Mad, Mad Marburg
Equatorial Guinea is currently experiencing its first recorded outbreak of Marburg virus (MARV) in its history.
Situation
Marburg virus (MARV) is the one of two members of the Marburgvirus genus, which is one of two genera in the Filoviridae family known to infect humans (the other being Ebolaviruses), and it is presently causing an outbreak of hemorrhagic fevers in Equatorial Guinea in the Kié-Ntem region, which borders Cameroon and Gabon. This is the first such outbreak of MARV in Equatorial Guinea, with 16 suspected cases, 9 deaths, and 16 close contacts noted. Historically, MARV infection has a case-fatality ratio of 88% (however that value has been as “low” as 24% when the patients were treated in Europe and the US, which is global health in a nutshell for you). In July 2022, 3 cases of Marburg were reported in Ghana, all from the same household, of which 2 died; fortunately there was no further documented transmission. The Nigerian CDC (NCDC) has offered a risk assessment for the outbreak for the Nigerian people, stating that the risk is moderate. The WHO’s last update for the situation was on February 13, 2023.
MARV and its filovirus relatives are among the viruses that cause viral hemorrhagic fevers, which are devastating conditions in which the inflammatory response essentially causes a circulatory collapse and potentially bleeding. These viruses were basically regarded as curiosities in the world of virology until we saw in the 2013-2016 Ebola virus outbreak (that caused a documented 28,646 cases) that they still have potential for global spread which basically reinforces one of my core beliefs about public health: problems “over there” don’t stay “over there,” meaning we should pursue medical countermeasures for these diseases regardless. I mention this to make explicit note that it is in the best interest of higher income nations to offer aid to lower and middle income nations to address their problems, although it would be nice if concern for fellow humans regardless of where they reside were enough.
The Virus and The Disease
Our knowledge of Marburg virus disease is limited largely because it is rare, occurs only sporadically, predominantly in lower income countries in Africa, and much of the information recorded about it isn’t in English. Consequently a lot of what we know about Filoviruses like MARV is based on an amalgamation of human and animal data across several filoviruses, which is… not ideal. MARV’s viral reservoir is not known, but it is probably (though not proven) to be fruit bats, specifically Egyptian rousettes, from whom humans can become infected directly or via an intermediate host (e.g. primates, which get exposed to the virus through bat urine). It has been suggested that older juvenile bats may be particularly high risk for infections because they harbor a higher viral load. MARV was the first filovirus to be catalogued after an outbreak in 1967 caused by grivets imported from Uganda resulted in the infection of researchers in Marburg and Frankfurt en Main in Germany and Belgrade in Serbia (then Yugoslavia). Filoviruses are present predominantly within the bodily fluids (specifically: amniotic fluid, blood/plasma/serum, breast milk, feces, saliva, semen, skin swabs, tears, urine, and even sweat) and do not aerosolize except possibly in hospital or laboratory settings where samples can undergo centrifugation or patients may undergo aerosolizing procedures (which is why airborne precautions are advised in the handling of patients infected with MARV but are not necessary in the community setting). For that reason, transmission generally occurs through direct contact and contact with contaminated surfaces (fomites) within the household and hospital setting; abrasions in the skin can facilitate infection. In general, spread of filovirus from asymptomatic individuals does not occur, but there are possible exceptions, particularly relating to sexual transmission from infected males (see details below).
Once the virus enters the body, it gets taken up by cells expressing attachment factors specific to the glycoprotein of the virus or lipids in the membrane of the virus and then eventually gets taken up into cells predominantly by macropinocytosis (cell drinking) and begins to replicate- initially within antigen presenting cells like macrophages and dendritic cells. Filoviruses spread throughout the body through the blood wherein they can access virtually every organ. MARV can infect endothelial cells which line the blood vessels and can cause a loss of endothelial integrity which results in shock and multisystem organ dysfunction; however, in other filovirus infections, endothelial cell infection is also seen but occurs after the stages of disease where hemorrhagic catastrophe occurs in animals, suggesting that other mechanisms may contribute. In addition to this, MARV infection can profoundly impair the ability of the liver to synthesize clotting factors and coupled with the hyperinflammatory state it is known to induce can promote their consumption along with depletion of platelets, which may offer an explanation for the mechanisms of hemorrhage.
Illness due to MARV is typically abrupt and usually occurs in 3 phases:
Generalized phase (day 1-4): Flu-like symptoms with high fever, chills, malaise, severe headache, muscle aches, and prostration. There may also be intense gastrointestinal symptoms which include abdominal pain, nausea, severe diarrhea, vomiting, and anorexia. Towards the end of the generalized phase, there may be pharyngitis. A maculopapular (red, flat patches, and bumps) rash commonly appears as well as inflammation of the glands and low platelets and white blood cells.
Early organ phase (days 5-13): Neurological symptoms begin to occur, including encephalitis, confusion, delirium, aggression, and irritability. Patients may become short of breath and develop capillary leak syndrome. At this point, hemorrhagic manifestations of the disease may occur with bleeding from the mucosae, blood vomit, and bruising. These symptoms are not universal. The kidney, liver, and pancreas may also become affected with elevated liver enzymes being commonly seen.
Late organ or convalescent phase (13+ days): In the later stages of the disease, death occurs or patients recovery. Patients who die will typically develop multiple neurological symptoms, clotting abnormalities, multisystem organ failure, shock, and coma. Death typically occurs 8 to 16 days after the onset of symptoms. In contrast, patients who survive generally have a prolonged recovery period with peeling of the rash, partial amnesia, and secondary infections being common.
The incubation period is typically 7 to 11 days but can be as long as 21 days (hence WHO policy requires 42 days, twice the maximum incubation period after the last detected case, to declare an outbreak over); transmission does not occur during the incubation period. Asymptomatic infections with MARV have not been documented, and death typically occurs on days 6 to 16 of symptoms.
There is a complicating factor here though. Filoviruses have a nasty property of being exceptionally persistent in some cases despite being RNA viruses (in 2021, cases in the Ebola outbreak in Guinea were traced an individual who had been infected 5 years prior and seemingly recovered) within immunologically privileged sites such as the testes, which raises challenging questions about how to best manage outbreaks. In fact, in 1967, a MARV-infected individual who recovered infected his wife by sexual transmission. Indeed, orchitis (inflammation of the testicle) seems to be a MARV-specific feature not seen in Ebola, as, in 1990, laboratory-acquired MARV infections in Russia, “testicular swelling to four times the normal size was observed 52 days after hospital admission for a duration of 4–5 days.” Nonhuman primates that develop persistent infection in the testes appear to have a breakdown in blood-testis barrier function (presumably because the immune system attempts to clear the virus) which recruits immunosuppressive machinery to promote repair and prevents effective clearance of virus. Researchers do note that it is likely that given enough time, the macaques would have likely eventually cleared the virus, but the potential for sexual transmission here cannot be disregarded because it is epidemiologically relevant and has human precedent. It is not known how long MARV can persist in the body, and with other filoviruses like EBOV, viral persistence can be quite different when comparing monkeys and humans.
Vaccines and Antivirals
The WHO had an important meeting about filoviruses in 2022 to discuss how to best manage them, which included a component about vaccines and antivirals.
The need for vaccines is self-evident: an effective vaccination program could prevent outbreaks before they start or help to rapidly control them (because filoviruses like Marburg are zoonotic, eradication is unrealistic). Currently there are vaccines for Ebola virus disease (Zaire) using recombinant vesicular stomatitis vectors (rVSV), adenovirus vectors, and modified vaccinia Ankara (MVA) vectors, but none are available for MARV. Mvabea (the MVA-vectored Ebola vaccine) encodes glycoproteins from multiple filoviruses (including MARV) and thus may offer cross protection against MARV but there are no data to support its effectiveness currently. Having vaccines that are ready to be deployed in the event of an outbreak, especially in countries where endemic, would be invaluable. As it happens, there are recent data on an chimpanzee adenovirus type 3 vectored MARV vaccine which are promising but clinical trials are badly needed, which also means an infrastructure prepared to carry out those trials within countries where virus is endemic (trials of vaccines are planned for the current outbreak). Furthermore, the possibility exists that because the incubation period of MARV can be quite long, post-exposure vaccination may be possible to help control outbreaks.
One of the major challenges we face with vaccines, however, is our incomplete understanding of what factors are needed for the immune system to successfully address infection by MARV. MARV and other filoviruses also cause a viremia (virus in the blood) which classic vaccinology suggests means it should be amenable to antibodies, which data from non-human primates support, as well as data for EBOV antiviral monoclonal antibodies (although obviously EBOV is a different virus). It should be noted however that results on this point are not entirely consistent and it has been suggested that EBOV requires a combination of monoclonal antibodies to be managed effectively, which may also potentially be true for MARV. Alternatively, it seems that for antibodies to be effective, killer T cell responses are critical. Survivors of filovirus infections develop long-lived antibody responses and cellular immunity, and the nucleoprotein appears to be most conserved across the filoviruses suggesting it could have some value as a vaccine antigen for the filovirus family. On the other hand, some data have found that neutralizing antibody responses among MARV infection survivors are not necessary for recovery, and rather develop a Th1-predominant immune response with relatively limited antibody and killer T cell responses. It is therefore possible that there are multiple ways to successfully address infection by MARV and other filoviruses, but the exact properties a vaccine need have to accomplish this remain nebulous, which complicates design. Additionally, the effectiveness of any vaccine approach for filoviruses can be hampered by existing immunological dysfunction in (the lower and middle income) countries where the disease is endemic such as from protein-calorie malnutrition, which speaks to a broader need for equity.
The persistence of filoviruses is another matter of great concern given that it has been shown that apparently recovered individuals can on some occasions still spread virus and cause new infections. It is probable that effective vaccines could be helpful in this respect by arresting infections before they have an opportunity to establish themselves in immunologically privileged sites like the testes, but once there, eliminating virus may be a challenge. Consequently, there is a clear unmet need for effective antiviral therapies, and ideally those which can pass the blood-testis barrier. Treatment of Ebola virus disease with monoclonal antibody cocktails has seen some success (and there are some preclinical data suggesting that they can work in MARV when combined with remdesivir), and thus it may be worthwhile to attempt to replicate this with other members of the filovirus family, but it is likely they would not be adequate to clear virus from immunoprivileged sites given that survivors do develop antibody responses and this doesn’t prevent the formation of reservoirs (the possibility does exist however that the antibody responses in question are simply delayed before they can prevent reservoir formation).
Nonetheless, there are several monoclonal antibody products for MARV that are in various stages of development, such as MR191-N and M4 which have shown some preclinical success. It is important to bear in mind that because MARV and other filoviruses emerge sporadically, opportunities to evaluate effectiveness of vaccines and therapies in the form of a classical phase 3 trial are generally limited, which requires thoughtful regulatory workarounds such as immunobridging or invocation of the Animal Rule. However, even though nonhuman primates are reasonably accurate in recapitulating human disease due to MARV (and so would likely be reasonably accurate in recapitulating the effectiveness of therapies), they are very expensive models and much more challenging and ethically fraught than rodents from an animal husbandry perspective.
Another key factor is that the ecology of MARV and other filoviruses remains incompletely understood. While it is thought that bats are the reservoir of the virus (specifically Egyptian fruit bats), this is still not proven, and how the virus goes from bats (or whatever the reservoir is) to humans is also inadequately characterized. Insufficient understanding of how the virus ends up in humans directly impedes efforts to prevent it from spilling over into them. Having said that, there are still prudent measures that can be taken to reduce infections. For example, ensuring wild game is appropriately cooked, use of condoms during sex acts, and observing isolation behaviors when individuals fall ill could all be valuable in helping with outbreak control. Historically, one of the challenges with filovirus control has related to the burial practices of affected peoples, which raises the need for appropriate education regarding how to prevent infections while also respecting customs.
Overall, I think the theme is that neglected tropical diseases have a tendency to remind us why we shouldn’t neglect them in ways that can be profoundly costly to the people they affect and so the global community has a moral imperative to not neglect them.