No licensed vaccine or drug exists as Marburg virus outbreaks multiply across Africa

Marburg virus disease has no licensed treatment and no licensed vaccine, yet the virus has triggered a string of outbreaks across Africa in just three years, with Ethiopia confirming its first-ever Marburg outbreak on 14 November 2025. A narrative review published in the journal Discover Viruses on 9 August 2026 synthesises what is currently known about the disease’s epidemiology, virology, pathogenesis, diagnostics and the investigational countermeasures now being evaluated in clinics and laboratories. The review, led by Amit Bansal and colleagues, draws on peer-reviewed literature published up to 13 February 2026, searched in MEDLINE and Embase, and deliberately excludes preprints and unpublished data.

Marburg virus disease is a severe and frequently fatal zoonotic haemorrhagic fever caused by two genetically distinct viruses, Marburg virus and Ravn virus, both members of the Filoviridae family and the genus Orthomarburgvirus. Genomic analyses show the two viruses differ by up to 21.3 percent in nucleotide sequence, yet they cause clinically indistinguishable illness. Case fatality rates vary widely, from 24 to 88 percent, reflecting ascertainment bias, outbreak size, access to supportive care and the completeness of surveillance. Globally, more than 600 cases have been recorded since the virus was first identified in August 1967, when laboratory workers in Marburg and Frankfurt in Germany fell ill after handling African green monkeys, with an average case fatality rate of around 50 percent. The review’s authors caution that the case fatality rate is difficult to interpret as a measure of care quality: in a small outbreak where the only detected case is discovered because the patient died, the rate would be 100 percent, a statistical artifact of limited sample size, delayed detection and reporting bias. In larger outbreaks, the rate often falls as a broader spectrum of disease severity is captured.

Since that first occupational outbreak, subsequent events have occurred predominantly in sub-Saharan Africa, with sporadic imported cases reported in high-income countries. Notable outbreaks include the Democratic Republic of the Congo from 1998 to 2000, Angola in 2005, Equatorial Guinea and Tanzania in 2023, Rwanda in 2024, and Tanzania and Ethiopia in 2025. The 2004–2005 outbreak in Uíge, Angola, disproportionately affected young children, and the intensive care those children required likely catalysed nosocomial and community amplification. Many filovirus outbreaks have been characterised by truncated transmission chains with limited onward propagation, which contrasts sharply with the sustained, widespread community transmission seen during the 2013–2016 West African Ebola epidemic. That difference matters for research: the prolonged Ebola footprint was what made it possible to design and implement robust clinical trials of therapeutics and vaccines, an opportunity Marburg has rarely afforded.

At the molecular level, Marburg virus is a filamentous, enveloped, single-stranded negative-sense RNA virus whose main structural proteins are the glycoprotein, the large polymerase, VP24, VP30, VP35, VP40 and the nucleoprotein. The replication cycle begins when the glycoprotein mediates attachment to host cell receptors, with entry reported to involve macropinocytosis, although clathrin-mediated or alternative pathways may dominate depending on cell type. Once inside endosomes, the GP1 subunit is cleaved by endosomal proteases, an essential step before binding to the Niemann-Pick C1 cholesterol transporter, the principal entry receptor. The GP2 subunit then drives pH-dependent fusion between viral and endosomal membranes, releasing the nucleocapsid into the cytosol, where the genome is transcribed and replicated before new virions bud from the plasma membrane, often at filopodia, in a process driven primarily by VP40.

The virus also dismantles the early immune response. VP35 and VP40 likely suppress type I interferon production or signalling, while VP24, unlike its Ebola counterpart, does not block STAT1 nuclear import and is instead associated with modulating Keap1-dependent stress pathways. Infected dendritic cells may fail to mature properly, impairing antigen presentation and cytokine production, and early lymphocyte depletion together with pro-inflammatory cytokine release can potentially contribute to cytokine storm and septic shock. Transmission from animals to humans is suspected to occur through Egyptian fruit bats as the reservoir, with monkeys serving as spillover hosts in laboratory-associated outbreaks, and the virus spreading via the saliva, urine and faeces of infected animals. Once people are infected, they transmit the virus through body fluids including saliva, blood, semen and breast milk, and through contaminated objects such as clothes, bedding and needles. Viremia increases throughout the course of disease, leading to increased shedding, and spread is often associated with caring for the sick and with burial preparations.

Clinically, the incubation period is typically 5 to 12 days, ranging from 2 to 21 days, and begins with fever, myalgia, headache and generalised weakness. By day 3 after symptom onset, patients frequently develop profuse watery diarrhoea, abdominal pain, cramping, nausea and vomiting, and a non-pruritic maculopapular rash, most prominent on the trunk, may appear between days 2 and 7. From around day 5, haemorrhagic manifestations may develop, though not in all patients, including haematemesis, melaena, epistaxis, gingival and vaginal bleeding, and haemorrhage at venepuncture sites. Neurological involvement can include confusion, irritability, aggression, delirium, encephalitis and seizures. Marked biochemical alterations accompany the disease, including elevated liver enzymes, increased creatinine and profound lymphocyte depletion, and later stages may lead to multi-organ failure and death. Viral persistence in immune-privileged sites such as the testes and eyes can lead to orchitis and uveitis during convalescence, sexual transmission is possible months after infection, and recent longitudinal studies have identified persistent cardiometabolic and haematological alterations, such as elevated blood pressure, reticulocytosis and biochemical markers of hepatic and renal stress, even years after recovery.

Diagnosis relies on laboratory confirmation because early symptoms are non-specific. RT-PCR is considered the gold standard for detecting Marburg virus RNA, typically targeting conserved genes such as NP, L and GP, with reported analytical sensitivities as low as 5 copies per reaction or 8.6 to 16 RNA copies per assay. ELISA targeting NP, VP40 and GP detects filovirus-specific IgM, visible as early as 2 to 4 days after symptom onset, and IgG, detectable from 8 to 10 days after onset and persisting for up to 2 years. Virus isolation requires a biosafety level 4 laboratory and is not used routinely. The review notes that diagnostic modalities remain limited in accessibility and scalability, particularly in resource-constrained settings, and that no WHO-prequalified and validated Marburg-specific antigen rapid diagnostic test exists for routine clinical deployment.

With no approved vaccines or specific antiviral treatments, early intensive supportive care remains the cornerstone of management and was shown to significantly improve patient survival. Intravenous fluid resuscitation, electrolyte correction, analgesia, antipyretics, nutritional support and psychosocial care form the standard package, and follow-up of contacts for early detection, isolation, testing and rapid access to supportive care is key to outbreak management. On the therapeutic front, the monoclonal antibody MBP091, isolated from a Marburg survivor four years after recovery, binds the viral glycoprotein and prevents its interaction with the NPC1 receptor, and a Phase I trial showed it was well tolerated in healthy adult volunteers. Remdesivir, a nucleotide analogue prodrug that inhibits the viral RNA-dependent RNA polymerase, showed promise in preclinical filovirus models, but human data specific to Marburg are limited, and a randomised controlled trial in the Democratic Republic of the Congo did not demonstrate a survival benefit for remdesivir in Ebola when compared against the monoclonal antibody ZMapp. The WHO-coordinated SOLIDARITY PARTNERS trial is now assessing these therapies for filoviral disease, randomising participants to receive MBP091 or no adjunctive therapy, and a 10-day course of remdesivir or no adjunctive therapy, with a host-directed intervention arm currently inactive.

During Rwanda’s 2024 outbreak, frontline healthcare workers were offered both agents under expanded access and clinical trial protocols: remdesivir was administered intravenously to 52 patients, including all who were critically ill, and MBP091 was given to 10 patients, six under expanded access and four enrolled in the WHO-coordinated randomised trial. Early clinical observations indicated that treated patients showed signs of improvement, but the authors stress that because of non-randomisation other factors may have contributed to the differences seen between treated and untreated patients. Rwanda also marked the first time post-exposure prophylaxis was administered to a large number of contacts during a filovirus outbreak, with remdesivir given to 150 contacts; no outcome data have been reported for these individuals, and the lack of randomisation warrants caution in interpreting any effect. In preclinical work, the oral prodrug obeldesivir, given for 10 days starting one day after infection, did not prevent disease from developing but showed 80 percent protection against a lethal Marburg challenge in non-human primates, and its efficacy when given later than one day post-infection still needs study.

On the vaccine front, no candidate has yet received regulatory approval, and the WHO Technical Advisory Group has prioritised four viral-vectored candidates encoding the Marburg surface glycoprotein, two based on vesicular stomatitis virus and two on non-replicating chimpanzee adenoviruses. An adenovirus-vectored vaccine developed by the Sabin Vaccine Institute has advanced furthest, demonstrating immunogenicity and safety in human trials, and a Phase I trial in the United States reported binding antibody responses in 95 percent of 40 healthy adult participants at one month, with 70 percent retaining elevated titres at one year. Phase II trials are underway in Uganda, Kenya and the United States, and Oxford University launched a first-in-human trial of its ChAdOx1 Marburg vaccine, which was given to 1,710 frontline workers and high-risk contacts during the 2024 Rwanda outbreak under emergency use authorisation in a phase 2 trial. DNA vaccines required three doses for peak antibody titres, with only 11 percent retaining antibodies at six months, while mRNA vaccines encoding the Marburg glycoprotein have shown promising results in animal models, including complete protection at low doses under specific experimental conditions, though human trials have not yet commenced. The review’s authors conclude that given the sporadic nature of outbreaks, targeted post-exposure measures may be more feasible than widespread pre-exposure immunisation, and that strengthening global preparedness will require integrated surveillance, rapid clinical trial frameworks and equitable access to vaccines and antivirals.

Subject of Research: Marburg virus disease aetiopathogenesis, diagnostics and investigational countermeasures

Article Title: Narrative review on Marburg virus disease aetiopathogenesis and management

Article References: Bansal, A., Ekanger, C. T., Dahl, E. H., & Blomberg, B. (2026). Narrative review on Marburg virus disease aetiopathogenesis and management. Discover Viruses, 3(1), Article 17. https://doi.org/10.1007/s44370-026-00055-y

Image Credits: AI Generated

DOI: 10.1007/s44370-026-00055-y

Keywords: Marburg virus disease, Orthomarburgvirus, haemorrhagic fever, MBP091, remdesivir, obeldesivir, post-exposure prophylaxis, ChAdOx1 vaccine, SOLIDARITY PARTNERS trial, RT-PCR diagnostics, Rwanda outbreak, Ethiopia outbreak