the standard of care and the other arm the experimental ZMapp treatment [136]. While randomized controlled trials are planned, with declining EVD case loads, a modification of the original trial designs may be required to evaluate the safety and efficacy of ZMapp.

Many potential therapeutic agents have shown some post-exposure efficacy in alleviating symptoms of EVD with direct antiviral action. Despite the high safety profile compared with other newly discovered antivirals, these agents cannot be used alone because of their low efficacy, but they can be used in combination with other treatments available including supportive care measures.

Interferon: Since its discovery in 1950, interferon has not been widely used because of the associated adverse events [123]. The potential use of interferons in the treatment of EVD is rooted in the evidence that EBOV interferes with functions of type I interferons [137‐139]. Previous pre-clinical research showed that exogenous interferon-α or interferon-β could delay the occurrence of viremia or prolong survival time, but could not save animals from death [140, 141].

Recombinant Nematode Anticoagulant Protein c2 (rNAPc2) and Recombinant Human Activated Protein c (rhAPC): EBOV infections cause coagulation diathesis as one of the important pathogenesis factors for the development of EVD [142]. The two originally licensed anticoagulant drugs available are rNAPc2 and rhAPC. These drugs have demonstrated in previous studies partial protection of NHPs from ZEBOV lethal challenge, with associated survival rates of 33% and 18%, respectively [143, 144]. In December 2014, the US FDA granted the manufacturer (ARCA Biopharma, Westminster, CO) orphan drug status for rNAPc2 as a potential post-exposure treatment for EVD [145].