Nipah virus vaccine development: current platforms, preclinical evidence, and clinical translation

1 Introduction

Nipah virus (NiV) is a highly pathogenic, zoonotic, negative-sense single-stranded RNA virus belonging to the genus Henipavirus within the family Paramyxoviridae (Chua et al., 2000; Eaton et al., 2006). The virus was first identified in 1998–1999 during a large-scale outbreak of acute encephalitis among pig farmers and abattoir workers in Peninsular Malaysia that subsequently spread to Singapore through the transport of infected swine (Chua et al., 1999; Paton et al., 1999) and a subsequent case-control study confirming that direct pig exposure was the dominant risk factor for infection. NiV and its close relative Hendra virus (HeV) are Biosafety Level 4 (BSL-4) pathogens because of their high pathogenicity, absence of licensed countermeasures, and potential for human-to-human transmission (Parashar et al., 2000).

Since the disease’s initial emergence in 2001, Bangladesh has reported nearly annual outbreaks, predominantly in the Rajshahi, Rangpur, and Khulna divisions during the winter months (Luby et al., 2009). A landmark epidemiological study spanning 14 years of NiV surveillance in Bangladesh demonstrated that most outbreaks originated from independent bat-to-human spillover events in the span of 2001–2007, whereas secondary and tertiary person-to-person spread was documented repeatedly after 2007 (Nikolay et al., 2019). In India, NiV outbreaks have been reported in West Bengal, specifically in Siliguri in 2001 (Chadha et al., 2006) and Nadia in 2007. Apart from this, the southern state of Kerala experienced four outbreaks in rapid succession in 2018, 2019, 2021, and 2023 (Arunkumar et al., 2019). During an outbreak in Kozhikode, 16 of the 18 confirmed cases resulted in death, with human-to-human transmission identified as the primary source of infection (Thomas et al., 2019). Collectively, the recurring nature of NiV outbreaks in South Asia, the consistently high mortality, and the potential for nosocomial amplification underscore the urgent need for effective preventive and therapeutic countermeasures.

The natural reservoir of NiV is the Pteropus genus of Old-World fruit bats (flying foxes). They harbour the virus asymptomatically and shed it intermittently through urine, saliva, partially eaten fruit, and fecal matter (Yob et al., 2001; Halpin et al., 2011). Ecological and epidemiological investigations in Bangladesh have established that the predominant route of spillover to humans is the consumption of raw date palm sap (tari) contaminated by bat excretions during the sap harvesting season (a cultural practice in rural communities of the Bengal Delta region) (Luby et al., 2006; Rahman et al., 2012). In addition to direct and indirect contact with infected bats and fruits contaminated by them, human-to-human transmission through close physical contact during caregiving and in nosocomial settings has been well documented (Homaira et al., 2010; Gurley et al., 2007). The demonstrated capacity for person-to-person transmission through close contact of respiratory droplets leads to a high case fatality rate (CFR) and a longer incubation period of 4–17 days (Ang et al., 2018). This has led the World Health Organization (WHO) to classify NiV as one of the top-priority pathogens with pandemic potential under its Research and Development (R&D) Blueprint for Action to Prevent Epidemics (Anish et al., 2024).

At the molecular level, NiV entry into host cells is mediated by two surface envelope glycoproteins that represent the principal targets for vaccine and therapeutic development. The attachment glycoprotein (G) binds to the highly conserved host cell receptors ephrin-B2 and ephrin-B3, while the fusion glycoprotein (F) facilitates pH-independent membrane fusion by undergoing receptor-triggered conformational changes (Bonaparte et al., 2005; Negrete et al., 2005). The broad tissue distribution of ephrin-B2, particularly on endothelial cells and neurons, accounts for the characteristic multisystem pathology of NiV infection, including systemic vasculitis, severe encephalitis, and interstitial pneumonia (Wong et al., 2002). Neutralizing antibodies targeting the receptor-binding domain of the G protein have consistently conferred protection in animal models. This established the G glycoprotein as the immunodominant target for vaccine design and passive immunotherapy (Guillaume et al., 2006; Geisbert et al., 2014).

Till now, no licensed vaccine or antiviral therapy against the NiV is available, despite being recognized as a priority pathogen by the WHO. Due to the sporadic nature of the outbreak, limited availability of BSL-4 containment for live virus studies, and increasing expenses, conventional Phase III efficacy trials are logistically infeasible (Singh et al., 2019; Satterfield et al., 2016a; Amaya and Broder, 2020). Patients solely rely on supportive care, isolation, and infection prevention and control (IPC) measures as provided by the clinical management team. During the initial Malaysian outbreak, ribavirin was the only antiviral administered to NiV patients. This drug showed an apparent reduction in mortality in an open-label, non-randomized study. These findings were not derived from controlled clinical trials and remain inconclusive, with no consistent validation in subsequent outbreaks (Chong et al., 2001). With growing challenges over the years, the countermeasure landscape has also evolved, propelled by various technologies and preparedness. Multiple vaccine platforms are now actively engaged in preclinical or clinical investigation. Subunit vaccines based on the HeV soluble G glycoprotein (HeV-sG) (Broder et al., 2012), recombinant viral-vectored vaccines using chimpanzee adenovirus (ChAdOx1 NiV) (Van Doremalen et al., 2019), and vesicular stomatitis virus (VSV) backbones (Loomis et al., 2020), and other novel approaches are investigated (Mishra et al., 2024). On the other hand, the humanized monoclonal antibody m102.4 that targets the receptor-binding domain of the Henipavirus G glycoprotein has demonstrated post-exposure prophylactic efficacy in both ferret and African green monkey models of lethal NiV infection (Bossart et al., 2009). This monoclonal antibody also completed Phase I clinical evaluation in healthy adults in Australia (Playford et al., 2020). Alongside vaccines and antibodies, nucleoside analogues, including favipiravir, showed promising efficacy against NiV and HeV replication in the Syrian hamster model (Dawes et al., 2018). Novel fusion inhibitors and host-directed antiviral approaches are being investigated with varying degrees of preclinical evidence.

Given the rapid and multifaceted evolution of this field, there is a pressing need for a comprehensive, current synthesis of the NiV vaccine and therapeutic landscape that critically appraises translational readiness. Very few updated reviews systematically presenting the full breadth of vaccine platforms and therapeutic modalities under development are reported. This review aims to address this gap. Initially, a concise overview of NiV virology, pathogenesis, and outbreak epidemiology is established, which is essential for the scientific and public health context. Later, this review provides a comprehensive update on the major vaccine candidates and platforms currently undergoing preclinical and clinical development, as well as recent advances in the application of artificial intelligence (AI) in this field.

2 Literature search

The review was conducted on peer-reviewed literature obtained from PubMed, Scopus, Web of Science, and ClinicalTrials.gov until March 2026. Literature search was done by using the following key terms in combination: “Nipah virus”, “Henipavirus”, “vaccine”, “subunit vaccine”, “viral vector vaccine”, “mRNA vaccine”, “DNA vaccine”, “virus-like particle”, “nanoparticle vaccine”, “epidemiology”, “viral pathogenesis”, “immunobiology”, and “clinical trial”. Additionally, other related literature was found by screening references from selected articles and international public health organizations such as WHO and CEPI. Preference was given to peer-reviewed primary research articles and studies as well as reports to give a complete overview of the current advancements in the development of NiV vaccines. Since this work aimed to review the available literature critically, there was no need to apply any systematic review methodology and PRISMA guidelines for selecting studies.

3 Epidemiology and transmission

The epidemiology of NiV first emerged in peninsular Malaysia and Singapore from September 1998 through June 1999 (Chua et al., 1999; Centers for Disease Control and P revention, 1999). It was observed that infected persons were likely community-farm controls to report increased numbers of sick or dying pigs (59% vs. 24%, P = 0.001) and more likely than case-farm controls to engage in activities involving direct contact with pigs (86% vs. 50%, P = 0.005). Only 8% of case patients reported no pig exposure (Parashar et al., 2000). Three individuals tested positive for IgG antibodies, as measured through ELISA, but did not show an IgM response. The next outbreak in 1999, other than Malaysia, was reported in Singapore, with 1.50% of the tested cohort showing antibodies by ELISA against HeV and NiV antigens (Paton et al., 1999). No further outbreaks have been reported from either Malaysia or Singapore since 1999. Thereafter, the epidemiological focus shifted to South Asia. In 2001, NiV outbreaks were identified in both India and Bangladesh. Since then, Bangladesh has experienced outbreaks almost annually, whereas India has reported sporadic outbreaks across several states, most recently in 2026 (Figure 1). The first South Asian outbreak was identified retrospectively in Siliguri, West Bengal, India, in 2001, involving 66 reported cases and 45 deaths (CFR ∼68%) (Chadha et al., 2006). Bangladesh subsequently reported its first confirmed NiV outbreak in Meherpur district in 2001 and has since experienced nearly annual outbreaks, predominantly in the western and northwestern divisions (Hsu et al., 2004). An additional outbreak in Nadia district, West Bengal, was reported in 2007 (Arankalle et al., 2011). Four subsequent outbreaks in 2018, 2019, 2021, and 2023 and two in 2024 were observed in Kerala, among which the 2018 outbreak had a CFR of 91% (17 deaths among 19 confirmed cases), while in the 2023 outbreak, six confirmed cases and two deaths were reported. Two independent NiV outbreaks were reported in Malappuram district, Kerala, in July and September 2024. The first outbreak involved a 14-year-old male student who presented with rapidly progressive acute encephalitis syndrome (AES), including fever, seizures, altered sensorium, myoclonus, and autonomic dysfunction. The second outbreak occurred approximately 2 months later and involved a 24-year-old male student who developed high-grade fever followed by vomiting and altered sensorium, culminating in fatal AES. Together, these outbreaks accounted for two laboratory-confirmed cases and two deaths, with no secondary human-to-human transmission, a pattern markedly different from the larger Kerala outbreaks of 2018 and 2023 (Sahay et al., 2025). Following the laboratory detection of the index cases through real-time RT-PCR, an ICMR-NIV mobile BSL-3 laboratory was dispatched to the outbreak site to help achieve rapid diagnosis and enhance containment efforts. Contact tracing was conducted, which resulted in the identification of 210 contacts in the July outbreak and 183 contacts in the September outbreak, all of whom were found to be negative for the presence of NiV infection. In addition, environmental investigation, including the examination of partially consumed fruits and bat excreta and the surveillance of bats, was conducted to identify any sources of spillover. Further, a network of ten and seven trained Virus Research and Diagnostic Laboratories (VRDLs) was established since November 2024 to screen NiV infection among cases of severe acute respiratory illness (SARI) negative for other common endemic viral respiratory pathogens. In addition, the Kerala government launched the ‘No Nipah Campaign’ to educate the public regarding probable sources of infection, transmission dynamics, and behavioural practices that minimize human-bat interactions (Sahay et al., 2025). There are significant implications that need to be considered for future vaccine implementation due to the repeated but regionally limited cases of NiV infections in Kerala. In consideration of the sporadic nature of NiV infections without any community transmission, it would be best to focus on targeted vaccinations rather than mass vaccination efforts. Upon licensing of the human NiV vaccine, key target populations for vaccination include healthcare workers, laboratory workers, NiV outbreak response teams, and people from regions that have experienced spillovers.

Outside South Asia, a single NiV outbreak was reported in the municipality of Senator Ninoy Aquino, Sultan Kudarat Province, Mindanao, Philippines, during March–May 2014. The outbreak was initially recognized following reports of human deaths associated with concurrent neurological disease and sudden deaths in horses. The outbreak involved 17 human cases, including 11 cases of acute encephalitis syndrome, five influenza-like illness cases, and one meningitis case. The overall case fatality rate was approximately 53%, while the case fatality rate among patients with acute encephalitis syndrome reached 82%. Epidemiological investigations indicated that transmission occurred primarily through exposure to infected horses, horse slaughtering, and consumption of contaminated horse meat, although limited human-to-human transmission was also documented among caregivers and healthcare workers. No subsequent confirmed human NiV outbreaks have been reported in the Philippines, indicating that this was an isolated zoonotic event (Ching et al., 2015).

Historically, NiV transmission has been defined through direct zoonotic outbreaks with limited human-to-human transmission through close contact with infected individuals. Initial studies have identified the natural reservoir to be fruit bats of the genus Pteropus, and transmission has occurred via intermediate hosts and consumption of foods contaminated by the bat excreta (Epstein et al., 2016). In Bangladesh, the main mode of zoonotic transmission involves ingestion of raw date palm sap, which is infected with Pteropus medius bat excrement during the winter sap harvesting season ranging from December to April (Rahman et al., 2012). Studies utilizing infrared camera technology have found that Pteropus bats regularly approach date palm sap containers during nighttime periods, thus infecting them with their urine, saliva, and partly eaten fruits (Khan et al., 2012). The above seasonal relation to date palm sap harvest explains the winter periodicity seen in NiV outbreaks in Bangladesh and has guided the formulation of preventive measures based on the local community (Satter et al., 2023). Person-to-person transmission has been observed in the majority of the NiV outbreaks in Bangladesh, as well as several outbreaks in India, via close physical contact while caring for acutely ill patients, and by sharing meals and exposure to respiratory droplets. The comprehensive 14-year epidemiological investigation by Nikolay et al. demonstrated that person-to-person transmission occurred in approximately one-third of identified Bangladeshi outbreak clusters. Physical proximity to patients during the acute febrile or respiratory illness phase represented the dominant risk factor for secondary infection (Nikolay et al., 2019). Nosocomial transmission has been particularly well documented during the 2001 Siliguri outbreak, where healthcare workers and hospital visitors were disproportionately affected. In multiple Bangladeshi outbreaks, transmission during funeral practices involving contact with deceased patients’ bodies was additionally documented (Sazzad et al., 2013) (Figure 2). Recent surveillance-based analyses report continued outbreaks in India and Bangladesh, with increasing geographic persistence and identification of new at-risk zones. Predictive modelling studies further indicate that densely populated regions with high human–animal interface may act as potential endemic hotspots for expanding the traditionally recognized epidemiological boundaries of the virus (Sun et al., 2024). One of the key developments in the existing literature base concerns the role of environmental and anthropogenic factors as the driving force behind transmission dynamics. Land-use changes, deforestation, agricultural intensification, and urbanization have increased human exposure to bats, which serve as reservoirs for these viruses, thereby enabling further spillovers (McKee et al., 2021). Such research findings have paved the way for eco-epidemiology by incorporating the effects of environmental changes on the outbreak of infectious diseases (Perfecto et al., 2023). While improvements in surveillance can be observed, gaps remain concerning the spillover process itself, especially with respect to the cases in Indian outbreaks where the exact pathway of infection cannot be found.

4 Clinical manifestations

The Malaysian outbreak was characterized primarily by encephalitis with relatively limited respiratory involvement, whereas the Bangladesh and India outbreaks have more frequently featured prominent acute respiratory distress. Infection in humans typically begins after an incubation period of approximately 4–14 days. Early illness is usually nonspecific, with fever, headache, myalgia, vomiting, sore throat, and cough, and may progress rapidly to severe disease. In many patients, the dominant syndromes are acute encephalitis and/or acute respiratory illness, with later manifestations including drowsiness, confusion, disorientation, seizures, and coma. In severe respiratory infection, the symptoms include atypical forms of pneumonia or ARDS-like disease, especially in the outbreaks recorded in Bangladesh and India. The fatality rates for human cases are commonly reported between 40% and 75% (https://www.who.int/news-room/fact-sheets/detail/nipah-virus). Studies on the pathogenesis in animal models reveal that NiV infection first occurs in the bronchiolar epithelium of the respiratory system, followed by endothelial cells, then circulates to the blood and other organs. Fatal human disease is characterized by systemic vasculitis, endothelial infection, and multiorgan (brain, lungs, kidneys, and spleen) involvement. Experimental hamster studies further showed that mononuclear leukocytes can carry the virus and contribute to dissemination without productive infection of the leukocytes themselves (Baseler et al., 2016).

Central nervous system (CNS) disease is a hallmark of severe NiV infection. Neuropathology includes multifocal necrotizing and inflammatory encephalitis, vascular injury, thrombosis, and blood–brain barrier disruption (Wong et al., 2002). Following initial infection through the respiratory epithelium or oropharyngeal mucosa, the virus undergoes local replication before disseminating hematogenously to target organs like the CNS, lungs, kidneys, and spleen. Infection and destruction of endothelial cells and consequent vasculitis are considered the hallmark pathological features of NiV disease (Wong et al., 2002; De Wit and Munster, 2015). In hamster models, it has been reported that NiV can enter the brain rapidly through the olfactory route, from where it spreads to the olfactory epithelium, olfactory bulb, and adjacent forebrain structures (Munster et al., 2012). Survivors may experience relapse or late-onset encephalitis (Tan et al., 2002) and persistent neurological sequelae, including seizures, cognitive impairment, and personality changes (Sejvar et al., 2007; Zhang et al., 2026).

5 Virology and pathogenesis

5.1 Classification and genome organization

NiV is classified within the genus Henipavirus, subfamily Orthoparamyxovirinae, family Paramyxoviridae, order Mononegavirales (Singh et al., 2019). The genus Henipavirus currently encompasses NiV, HeV, Cedar virus, Ghanaian bat henipavirus, Mojiang henipavirus, and the recently identified Langya virus. Among these, NiV and HeV remain the only henipaviruses known to cause severe and frequently fatal disease in humans. Two major genetic lineages of NiV were recognized: NiV-Malaysia (NiV-M), and NiV-Bangladesh (NiV-B) (Satterfield et al., 2016b). Later, one has been responsible for all subsequent outbreaks in South Asia. Both strains share approximately 97.7% and 91.8% nucleotide identity with the complete genomes of the Cambodia strain (CSUR381), respectively (Gaudino et al., 2020).

The NiV genome is a non-segmented, negative-sense, single-stranded RNA molecule of approximately 18.2 kilobases (kb) (Sun et al., 2018). The NiV-B genome is 18,252 nucleotides, which is exactly six nucleotides longer than the NiV-M prototype strain (Harcourt et al., 2005). The genome encodes six major structural proteins: nucleoprotein (N), phosphoprotein (P), matrix protein (M), fusion glycoprotein (F), attachment glycoprotein (G), and large protein (L). It comes among the largest known paramyxovirus genomes organized in the conserved gene order 3'-N-P-M-F-G-L-5' (Uchida et al., 2018). N is responsible for wrapping the viral genome and serving as a template for the RNA-dependent RNA polymerase during replication and transcription. P acts as a polymerase cofactor for the L protein-encoding polymerase. M, F, and G encode the structural proteins necessary for viral assembly and entry (Harcourt et al., 2000). Beyond the structural proteins, the P gene gives rise to additional gene products through co-transcriptional mRNA editing and alternative open reading frames. This process gives rise to accessory proteins V, W, and C, which are essential for suppressing the host’s innate immune response (Satterfield et al., 2016b; Yoneda et al., 2010). V protein antagonizes innate antiviral signalling by inhibiting RIG-I/MDA5 sensing and sequestering STAT2, whereas W protein translocates to the nucleus to suppress IFN-β and interferon-stimulated gene expression through blockade of TLR3 and IKKε pathways. C protein contributes to pathogenicity by regulating early proinflammatory host responses. Loss of V or C markedly attenuates virulence in animal models (Satterfield et al., 2015). The genomic organization and functional roles of major NiV proteins are summarized in Figure 3.

5.2 Viral entry and key drug/vaccine targets

Cellular entry by NiV is mediated by the coordinated and sequential action of G and F glycoproteins that constitute the principal molecular targets for both vaccine development and therapeutic intervention (Salle and h, 2025). The G glycoprotein is a type II transmembrane protein. It binds with high affinity to the host-cell surface receptor ephrin-B2 and, to a lesser extent, to ephrin-B3. These receptors are highly conserved members of the Eph receptor tyrosine kinase ligand family, expressed across a broad range of mammalian species (Bowden et al., 2008). Structural elucidation of the NiV G glycoprotein in complex with ephrin-B2 has revealed the atomic-level determinants of receptor engagement and has provided a rational basis for the design of both entry inhibitors and structure-guided immunogens targeting the receptor-binding interface (Bonaparte et al., 2005). The remarkable cross-species conservation of ephrin-B2 and ephrin-B3 explains the unusually broad host range of NiV, which is capable of productively infecting humans, pigs, horses, cats, dogs, ferrets, hamsters, and nonhuman primates (NHPs) (Hoque et al., 2023).

Upon receptor engagement by G glycoprotein, the F glycoprotein is initially synthesized as an inactive precursor (F₀) that requires endosomal cathepsin L–mediated proteolytic cleavage to generate the biologically active, disulfide-linked F₁–F₂ heterodimer (Pager et al., 2006) and undergoes a series of irreversible conformational rearrangements that drive pH-independent membrane fusion of the viral envelope with the host cell membrane (Lee and Ataman, 2011). Recent structural characterization of the Henipavirus F glycoprotein in its metastable prefusion conformation has yielded critical insights for the rational design of both fusion-inhibitory therapeutic peptides and stabilized prefusion immunogens for next-generation vaccine development (Byrne et al., 2023; Bossart et al., 2005). After binding to ephrin-B2/B3 receptors, NiV can disseminate to the central nervous system. Evidence from brain tissues of infected patients suggests that neuronal tropism is a key feature of NiV pathogenesis, underscoring the importance of targeting viral G and F glycoproteins and their interactions with ephrin-B2/B3 receptors for therapeutic and vaccine strategies (Ong et al., 2022).

A distinguishing feature of NiV common to many other paramyxoviruses is its potent capacity to antagonize host innate immune defences through the P gene-encoded accessory proteins V, W, and C (Lo et al., 2012; Satterfield et al., 2016c). The V protein prevents nuclear translocation of STAT1 and STAT2 by sequestering them and thereby inhibiting both type I and type II interferon (IFN) signalling pathways. W protein achieves a parallel but spatially distinct inhibitory effect through nuclear sequestration of STAT1 (Rodriguez et al., 2004). Additionally, the V protein has been shown to inhibit MDA5-mediated sensing of cytoplasmic viral double-stranded RNA, thereby suppressing the upstream induction of IFN-β (Andrejeva et al., 2004). C protein-deficient NiV mutants displayed attenuated pulmonary disease in ferret models. This shows that C protein modulates inflammatory cytokine responses and contributes to respiratory pathogenesis (Satterfield et al., 2016c; Mathieu et al., 2012). Collectively, these multifaceted immune evasion strategies enable NiV to establish productive systemic infection before the host can mount an effective antiviral response. This profoundly shows the importance of post-exposure interventions and prophylactic vaccination.

6 Vaccines against Nipah virus

6.1 Vaccines platform under development

NiV vaccine development has primarily focused on the two glycoproteins F and G as the main targets. Several candidates, including HeV-sG-V, PHV02, and mRNA-1215, along with the monoclonal antibody m102.4, are currently undergoing Phase I clinical evaluation. Currently, no human NiV vaccine has received approval despite showing a pandemic trend in recent years. However, there are four NiV vaccine candidates that are in Phase I clinical trials for healthy adults​​. Few vaccine candidates have been shown to provide complete protection in preclinical trials in cats, African green monkeys, ferrets, hamsters, and porcine models. Multiple vaccine platforms currently under investigation for NiV, including recombinant viral vectors, subunit vaccines, and virus-like particles, are reviewed in detail (Figure 4). The major NiV vaccine platforms currently under development or clinical evaluation are compared in Table 1, highlighting their principal antigens, strongest preclinical or clinical evidence, current stage of development, major advantages, and key limitations.

TABLE 1. General overview of NiV vaccine platforms under preclinical and clinical development, including target antigens, current development status, principal advantages, and limitations.

6.1.1 Subunit vaccines

Subunit epitope-based vaccines are a good alternative to traditional vaccine platforms due to several advantages over conventional vaccines. By targeting specific antigenic determinants rather than whole inactivated or live-attenuated pathogens, these vaccines minimize the risk of adverse effects. This makes them suitable for individuals with a weak immune system. Apart from this, the overall speed and cost of the manufacturing process are low (Pollet et al., 2021). Although subunit vaccines against NiV have not reached clinical trials, in the case of SARS-CoV-2, subunit vaccines manufactured by Biological E Limited in India (Corbevax) and PT Bio Farma (IndoVac) in Indonesia have been successfully approved and are in use by people. In the case of IndoVac, the yeast-derived antigen is produced without the use of animal- or human-derived cells or components, which may support compliance with halal requirements and facilitate vaccine acceptance and distribution in Muslim-majority countries. This production approach may also improve acceptability among individuals who avoid animal-derived medical products for religious or ethical reasons (Hotez et al., 2023). According to research, it has been found that the soluble recombinant G glycoprotein of HeV (HeV-sG) subunit vaccine is one of the most extensively studied among all NiV/HeV vaccines (Broder et al., 2012; Broder et al., 2016). The first subunit vaccine designed for NiV consisted of soluble G glycoprotein (sG) and was inoculated subcutaneously in two adult cats. Administration of this glycoprotein resulted in the production of high levels of serum-neutralizing antibodies (titres of ∼20,000) that sustained for 2 months (Mungall et al., 2006). Due to its ability to produce cross-protective antibodies against NiV, a subunit vaccine based on HeV-sG, adjuvanted with CpG, was evaluated in a ferret model that mimics human and equine disease. Immunization with 100 μg, 20 μg, or 4 μg doses induced protective immunity, with the 100 μg and 20 μg doses conferring complete protection against a 5,000 50% tissue culture infectious dose (TCID₅₀) challenge. No viral genome was detected in tissues or body fluids, and it provided long-term protection and reduced viral replication for 14 months post-immunization (Pallister et al., 2011). In the ferret model, a subunit vaccine based on recombinant soluble HeV-sG, formulated with the CpG adjuvant, was evaluated. Immunization with 4, 20, or 100 μg doses conferred complete protection against NiV challenge at both 20 days and 434 days post-vaccination. No viral genome was detected in tissues or fluids at early challenge in vaccinated animals. The absence of increased antibody titres post-challenge indicated viral replication inhibition and no recall response (Pallister et al., 2013). Evaluation of a subunit vaccine based on adjuvanted recombinant soluble HeV-sG in pigs demonstrated that vaccination induced neutralizing antibody titres (∼320) but failed to activate a detectable cell-mediated immune response, resulting in only partial protection against HeV challenge. Furthermore, cross-neutralizing antibody levels against NiV were insufficient, and vaccinated animals were not protected against NiV infection. In contrast, pigs orally infected with NiV developed both robust humoral and cellular immune responses. Activation of IFN-γ and the CD25 marker on CD4⁺ and CD8⁺ memory T cells were detected, which protected the pigs. This study highlights that both antibody-mediated and cell-mediated immunity are essential for effective protection against henipaviruses. It indicates the limitations of subunit vaccines and the need for improved formulations to induce both humoral and cellular immunity (Pickering et al., 2016). This disparity between porcine vaccine failure and primate protection points towards fundamental differences in Henipavirus immunobiology across mammalian species. It raises the question of whether humans, like pigs, require both cellular and humoral immunity for protection. NiV infection of porcine immune cells leads to altered Th1/Th2 response patterns, skewed cytokine responses, and altered regulatory pathways in the infected immune cells, which is not the case in other species. Human immune responses are fundamentally closer to primate than porcine biology. Human HLA-MHC structure, TLR9 signalling follows primate mechanism, and critically, mimics primate pathophysiology, not porcine. While porcine NiV manifests as acute respiratory illness (peak viral replication on day 3–5), in humans it involves systemic vasculitis with hematogenous dissemination to multiple organs (day 10–14). However, studies in AGM have not systematically characterized vaccine-induced T-cell responses in protected animals, leaving unresolved whether primate protection depends on antibodies alone or requires unmeasured cellular immunity that Pickering’s pig data indicate may be necessary. Therefore, advancing HeV-sG subunit vaccines to human Phase II trials must establish absolute protective neutralizing antibody titres through comparison of Phase I human sera to AGM efficacy models and characterize T-cell responses via IFN-γ ELISPOT and flow cytometry to determine whether vaccine-induced cellular immunity contributes to protection. In striking contrast to the porcine vaccine failure, African green monkeys vaccinated with an identical HeV-sG formulation demonstrated robust protection against lethal NiV challenge. Where pigs achieved neutralizing antibody titres of ∼320 but failed to activate cellular immunity and conferred zero NiV protection, primates achieved complete protective efficacy. This hints at the feasibility of using HeV-sG for immunization and advances towards clinical trials in humans (Bossart et al., 2012). Another study demonstrated African green monkeys as a model to assess a subunit vaccine. Animals were immunized with two doses of HeV-sG formulated with alhydrogel, with or without CpG adjuvant, and subsequently challenged with a high dose of infectious virus. All vaccinated animals (n = 8) remained clinically healthy and survived. HeV G-specific Ig antibodies were undetectable before vaccination but became measurable after the prime dose and significantly increased following booster immunization (Mire et al., 2014). Recently, for human use, NiV/HeV subunit vaccine efficacy has been demonstrated using a human-compatible formulation of the recombinant HeV-sG in the AGM model. Two independent challenge studies revealed that a single dose induces a robust humoral immune response comparable to prime–boost immunization strategies, and protective immunity was achieved as early as 7 days post-immunization in vaccinated animals surviving lethal NiV challenge. This highlighted a significant potential of the HeV-sG subunit vaccine to be used in humans, especially during future emergency outbreaks, in alignment with WHO target product profiles (Geisbert et al., 2021).

The convergent evidence from ferret and primate studies demonstrates complete protection with parenteral HeV-sG subunit vaccination, supporting advancement to human Phase II trials. However, Pickering’s critical finding that porcine vaccine failure reflects inadequate cellular immune activation despite adequate humoral responses mandates that human trials comprehensively characterize both antibody and T-cell responses.

6.1.2 Viral vector vaccines

Recombinant viral vector vaccines offer various advantages over traditional vaccines due to their ability to elicit a potent cellular response in addition to an antibody response. The presence of antigens on the surface of vector particles, combined with efficient intracellular expression, enhances antigen presentation and promotes the activation of cytokines and co-stimulatory molecules. However, when replication-deficient viral vectors are used, their reduced immunostimulatory capacity requires the addition of adjuvants to induce optimal immune responses. (Travieso et al., 2022). Many studies have been conducted on recombinant NiV viral vector vaccines, using viruses such as rabies virus, VSV, adeno-associated virus, canary pox virus, chimpanzee adenovirus, cowpox virus, Newcastle disease virus, and measles virus. In the FDA- and EMA-approved rVSV-ZEBOV (Ervebo) vaccine, the native VSV glycoprotein (G) is replaced with the Ebola virus (EBOV) glycoprotein (GP) (Saphire, 2020; Jones et al., 2005), which demonstrated protective efficacy within 10 days following a single-dose administration in infected individuals during the West African Ebola epidemic, showing a suitable and safe candidate for humans (Henao-Restrepo et al., 2017; Poetsch et al., 2019). Building on this platform, VSV-EBOV has been adapted for NiV infection vaccine development by incorporating NiV glycoproteins G or F. These constructs retain EBOV GP for viral entry while expressing NiV antigens, thereby inducing neutralizing antibodies and conferring protection in both hamster and non-human primate models after a single dose (DeBuysscher et al., 2014; Prescott et al., 2015). To increase safety, replication-deficient VSV vectors were demonstrated in hamsters by deleting the glycoprotein gene, leading to single-cycle replication-deficient vectors (VSV-ΔG). Vaccinated hamsters did not show any clinical symptoms and survived post-infection after 32 days (Lo et al., 2014). In ferrets, all three groups of animals receiving a single dose of rVSV vector vaccine designed as G_Ind* rVSV-NiV_B/F-GFP, G_Ind* rVSV-NiV_B/G-GFP, and rVSV-ΔG-NiV_B/F/G-GFP, respectively, generated humoral responses against rVSV-ΔG-NiV_B antigens after 28 days post vaccination. (Mire et al., 2013). Recently, a study reported the construction of a single-cycle VSV G-complemented rVSV-vectored NiV_B G vaccine lacking GFP so that it can be used in clinical trials in the future. This was injected in AGM, which survived the NiV challenge 3 days post-vaccination and also showed the presence of NiV G-specific antibodies, natural killer cells, and cytotoxic T cells (Foster et al., 2022). Vaccine studies with the NiV G glycoprotein showed enhanced neutralizing antibody responses against both homologous (NiV_B) and heterologous (NiV_M) strain challenge in AGM compared to the vaccine expressing NiV F, which showed partial protection against heterologous challenge and complete protection against homologous challenge (De Wit et al., 2023).

The chimpanzee adenoviral vector (ChAdOx1) vaccines have emerged as a promising platform due to their enhanced immunogenicity and low reactogenicity because of the presence of a replication-deficient vector. Further, they can be manufactured on a large scale with the facility to store under standard refrigerated conditions, and potential for short-term ambient-temperature transport (Folegatti et al., 2022). They also elicit long-lasting cellular immunity, as demonstrated by the ChAdOx1 expressing conserved influenza virus antigens, nucleoprotein (NP), and matrix protein-1 (M1) vaccine given to young and old individuals (Coughlan et al., 2018). ChAdOx1 encoding the glycoprotein of NiV_B demonstrated strong protective efficacy in the Syrian hamster model. Both single-dose and prime–boost regimens conferred complete protection against lethal homologous challenge. No infectious virus was detected in oral swabs, lung, or brain tissues. A single-dose vaccination also provided complete protection against heterologous NiV_M. Additionally, partial cross-protection was observed against HeV infection (Van Doremalen et al., 2019). The efficacy of ChAdOx1 NiV (N and F protein) was also investigated in the AGM model by giving two shots of the vaccine. The first one was given 4 weeks before the challenge, and the next one (prime-boost vaccination) was given 4 or 8 weeks before the challenge. NiV G-specific antibodies were detected 14 days post-vaccination, which significantly increased after the booster dose (Van Doremalen et al., 2022). Following encouraging preclinical efficacy in African green monkeys, the ChAdOx1 NipahB vaccine progressed to human clinical development. Ongoing Phase I evaluation (NIV001) in the United Kingdom is followed by the launch of the world’s first Phase II clinical trial in Bangladesh to assess its safety and immunogenicity in healthy adults (Registry, ISRCTN, 2026).

Recombinant measles virus (MV) based vaccines are being explored as a vaccine strategy for NiV prevention. Measles virus vectors present many advantages, such as maintaining high genetic stability compared to other RNA viruses, conferring high immunogenicity by inducing both antibodies and T cells, and their replication is restricted to the cytoplasm, thereby decreasing the risk of genomic integration or reversion (Baldo et al., 2016). Studies have shown that recombinant MV expressing NiV glycoproteins (F and G) can successfully induce strong NiV-specific neutralizing antibody responses. This vaccine strategy offered total protection against lethal NiV challenge in both hamster and AGM models. No viral load post-infection was detected. (Yoneda et al., 2013). Although till now MV-based vaccines for NiV have not yet entered human clinical trials, recent reports are confirming MV-based chikungunya vaccine (MV-CHIK) is eliciting chikungunya-specific immune responses in adults that entered Phase I and II clinical trials (Schmitz et al., 2023). Another confirmed example is the measles-vectored SARS-CoV-2 vaccine candidate V591/TMV-083 that showed Phase 1/2 safety and immunogenicity results (Vanhoutte et al., 2022).

Modified vaccinia Ankara (MVA) based vaccines are considered a safe and effective approach for NiV vaccine development due to their inability to replicate in mammalian cells and reduced virulence compared to the wild-type vaccinia virus (Lv et al., 2025; Orlova et al., 2022). Old reports demonstrated that hamsters vaccinated with either VV (vaccinia virus)-NiV.G or VV-NiV.F were completely protected from a lethal infection. Animals were able to show a strong humoral response against challenged lethal NiV infections (Guillaume et al., 2004). Later, the vaccinia virus was modified to design a safe strain known as MVA that could be used in clinical vaccine research. MVA–NiV-G candidate vaccines expressing NiV glycoprotein were injected into mice to find the role of T-cell-mediated immunity in fighting against NiV. Post immunization, H2-b-restricted epitope in the NiV-G stimulated antigen-specific CD8 T cells, and an H2-IAb-restricted epitope stimulated antigen-specific CD4 T cells were identified. Interestingly, they also found that full-length NiV-G induced stronger epitope-specific T-cell responses compared to the soluble antigen NiV-sG (Kalodimou et al., 2019).

Rabies virus-based vectors are being explored as a viable strategy for the NiV vaccine, as it can replicate and express foreign protein in target cells. The rERAG333E strain of the oral rabies vaccine produced by mutation from arginine to glutamic acid at residue 333 of glycoprotein (G_333E) was reported to be safe for oral vaccination as well as immunogenic for rabies virus Evelyn-Rokitnicki-Abelseth (rERA) infected dogs (Shuai et al., 2015). Live recombinant RABV/NiV vaccine expressing NiV G proteins was injected intramuscularly in C57BL/6 mice. Approximately 2 weeks of immunization, ELISA detected high anti-NiV G antibody titres, and immunofluorescence confirmed NiV G-specific immune sera against HeV (Keshwara et al., 2019). The immunization efficacy of two recombinant vaccines, rERAG_333E/NiVG and rERAG_333E/NiVF, was demonstrated in pigs and mice through oral administration. Both the vaccines, either singularly or together, produced significant virus-neutralizing antibodies against NiV, persisting for around 11 weeks post vaccination in mice and pigs. Further, a high level of NiV-G or NiV-F-specific IgG antibody was detected throughout the overall 11-week observation period (Shuai et al., 2020). These results support the ongoing evaluation of RABV-based vaccines as an effective approach for NiV prevention and indicate their suitability for AGM and human clinical trials.

6.1.3 mRNA vaccines

The mRNA vaccines have several advantages over DNA vaccines, live or killed attenuated or subunit vaccines. Being mRNA, it can be easily transported within a cell by formulating it within carrier molecules, taken up by the cell, and also offers flexible multi-antigen design (F + G or chimeric constructs). It is non-infectious as it cannot integrate into the genetic material and therefore has no risk of genomic integration. Further, it degrades in the cytoplasm through cellular processes, facilitating modification to control its life cycle in vivo (Pardi et al., 2018). However, early mRNA vaccine development for Henipavirus revealed a critical efficacy gap with major implications for clinical translation. Before optimized NiV-G encoding mRNA vaccine constructs were established, a single-dose mRNA vaccine containing an mRNA of the glycoprotein of the HeV showed partial protection (∼70%) against NiV challenge in Syrian hamsters, primarily through the induction of neutralizing antibodies, although the associated cell-mediated immune response remained poor (Lo et al., 2020). Although this study established the feasibility of mRNA-based vaccination against henipaviruses, the observed efficacy remains below the WHO Target Product Profile (TPP), which emphasizes high protective efficacy following a single-dose regimen for rapid outbreak response. This outcome was interpreted as suboptimal immune priming rather than an inherent limitation of the mRNA platform. In particular, in their study, the vaccinated animals did not have any detectable neutralizing antibodies against the NiV before viral infection, suggesting that there was no effective protection against the virus via the humoral immune response at the time of exposure. In addition, this was the first evaluation of a lipid nanoparticle (LNP)-formulated mRNA vaccine in the Syrian hamster model, and the authors acknowledged that immunogenicity in this model had not yet been optimized. Finally, intramuscular delivery was chosen mainly for convenience and biosafety reasons within the BSL-4 containment, while previous studies demonstrated better immunogenicity with multi-site intradermal inoculation. Collectively, these findings highlighted several areas that require improvement, including immunogen design, vaccine formulation, immune priming, and delivery. This information provides insights to develop a new generation of vaccines based on optimized immunogens and LNP technology. Understanding these limitations, further development of mRNA vaccines focused on antigen engineering. A key advancement involved a chimeric mRNA vaccine that was constructed by covalently linking the stabilized pre-fusion F (Pre-F) protein of NiV to the C-terminus trimerization domain of three attachment protein monomers and injected in mice to assess its immunogenicity. The pre-F/G mRNA in the dose range of 1–10 μg elicited a strong neutralizing antibody response as well as NiV-specific Tfh and other effector T-cell responses compared to G alone (Loomis et al., 2021). Immunogenicity of nucleoside-modified mRNA vaccine encoding soluble G glycoprotein from NiV-M was demonstrated in pigs. The first dose of vaccine was able to elicit antibodies to neutralize both NiV-M and NiV-B strains, but not efficiently against the related HeV. A booster dose leads to the production of NiV G-specific T-cell cytokine (Pedrera et al., 2024). Preclinical studies using lipid nanoparticle (LNP)-encapsulated mRNA vaccines encoding NiV F, G, or a combination of both (GF) have demonstrated robust immunogenicity in both mice and Syrian hamsters. Additionally, sera from Syrian hamsters showed a high level of cross-neutralizing antibodies against both NiV-M and NiV-B strains (Sun et al., 2025). Recently, an mRNA vaccine encoding the 60-mer head domain of NiV-G, which is the main target of antibodies neutralizing NiV-G antigen, was constructed within a nanoparticle. This mRNA vaccine was able to elicit a robust antibody response in the pseudo-typed NiV system (Brandys et al., 2025). Lipid nanoparticle-based mRNA vaccine, mRNA-1215, encoding a chimeric pre-F protein linked to glycoprotein G of NiV-M, recently entered a human phase 1 clinical trial. 40 healthy adults were considered and divided into four groups consisting of 10 individuals each, in which each group individual was subjected to 10-, 25-, 50-, or 100-μg doses, respectively. It was observed that mRNA-1215 elicited neutralizing antibodies after 2 weeks in all dose groups and remained elevated for at least 1 year after vaccination (Ploquin et al., 2026). Although there is great promise with mRNA vaccines, there are some concerns with their cold chain storage requirements and questions surrounding the longevity of immunity. The potential for reactogenicity and requirement for boosters must be considered in future clinical studies of NiV vaccine.

6.1.4 Virus-like particles (VLPs)

Virus-like particles are attractive vaccine platforms because they can mimic the size, morphology, and antigenic organization of authentic virions while lacking the viral genetic material required for replication (Walpita et al., 2011). For NiV, the major structural components used in VLP vaccine designs are the matrix protein M, the fusion glycoprotein F, and the attachment glycoprotein G. Groundwork on VLPs as a vaccine was reported by a study that established recombinant systems to study NiV particle assembly and showed that M, F, and G proteins could be released in membrane-associated particles. The M protein, being important for particle assembly and budding, plays a central role in producing organized VLPs. This study concluded that co-expression of NiV structural proteins supports a more organized budding process and that the VLP system allows examination of NiV assembly outside high-level biological containment (Patch et al., 2007).

Virus-like particle vaccines for NiV remain a promising platform, with the strongest earlier evidence coming from mammalian-cell-derived NiV VLPs composed of matrix protein M, fusion glycoprotein F, and attachment glycoprotein G, which protected hamsters after both single-dose and multiple-dose vaccination schedules (Walpita et al., 2017). A study developed highly sensitive HiBiT-tagged NiV VLPs using plasmid-based expression of NiV structural proteins, allowing VLP production and functional testing under BSL-2-compatible conditions rather than relying entirely on live NiV work in BSL-4 laboratories. This platform is important for vaccine development because it enables rapid measurement of viral binding, entry, and antibody neutralization (Rajan et al., 2024). Another advance in this vaccine is the development of a transcription- and replication-competent NiV VLP system that models parts of the NiV life cycle without requiring infectious virus. This provides an additional experimental platform for studying NiV replication biology and evaluating antiviral or immunological interventions under safer laboratory conditions (Wang et al., 2024). Recently, Bezeljak and colleagues reported production of enveloped NiV-like particles in Sf9 insect cells using a single recombinant baculovirus co-expressing NiV F, G, and M. The new insect-cell platform incorporated a multi-step monolith chromatography purification workflow, including steric exclusion chromatography followed by cation- and anion-exchange steps, as opposed to previous NiV VLP approaches dependent on mammalian-cell expression and density-gradient ultracentrifugation. In Syrian golden hamsters, a single 25 µg dose of these insect-cell-derived NiVLPs induced systemic anti-NiV G IgG responses within 14 days. This suggests rapid binding-antibody immunogenicity that could be useful for outbreak-response vaccination strategies. Neutralizing antibody titres remained limited compared with total IgG responses, showing that antigen quantity alone may not be sufficient (Bezeljak et al., 2025). Glycoprotein conformation, glycosylation, adjuvant choice, and particle purity require further optimization. Therefore, in the last 2 years, NiV VLP vaccines have not moved into clinical trials, but they have provided important enabling advances like safer BSL-2-compatible VLP assays, life-cycle-modelling VLP systems, and a more scalable insect-cell/chromatography manufacturing route. Overall, this shows NiV VLP vaccines remain behind mRNA, viral-vector, and protein-subunit candidates in clinical translation, but recent work has strengthened the platform’s manufacturability.

6.1.5 DNA vaccines

DNA-based vaccines have several advantages, such as rapid production, easy design, no need for ambient cold temperature, and the capability to neutralize viral particles by eliciting an appropriate immune response, making it important especially for outbreak areas (Porter and Raviprakash, 2017). Early NiV DNA vaccine studies involved cloning the G and F glycoprotein genes from infected Vero E6 cells via RT-PCR. This followed insertion into vectors such as the pT-Adv plasmid for sequence validation, which showed only a silent mutation in the G gene. These genes were subcloned into the DNA vaccine vector pV1J plasmid and delivered in mice using gene gun immunization. Enhanced antigen expression was achieved by parallel construction of recombinant vaccinia viruses using the NYVAC vaccinia virus system. Immunogenicity was confirmed by ELISA and neutralization assays by detecting G- and F-specific antibodies (Guillaume et al., 2004). A study in pigs demonstrated strong neutralizing antibody responses by recombinant vaccines expressing viral glycoproteins. It was observed that the G glycoprotein alone conferred complete protection against viral challenge, whereas the F protein provided only partial protection. These findings established G protein as the principal immunogenic determinant to be used in subsequent DNA vaccine design (Weingartl et al., 2006). Early development of NiV DNA vaccines was further advanced through the use of mammalian codon-optimized constructs encoding the viral glycoproteins F and G. The optimized genes were cloned into the pCAGGS vector to generate the pCAGGS-F and pCAGGS-G constructs. Co-expression of these constructs in BHK-21 cells was confirmed by functional protein expression in these cells by employing cell fusion and Western blot techniques. Intramuscular DNA immunization in BALB/c mice elicited antigen-specific IgG responses. G construct induced significantly higher antibody and neutralizing titres compared to F. Neutralization assays using pseudotyped VSV demonstrated antibody generation against the G protein that effectively blocked viral entry. Responses to co-immunization with F and G were not greater than those for G alone (Wang et al., 2006). In another study, a highly conserved F protein sequence was designed using bioinformatics and structural modelling. It showed strong similarity across circulating strains and predicted stability in a pre-fusion conformation. The optimized gene was expressed using the pVAX1 vector and in parallel with a chimpanzee adenoviral vector. These expression vectors were successfully expressed and functionally validated in mammalian HEK293T cells through Western blot analysis and immunofluorescence assays. Immunization studies revealed that the DNA vaccine induced measurable F-specific humoral and cellular immune responses in infected Syrian hamsters, which were significantly enhanced upon booster doses (Lu et al., 2023a). Further, DNA vaccine targeting the NiV G glycoprotein was evaluated using a codon-optimized construct (DNA-G) alongside a recombinant chimpanzee adenoviral vector (AdC68-G) expressing the same antigen. Both vaccines induced long-lasting neutralizing antibody responses and robust G-specific T-cell immunity in BALB/c mice. In Syrian golden hamsters, immunization conferred complete protection against lethal challenge with both NiV-M and NiV-B strains. This demonstrated the efficacy of G-based DNA vaccination as a priming component in combination vaccine strategies (Lu et al., 2023b). Engineered synthetic DNA vaccines against NiV were developed by using the consensus sequence of the F and G glycoproteins retrieved from the NCBI database and further modified by RNA optimization to improve RNA stability and the addition of an IgE leader sequence to enhance translation efficiency in human cells. Electroporation-enhanced intramuscular injection of each NiV-F and NiV-G into mice resulted in a potent cellular immune response against multiple epitopes of these antigens (Choi et al., 2022). Immunoinformatics-driven strategies have enabled the design of multi-epitope DNA vaccine candidates against NiV by integrating conserved B- and T-cell epitopes derived primarily from the G and F glycoproteins. Selected epitopes showing antigenicity, non-allergenicity, and population coverage were assembled into a chimeric construct incorporating suitable linkers and immune-enhancing adjuvant sequences. The optimized gene sequence was further adapted for expression in a plasmid backbone, such as the pET28a vector. This was then followed by in silico codon optimisation and structural modelling to ensure stability and efficient expression. Immune simulation analyses predicted activation of cytotoxic T lymphocytes and helper T cells. Although this study lacks experimental validation, it provides a rational framework for next-generation epitope-based DNA vaccine design targeting NiV (Shabbir et al., 2025). Many studies showed that DNA vaccines induced effective immune responses neutralizing the virus, but have not progressed to clinical trials. Their long-term efficacy and population-level applicability remain to be fully evaluated to move forward with a clinical trial (Lv et al., 2025).

6.1.6 Nanoparticle and novel platforms

Nanoparticle vaccines represent one of the advanced strategies for NiV vaccine development. It provides benefits in antigen delivery, stability, and immune activation, and allow multivalent antigen display for an improved B-cell receptor-mediated immune response compared to conventional vaccines. Nanoparticle-based delivery systems of vaccines enhance humoral and cellular immune responses by promoting efficient uptake of antigens. They protect antigens from early degradation, provide controlled release, and facilitate efficient uptake and processing by immune cells. Furthermore, their physicochemical properties can be altered to target specific cells, leading to enhanced vaccine efficacy (Diaz-Arévalo et al., 2020; Zhang and Cui, 2023). It has been reported that a homotypic virus-like nanoparticle vaccine displaying NiV G from both strains was constructed utilizing the self-assembling properties of ferritin. Compared to the NiV G subunit vaccine, this nanoparticle formulation generated significantly higher neutralizing antibody levels and provided complete protection against lethal NiV infection in Syrian hamsters. It also induced antibodies with strong cross-reactivity against both homologous and heterologous henipaviruses (Yin et al., 2024). Following this, another study demonstrated that ferritin-based nanoparticles displaying the NiV G head domain on the surface (NiV G-ferritin) and soluble NiV G head domain (NiV sG) provided complete protection against lethal NiV challenge in Syrian golden hamsters. NiV G-ferritin induced faster, stronger, and broader neutralizing antibody responses against multiple henipaviruses compared to NiV sG and sustained long-lasting immunity for at least 6 months in Syrian hamsters and mice (Zhou et al., 2024). SpyCatcher/SpyTag technology: FeNP-sG and FeNP-Ghead ferritin nanoparticle constructs have superior immunogenicity compared to the antigen alone. Studies have shown that FeNP-sG (residues 71–602) induces specific T-cell and humoral responses better than FeNP-Ghead (residues 182–602). Complete protection against NiV challenge in hamster models was achieved with two doses of 5 μg of FeNP-sG (Chen et al., 2024). To reduce possible side effects or off-target immune responses to the SpyCatcher linker, an intein-spliced FeNP-Ghead vaccine was developed. This two-layer nanoparticle gave 100% protection in hamsters at a dose as low as 1 μg (Chen et al., 2025).

Non-ferritin-based nanoparticles are used for the delivery of DNA and RNA vaccines. In pseudotyped virus systems, an mRNA vaccine encoding a 60-mer nanoparticle with the NiV G head domains has induced a strong neutralizing antibody response (Brandys et al., 2025). Nucleoside-modified mRNA encoding the soluble G glycoprotein formulated in LNPs has demonstrated potent immunogenicity in both pigs and hamsters (Lo et al., 2020). In pigs, these vaccines induced significant CD4 and CD8 T-cell responses and neutralized cell fusion mediated by both NiV-M and NiV-B (Pedrera et al., 2024). In hamster models, a single dose of mRNA-LNP encoding a homologous Henipavirus glycoprotein was able to protect up to 70% of animals challenged with lethal NiV (Sun et al., 2025). In conclusion, nanoparticle vaccines represent a versatile platform with the potential to overcome limitations of conventional vaccine platforms.

6.2 Safety and reactogenicity profiles of clinical-stage vaccine candidates

NiV vaccine development occurs within a unique regulatory and clinical context. Unlike other vaccines where endemic diseases are found, the NiV outbreak occurs in a specific geographic location, unpredictable in time, and unethical to do a placebo-controlled efficacy trial. Thus, the process of moving vaccine candidates from preclinical to clinical study phase has been tough, making it very crucial for Phase I safety characterization. Furthermore, the rapid evolution of vaccine platform technologies, particularly mRNA and viral vector approaches, necessitates careful comparative safety assessment to inform clinical trial design, regulatory approval pathways, and post-authorization surveillance strategies (Chua et al., 2000; Eaton et al., 2006; Chua et al., 1999). The four most clinically advanced NiV vaccine candidates now under human evaluation represent distinct technological platforms with differing inherent safety considerations. To provide an overview of the currently available clinical evidence, Table 2 summarizes the Phase I safety and reactogenicity profiles of the major clinical-stage NiV vaccine candidates. Detailed descriptions of each vaccine candidate are presented in the following sections.

TABLE 2. Comparative clinical safety profiles of clinical-stage NiV vaccine candidates.

Comparison of the available Phase I clinical safety data for HeV-sG-V, mRNA-1215, PHV02, and ChAdOx1 NipahB, including reactogenicity, adverse events, serious adverse events, laboratory safety findings, follow-up duration, and overall safety outcomes.

6.2.1 HeV-sG-V subunit vaccine

The recombinant HeV soluble glycoprotein G vaccine (HeV-sG-V), formulated with aluminium hydroxide (alhydrogel), demonstrated a favourable safety and tolerability profile in a Phase I, randomized, observer-blind, placebo-controlled clinical trial involving 192 healthy adults. Safety assessments included solicited local and systemic adverse events, unsolicited adverse events, medically attended adverse events (MAAEs), serious adverse events (SAEs), and clinical laboratory evaluations. Overall, 173 (90%) participants reported at least one adverse event, although only six (3%) participants experienced mild immediate adverse events, primarily injection-site pain. Among the participants during the 7-day post-vaccination observation period, 86% experienced solicited adverse events; local adverse events were noted in 127/160 (79%) participants who received the vaccine as opposed to 10/32 (31%) participants receiving the placebo, while systemic adverse events were noted in 106/160 (66%) participants who received the vaccine as well as in 21/32 (66%) participants who received the placebo.

The most common local adverse event was injection site pain that was seen in 127/192 (66%) participants; there were significantly higher occurrences of injection site pain in vaccine participants as compared to placebo participants (p < 0.05). There was an increase in the incidence of injection site pain in correlation with the increasing doses of the antigen; 75/90 (83%) participants with the 100-µg vaccine experienced injection site pain. There was injection site erythema in 13%–33% of participants who had the vaccine as opposed to 0%–11% of participants who had the placebo. Severe injection site erythema or swelling was noted in only 4 participants; no clustering of these events was observed in any group.

The most frequent systemic adverse events among those vaccinated included fatigue (73/160; 46%) and headache (70/160; 44%), but similar frequencies were noted among the placebo group (fatigue 15/32; 47% and headache 16/32; 50%), showing that there is no vaccine-specific signal for these adverse events. Only myalgia was significantly more common in the vaccinated than in the placebo group (p = 0.03). Febrile reactions were not common and were observed only in 2 cases (1% of those vaccinated). All systemic solicited adverse events were classified as mild to moderate in intensity, usually occurred within 2 days after vaccination and were self-resolving; four events (three headaches and one fatigue) persisted longer than 7 days of observation.

The incidence of unsolicited adverse events was similar among the groups of vaccinees and placebo subjects: 56/160 (35%) of the vaccinees and 10/32 (31%) of the placebo subjects in the 28 days after vaccination. The unsolicited adverse events which had been associated with the vaccine and which occurred in more than one individual were nausea in three individuals and pruritus in two individuals. There were no haematological or biochemical laboratory abnormalities, no hospitalizations, no deaths, and no vaccine-related serious adverse events reported.

No dose-limiting toxicity was seen at any of the dose levels studied (10 μg, 30 μg, or 100 µg), thus providing evidence for further dose escalation and testing. Follow-up and laboratory tests continued through Day 197 after vaccination (six and a half months), and immunogenicity follow-up was done at various intervals during this period (Frenck et al., 2025). These findings indicate that HeV-sG-V exhibits an acceptable safety profile, with predominantly mild-to-moderate It shows transient reactogenicity despite a dose-dependent increase in local adverse events, thereby supporting its continued clinical development.

6.2.2 mRNA-1215 vaccine

The mRNA-1215 vaccine formulated as a lipid nanoparticle carrying mRNA encoding the chimeric pre-fusion F protein fused with NiV glycoprotein G (Malaysian strain) was shown to be well tolerated in an open-label Phase I clinical trial including 40 healthy adult volunteers (18 women, 22 men; mean age 37 years, range 22–59 years) recruited from 11 July 2022 to 22 August 2023. The participants were randomized into four groups according to the escalating dose: 10 μg (n = 10), 25 μg (n = 10), 50 μg (n = 10), and 100 μg (n = 10) intramuscularly in a two-dose prime-boost regimen separated by 4 weeks, whereas one volunteer from the 10-μg dose group received the first dose only. Safety and immunogenicity data were collected during 14–16 months after enrolment until September 17.

Mild pain and/or tenderness at the injection site was the most common local adverse event, occurring in 33 out of 40 participants (82.5%). The incidence of injection site pain varied from 33% to 50% in the 10 μg dose group to 70%–100% in the 25, 50, and 100 μg dose groups. Local adverse events occurred primarily during the 7 days following the administration of the vaccine. However, there were 2 participants with local adverse events that lasted for more than 7 days, occurring up to 11 days post-injection. There were also 2 participants in the 100 μg dose group with severe local reactions, 1 with severe erythema and 1 with moderate pain.

Similarly, systemic reactogenicity was observed to be relatively mild to moderate and recorded among 20 out of the 40 subjects (50%), with a rise in frequency and severity based on vaccine dosage escalation and second vaccination. Malaise was the most common systemic adverse event and was observed in 13 subjects (33%) following the first vaccination and 15 subjects (39%) following the booster vaccination. This was followed by headache (35%) and myalgia (30%). There was a rise in the percentage of subjects with any systemic symptoms from 30% for the 10 μg group to 40%, 60%, and 80% for the 25 μg, 50 μg, and 100 μg groups, respectively. There was severe systemic reactogenicity, which was only noted in one subject in the 100 μg group. Five subjects had headache and malaise beyond day 7 post-vaccination but resolved in 8–12 days.

Ten mild to moderate unsolicited adverse events related to vaccination occurred in 9 participants (22.5%). One participant experienced two related events. The most commonly reported related unsolicited AE was leukopenia and occurred in 4/40 (10%) of participants (3 following the prime dose, one following the booster dose; most common in the 100 μg dose group, n = 2). Leukopenia was mild in severity, occurred at day 7 following the vaccination, and resolved on day 7 in 3 events and on day 22 in one event. Skin and subcutaneous tissue disorders were the second most commonly reported class of related unsolicited AEs and occurred in 3/40 (8%) of participants. Two participants (5%) in the 10 μg dose group experienced an erythematous axillary rash ipsilateral to the vaccine injection site following the first dose; rash onset occurred within hours of vaccination and resolved after 19 days in one case and 10 days after vaccination in another case, and was resolved within 6 days. One participant in the 100 μg dose group had moderate urticaria onset 12 days after the first dose; worsening followed the second dose, and the urticaria lasted approximately 6 months. Additional vaccine-related unsolicited adverse events included moderate diarrhea (1 participant, 10-μg group; onset Day 2, resolved same day), mild hypertension (1 participant, 100-μg group; onset Day 0, resolved Day 7), and moderate dizziness (1 participant, 100-μg group; onset Day 0, resolved Day 1).

There were no occurrences of serious adverse events, adverse events of special interest, new chronic medical conditions for more than 1 year, or study interruptions due to reactogenicity or adverse events. Dose-limiting toxicities were not noted. The doses of 10, 25, 50, and 100 μg of vaccine were well-tolerated; however, reactogenicity tended to increase at the highest dose of 100 μg.

Safety follow-up and immune response evaluation continued up to 14–16 months from the first immunization, including 13–15 clinic visits. The last study visit was performed on 17 September 2024. Immune responses were followed for at least 56 weeks (1 year) after the first vaccination, and continued presence of increased neutralizing antibodies in comparison with the baseline level was observed (Ploquin et al., 2026).

6.2.3 PHV02 (rVSV-based) vaccine

PHV02 (rVSV-ΔG-EBOV GP-NiVG) has advanced to Phase I clinical trials in healthy adults as a live attenuated rVSV-vectored NiV vaccine. Phase 1b trial (NCT06221813) is a randomized, observer-blind, placebo-controlled trial conducted to assess the safety and immunogenicity of three different dose levels of PHV02 in healthy adults aged 18–59 years. A total of 120 healthy subjects were recruited for the clinical trial up to June 2025 and received two intramuscular vaccinations at days 1 and 29 with two doses of 2 × 10⁵, 2 × 10⁶, 2 × 10⁷ PFU, or placebo for the follow-up period up to Day 197 (approximately 28 weeks post-prime vaccination). Prespecified safety outcomes were local and systemic reactogenicity, unsolicited adverse events, rash, arthralgia, neurologic adverse events, medically attended adverse events, serious adverse events, and viral shedding, as well as NiV-specific IgG and neutralizing antibody responses. As of now, the quantitative safety and immunogenicity data for this Phase I trial have not been published in any peer-reviewed literature or available on ClinicalTrials.gov (Public Health Vaccines LLC, 2025). Therefore, a direct comparison of PHV02 reactogenicity and safety profile with other NiV vaccine approaches is currently unavailable. This represents an important gap in the current comparative safety assessment of clinical-stage NiV vaccine candidates and underscores the critical need for timely publication of Phase 1b trial results to inform regulatory and clinical decision-making.

6.2.4 ChAdOx1 NiV vaccine

Currently, ChAdOx1 NipahB, a replication-deficient chimpanzee adenovirus vector expressing the NiV Bangladesh strain glycoprotein (G), represents the most clinically developed NiV vaccine candidate. The candidate showed promising efficacy in preclinical studies conducted in hamsters and African green monkeys, leading to the initiation of the ChAdOx1 NipahB vaccine first-in-human Phase I clinical evaluation in the United Kingdom (NIV001). The study enrolled 51 healthy adult volunteers, comprising an open-label sentinel cohort (n = 6) and a randomized, observer-participant-blinded main cohort (n = 45), to evaluate single-dose and prime-boost vaccination regimens. Interim analyses from four independent Data Safety Monitoring Board (DSMB) reviews reported no serious adverse events or safety concerns. Preliminary immunogenicity data demonstrated robust antibody responses against both NiV Bangladesh and Malaysia glycoprotein G after a single dose. Further enhancement of neutralizing antibody titres following booster vaccination was observed. Based on these favourable interim safety data and immunogenicity data from this ongoing study, the vaccine subsequently advanced to an ongoing Phase IIa randomized, double-blind, placebo-controlled clinical trial (NIV002; ISRCTN62461807) in Bangladesh to further evaluate its safety, tolerability, and immunogenicity. It is planned that 306 healthy individuals aged 18–55 years will participate in the study to determine the safety, tolerability, and immunogenicity of one- and two-dose vaccination schedules. The parameters of safety evaluation include solicited local and systemic reactogenicity within 7 days after each dose of the vaccine, unsolicited adverse events within 28 days, haematological and biochemical laboratory abnormalities, serious adverse events (SAEs), and adverse events of special interest (AESIs) throughout the whole study period. The main criterion of immunogenicity evaluation is the determination of NiV glycoprotein G-specific IgG antibody levels by ELISA 28 days after each vaccination, while the secondary criteria of immunogenicity include antibody kinetics and durability of immune response (Registry). Since the trial is still ongoing, there are no safety and immunogenicity data available to describe the clinical safety profile of ChAdOx1 NipahB. Specific quantitative data on the percentage of participants experiencing local adverse events, systemic adverse events, serious adverse events, dose-limiting toxicities, and total follow-up duration are therefore unavailable at this time.

6.3 Translational fidelity of animal models in NiV vaccine development

The vaccine efficacy data presented in this review are derived from several animal models, which include Syrian hamsters, ferrets, pigs, cats, and African green monkeys. While all of these animal models provide valuable data for vaccine design, they vary greatly in terms of their translational accuracy. The African green monkey is considered the highest translational model for vaccine efficacy prediction in humans due to its great homology with humans. AGM share a high degree of genomic similarity with humans on the genetic level and possess orthologs of MHC Class I and MHC Class II genes similar to HLA loci in humans (Heijmans et al., 2020). Because of the genetic and immunological similarity between the two species, one can reliably predict peptide binding in MHC and the type of T-cell response that is expected from a human population. Such prediction was made in the case of HeV-sG vaccine when 100% survival with rapid development of neutralizing antibodies by day 21 post-vaccination was seen in several AGM experiments (Bossart et al., 2012; Mire et al., 2014; Geisbert et al., 2021; Frenck et al., 2025) in Phase I human trials. Syrian hamsters are often used for rapid proof-of-concept experiments due to their susceptibility to NiV infection and development of respiratory and neurological forms of the disease. They have much lower MHC diversity, compared to hundreds of human HLA alleles distributed across multiple polymorphic loci. This extreme MHC constraint suggests immune responses shaped by a limited MHC repertoire cannot reliably predict the diverse HLA-mediated immune responses across human populations. Ferrets reproduce many respiratory and neurological manifestations of human NiV disease and therefore provide greater translational relevance. Pigs are a natural amplifying host of NiV and are particularly valuable for evaluating vaccines aimed at reducing zoonotic transmission and outbreak control. Consequently, protective efficacy demonstrated in rodent models should be interpreted as preliminary evidence to avoid overvaluing proof-of-concept studies and ideally confirmed in non-human primates before extrapolation to human clinical efficacy.

7 AI-assisted vaccine design for Nipah virus

Immunoinformatic-based vaccine design has emerged as a recent supportive strategy in vaccine research. It helps in rapid epitope discovery, antigen prioritization, and rational multi-epitope construct development. Unlike conventional vaccine-development pipelines that require prolonged antigen screening and experimental optimization, computational vaccinology approaches can rapidly analyse NiV proteomic sequences to identify conserved B-cell, cytotoxic T-lymphocyte (CTL), helper T-lymphocyte (HTL), and IFN-γ-inducing epitopes. Further, it leads to antigenicity, immunogenicity, allergenicity, toxicity, and HLA population coverage prediction. It utilizes molecular docking, molecular dynamics simulation, immune simulation, codon optimization, and in silico cloning. Recent studies have applied immunoinformatics and reverse-vaccinology workflows to design multi-epitope NiV vaccine candidates targeting surface glycoproteins G and F, nucleoprotein N, and the complete NiV proteome. Early computational studies identified conserved G-protein epitopes as potential peptide-vaccine components, while later studies expanded toward multitope or chimeric vaccine constructs incorporating CTL, HTL, and B-cell epitopes from multiple NiV proteins. In a study, five CTL, five HTL, and six linear B-cell epitopes were used to design a chimeric construct predicted to interact with TLR2 and TLR4 (Shahab et al., 2024). In 2025, computational studies continued to broaden antigen selection, including nucleoprotein-based vaccine design and multi-antigen constructs targeting G, F, M, L, P, and N proteins. This suggests that AI-assisted pipelines may help identify conserved epitopes beyond the traditionally dominant G and F antigens (Shabbir et al., 2025). A list of AI-assisted and immunoinformatic-based NiV vaccine-design studies is provided in Table 3. A particularly important translational update is the 2025 CEPI-supported programme involving Gennova Biopharmaceuticals and Houston Methodist Research Institute, in which AI technology is being used to optimize NiV protein targets for a self-amplifying mRNA vaccine candidate intended for preclinical and clinical development. This represents a shift from purely in silico vaccine design toward integration of AI-guided antigen optimization with an experimentally developable RNA vaccine platform (Coalition for Epidemic Preparedness Innovations (CEPI), 2025). Despite these advances, AI-designed NiV vaccines remain at an early stage, because most published candidates are computational predictions and lack experimental confirmation of antigen expression, immunogenicity, neutralizing-antibody induction, protective efficacy, safety, and manufacturability. Therefore, AI should currently be viewed not as an independent validated vaccine platform, but as an enabling technology that can accelerate antigen discovery, improve epitope prioritization, guide structure-based immunogen design, and support next-generation platforms such as mRNA, self-amplifying RNA, DNA, nanoparticles, and viral vectors. Future progress will require systematic validation of predicted epitopes through HLA-binding assays, T-cell stimulation assays, antibody-binding studies, pseudovirus and live-virus neutralization assays, and protective-efficacy testing in relevant animal models such as hamsters, ferrets, pigs, and African green monkeys. The major recent advance in AI-guided NiV vaccinology is characterized by a decisive paradigm shift away from purely descriptive, in silico immunoinformatic screenings toward platform-linked antigen optimization seamlessly integrated with physical delivery systems. Historically, computational studies on NiV were restricted to static, theoretical mapping of potential B- and T-cell epitopes, utilizing basic algorithms to predict theoretical criteria like antigenicity, allergenicity, and toxicity without validating how these sequences behave in an actual biological system (Sharma et al., 2025; Srivastava et al., 2023). Modern AI frameworks have advanced this field by treating vaccine design as a holistic engineering problem, coupling machine-learning-driven sequence design directly with next-generation delivery platforms—such as mRNA scripts, viral vectors, or adjuvanted proteoliposome nanoparticles. By incorporating structural biology parameters like 3D protein folding, consensus-sequence engineering, and molecular dynamics of viral envelope proteins into generative AI workflows, researchers can now optimize structural stability and conformational precision in silico before feeding these precise instruction cassettes straight into advanced cell-free synthesis or nanoparticle-based manufacturing pipelines (Cao et al., 2025; Hu et al., 2025). This structural and platform-centric convergence bridges the gap between digital prediction and biological reality, transforming computational vaccine candidates into hyper-stable, physically expressible countermeasures capable of eliciting robust, broad-spectrum protective immunity against lethal Henipavirus threats (Srivastava et al., 2023; Cao et al., 2025; Hu et al., 2025).

TABLE 3. Overview of immunoinformatics and reverse vaccinology-based vaccine design strategies developed against NiV, highlighting target antigens, computational approaches, and major predicted immunological outcomes.

The current progress in immunoinformatic databases specialized for the NiV is exemplified by dedicated repositories like NipahVR, which catalog structurally curated genomic, diagnostic, therapeutic, and epitopic features of the virus (Gupta et al., 2020). These databases extract and index experimentally validated or computationally mapped linear B-cell epitopes, CTL, and HTL binders from critical NiV structural targets such as the G, F, and N proteins, collated alongside general sequence resources like GenBank and the Immune Epitope Database (IEDB) (Gupta et al., 2020; Kumar et al., 2024). Architecturally, the development of these immunoinformatics databases relies on a decoupled front-end and back-end framework designed to optimize data retrieval and user experience. The back-end is constructed using structured relational database management systems like MySQL or PostgreSQL, coupled with server-side scripting to seamlessly handle intensive queries, BLAST alignment tasks, and dynamically fetch multi-targeted therapeutics like siRNAs, sgRNAs, and epitope models. Conversely, the front-end interface is built using standard web technologies, including HTML5, CSS3, and JavaScript libraries (such as Bootstrap or jQuery) to provide an interactive, responsive, and user-friendly GUI that allows global researchers to query complex immunogenic variants and downstream molecular docking predictions cleanly across multiple device platforms (Racz et al., 2014; He et al., 2010).

These computational predictions of vaccines show an increasingly prevalent shift from predicting epitopes for a single antigen towards the reverse vaccinology-based proteome-wide approach combined with machine-learning-assisted identification of the best antigens and multi-epitope vaccine design. Although most computational analyses identified G and F glycoproteins as the most immunogenic vaccines, some recent studies have attempted to increase the diversity of epitopes by considering more antigens, including structural and non-structural N, P, V, W, C, and L proteins. Nonetheless, these results are largely theoretical, as most of the identified vaccine candidates have yet to undergo any experimental verification. Indeed, without performing experiments involving antigen production, testing immunogenicity using animal models, and conducting neutralization and challenge studies, the results on immunogenicity, protective efficacy, or safety will remain purely theoretical.

8 Current challenges and future perspectives

Despite substantial progress in NiV vaccine development, no vaccine has yet been licensed for human use. The recent entry of multiple candidates into human evaluation marks an important transition from purely experimental vaccine research toward clinical translation (Ploquin et al., 2026; Frenck et al., 2025). Among the most advanced platforms, recombinant soluble glycoprotein subunit vaccines, mRNA vaccines, recombinant vesicular stomatitis virus vectors, and chimpanzee adenoviral vectors have shown encouraging progress and should remain priority candidates for further development (Brandys et al., 2025; Lu et al., 2023b; Pigeaud et al., 2026). Nevertheless, an important milestone has already been achieved in the broader Henipavirus field through the licensure of Equivac® HeV, the only licensed Henipavirus vaccine, approved for use in horses in Australia (Middleton et al., 2014). It is a recombinant subunit vaccine composed of the HeV-sG formulated with a proprietary adjuvant (Zoetis). The recombinant HeV-sG antigen is produced using mammalian expression systems, with the commercial vaccine manufactured using a Chinese hamster ovary (CHO) cell expression system to achieve high antigen yield and consistency. This vaccine was initially registered for limited use as an emergency registration with the Australian Pesticides and Veterinary Medicines Authority (APVMA) in 2012 due to the pressing need for reducing spillover of the HeV from horses to humans. Following the accumulation of additional data on vaccine safety, efficacy, immunogenicity, and duration of immunity, several amendments were made to the product label, including approval for use in pregnant mares and modifications to the booster schedule. The vaccine received full registration from the APVMA in August 2015, and annual (12-monthly) booster vaccinations were subsequently approved in May (APVMA Gazette No. 17 (25 August 2015)). Experimental challenge studies demonstrated that two doses of the commercially formulated recombinant HeV-sG subunit vaccine completely protected all seven horses challenged with HeV 21 days after the second vaccination. Protection persisted for at least 6 months in two of three vaccinated horses, while the third horse remained clinically healthy with only transient low-level detection of viral genome in nasal secretions and no recoverable infectious virus. Vaccinated horses showed no clinical disease following challenge, indicating that the vaccine effectively prevented both disease and significant viral replication. Vaccinated horses with pre-challenge neutralizing antibody titres of 16 or 32 were protected from clinical disease. No vaccine-related serious adverse effects were reported in the vaccinated horses used in preclinical studies. Post-registration surveillance demonstrated a favourable safety profile, and subsequent safety data supported approval of the vaccine for use in pregnant mares. Given the high antigenic similarity between the HeV and NiV glycoprotein G proteins, the successful development and licensure of Equivac® HeV provides strong proof-of-concept for recombinant G glycoprotein-based vaccine strategies and establishes an important translational and regulatory pathway for the development of human NiV vaccines. A major future priority is to define reliable and long-term immune correlates of protection. Most NiV vaccine studies measure serum neutralizing antibodies directed against the attachment glycoprotein G, because G mediates receptor binding and, along with the fusion protein, nowadays acts as a dominant target of protective humoral immunity (Bossart et al., 2012; Loomis et al., 2021; Lu et al., 2023a). However, protective thresholds remain incompletely defined across platforms and animal models. Future studies should therefore harmonize neutralization assays, standardize results produced by pseudovirus and live-virus testing systems, and systematically compare immune responses across hamsters, ferrets, pigs, and AGM (Alam et al., 2025; Nie et al., 2019). This is especially important because sporadic NiV outbreaks make conventional phase III efficacy trials difficult. Beyond identifying correlates of protection, the development of next-generation vaccine platforms that enhance antigen presentation and immune activation will also be critical. Unlike conventional vaccines that rely on passive uptake of antigens by antigen-presenting cells (APCs), the CD40-targeted vaccine strategy exploits the CD40 receptor expressed on dendritic cells, macrophages, and B cells to facilitate receptor-mediated antigen delivery. CD40 engagement promotes APC maturation, enhances antigen processing and presentation, and increases the expression of co-stimulatory molecules required for efficient activation of both CD4⁺ and CD8⁺ T-cell responses. In this regard, the CD40-targeted antigen-presenting-cell vaccine strategy reported by Pastor et al. represents a conceptually distinct approach compared to conventional antigen-based vaccines. The rationale underlying this strategy is to leverage CD40, a co-stimulatory molecule constitutively expressed on the surface of APCs (dendritic cells), to specifically direct vaccine antigen to these critical immune cells rather than relying on passive uptake of soluble protein. By fusing the NiV G glycoprotein to an anti-CD40-targeting antibody, receptor-mediated endocytosis delivers the antigen directly to the endosomal compartment of dendritic cells, thereby enhancing cross-presentation to cytotoxic T lymphocytes (CD8⁺ T cells). Importantly, CD40 ligation simultaneously delivers a maturation and co-stimulatory signal to the dendritic cell itself, amplifying both the antibody and T-cell priming response from a substantially lower antigen dose compared to soluble glycoprotein alone. Experimental validation of this approach demonstrated that receptor-mediated delivery of the G glycoprotein significantly enhanced immunogenicity compared with soluble G protein. It induced higher G-specific IgG titres by Day 21 (p < 0.01), greater antibody avidity after booster immunization (Day 42; p < 0.05), stronger neutralizing antibody responses (p < 0.01), and robust B-cell and IFN-γ-producing T-cell responses. In African green monkeys, the vaccine generated detectable 1gG and IgA responses within 10 days of primary immunization. It maintained neutralizing antibodies for approximately 100 days after boosting, and provided 100% protection against lethal NiV-B challenge. This highlights the potential of dendritic cell-targeted vaccination to improve vaccine potency while reducing antigen dose requirements (Pastor et al., 2024). Compared to conventional NiV vaccine platforms, this CD40-targeted approach offers key advantages. Improved potency enables efficacy at lower antigen doses. It does not require a viral vector backbone (avoiding vector-immunity concerns), and it can function as a standalone recombinant protein–antibody fusion product. However, the approach also carries limitations relevant to clinical translation. Manufacturing and quality control of a recombinant monoclonal antibody fusion protein is substantially more complex and costly than soluble glycoprotein alone. The regulatory pathway for such bispecific constructs in human vaccines is not yet established, and the strategy has not yet entered human clinical trials. Such antigen-targeting strategies provide proof-of-concept of dendritic cell-targeted vaccination to improve vaccine potency while reducing antigen dose requirements and may complement existing subunit, mRNA, nanoparticle, and viral-vector platforms in future NiV vaccine development.

Another important future direction is the development of vaccines suitable for rapid outbreak deployment. WHO’s target product profile for NiV vaccines emphasizes a rapid onset of protection, ideally within 2 weeks after the first dose, and preferably strong efficacy after a single dose for reactive immunization in outbreak settings (World Health Organ ization, 2017). This requirement has important implications for platform prioritization. Viral-vectored vaccines, mRNA vaccines, and optimized subunit vaccines should be evaluated not only for peak antibody titres but also for speed of immune induction, durability, safety, and feasibility of use in rural or resource-limited outbreak areas. In this context, the CEPI-led “100 Days Mission” has further emphasized the importance of vaccine platforms capable of rapid antigen redesign, scalable manufacturing, and accelerated clinical deployment during emerging infectious disease outbreaks. These preparedness frameworks may be particularly relevant for the NiV because of its epidemic potential and unpredictable outbreak dynamics (Williams et al., 2026; Kim et al., 2025). A vaccine requiring multiple doses may still be valuable for high-risk populations, healthcare workers, laboratory personnel, and animal handlers, but a single-dose or rapidly protective vaccine would be more useful during sudden outbreaks.

Given that NiV outbreaks are often sporadic, geographically isolated, and have a small chain of transmissions, a traditional phase III efficacy trial is probably not feasible. In this setting, ring vaccination has been put forward as a possible method for vaccine deployment whereby vaccination is carried out on the index patients, their contacts, and contacts of contacts. This means that, at the same time, the vaccination method can contribute to both controlling the outbreak and assessing its clinical effects. Application of this approach to NiV infection poses particular difficulties. The small scale of the outbreaks and their unpredictability create problems for statistical analysis of efficacy; quick detection of cases, contact tracing, confirmation of the lab results, and vaccine administration are crucial components of a successful ring vaccination campaign. Additionally, there are no established immune correlates of protection in this case, which creates difficulties in interpreting the immunogenicity data collected during outbreak-based trials. Recent modelling of the virus transmission shows that, under the present epidemiological conditions, reactive vaccination is more beneficial to public health than proactive vaccination, especially in combination with an efficient surveillance system and emergency stockpiling of vaccines (Cortes-Azuero et al., 2026). These considerations support the incorporation of ring vaccination into future preparedness frameworks while highlighting the need for integrated one health surveillance, adaptive regulatory pathways, and outbreak-ready clinical protocols to enable timely evaluation and deployment of candidate vaccines.

Future vaccine design should also address antigenic breadth. NiV-M and NiV-B differ epidemiologically and clinically. Vaccine candidates designed should ideally protect against both lineages. Broadly protective vaccines targeting conserved regions of G and F, or multivalent formulations incorporating antigens from multiple henipaviruses, may reduce the risk of immune escape and provide preparedness against related emerging henipaviruses. Structure-guided immunogen design, stabilized prefusion F constructs, G-head nanoparticles, and mosaic or ferritin-based display platforms may help improve the magnitude and quality of neutralizing antibody responses. In parallel, immunoinformatics-driven multi-epitope vaccine design approaches are increasingly being explored to identify conserved B-cell and T-cell epitopes with broad population coverage, particularly for South and Southeast Asian populations, where NiV outbreaks are most prevalent (Afnani et al., 2026). AI-assisted vaccine design may complement conventional vaccine platforms by enhancing cross-population immune responsiveness and improving antigen precision. AI-assisted vaccine design complements conventional platforms by shifting discovery from empirical screening to computational engineering, specifically addressing limitations in population-level coverage and antigen precision (Gote, 2026). While conventional methods often struggle with highly variable structures and diverse human genetics, deep learning architectures optimize multi-epitope topologies by predicting pan-specific HLA binding profiles to ensure broad cross-population immune responsiveness across polymorphic genetic backgrounds (Gote, 2026; Yang et al., 2026). Furthermore, by leveraging Graph Neural Networks (GNNs) and de novo structural generative models, AI enhances antigen precision, mapping sequence spaces to isolate the exact minimum structural units needed to elicit protective immunity while pruning away non-neutralizing decoys, structural toxins, or allergens (Gabellieri et al., 2026; Lastra et al., 2026). When integrated upstream of conventional mRNA, viral vector, or subunit platforms, this computational pipeline transforms delivery systems into hyper-targeted, globally resilient countermeasures (Elfatimi et al., 2025).

Along with manufacturing and deployment considerations, a successful NiV vaccine should be scalable, affordable, thermostable, or compatible with standard 2 °C–8 °C cold-chain systems so that it could be deployed in Southeast Asian regions where spillover risk is highest. Recent pandemic preparedness perspectives have additionally highlighted the importance of strengthening regional manufacturing capacity, technology transfer, and equitable vaccine access in low- and middle-income countries, particularly for outbreak-prone regions where delayed vaccine availability could substantially worsen epidemic outcomes (Hernandez-Ruiz et al., 2025). Future trials should also include populations relevant to outbreak response. Adults, healthcare workers, caregivers, and potentially occupational groups with high animal or bat-exposure risk from endemic regions should be included in clinical trials. In parallel, vaccine stockpiling, ring-vaccination strategies, and emergency-use regulatory pathways should be planned before outbreaks occur. The future of NiV vaccine development should move from platform demonstration toward translational readiness. The most urgent needs are standardized immune correlates, comparative evaluation of leading candidates, demonstration of protection against both major NiV lineages, and clinical testing in at-risk regions. If these gaps are addressed, NiV vaccine development may progress from early-phase clinical evaluation to practical outbreak-preparedness tools capable of reducing mortality and preventing human-to-human amplification.

9 Conclusion

NiV remains one of the most important zoonotic viral threats because of its high case-fatality rate. Recurrent spillover in South and Southeast Asia, leading to increased human-to-human and nosocomial transmission, has increased the need for licensed human vaccines. Over the past two decades, vaccine research has progressed substantially from early antigen-identification studies to diverse preclinical platforms and, more recently, to early-phase clinical evaluation. Most vaccine strategies have focused on the viral G and F glycoproteins, which mediate receptor binding and membrane fusion, respectively. These proteins represent the major targets of neutralizing and protective immune responses. A comparative assessment of the vaccines indicates that subunit, viral-vectored, and mRNA-based vaccines currently represent the most advanced NiV vaccine platforms. Multiple complementary indicators, including immunogenicity endpoints, protection rates, speed of immune induction, and cold-chain requirements across animal models and species, support this ranking. Among these, subunit vaccines demonstrate the strongest overall evidence due to consistent complete protection in multiple animal models, including ferrets and African green monkeys, and have advanced into human clinical testing. Viral-vectored vaccines similarly exhibit high translational potential. Complete protection in relevant animal models while eliciting both humoral and cellular immune responses is observed in the case of these vaccines. VSV and ChAdOx1-based vaccines offer important advantages for outbreak preparedness because they can induce robust immune responses and may provide rapid protection with a limited dose. Although mRNA vaccines remain comparatively early in development, recent advances have accelerated their clinical translation. Phase I evaluation of mRNA-1215 demonstrated rapid induction of neutralizing antibodies, highlighting the adaptability of the platform for emerging viral threats. In parallel, VLP and nanoparticle vaccines provide native-like or multivalent antigen display that improves B-cell activation and neutralizing antibody responses. A detailed comparative evaluation of the leading experimentally validated NiV vaccine platforms, including immunogenicity, protective efficacy, speed of immune induction, duration of immunity, storage requirements, and translational readiness, is presented in Table 4. Before clinical translation, these platforms require additional optimization of antigen conformation, glycosylation, manufacturing scalability, and protective efficacy. Despite these advances, several major challenges remain. The sporadic and geographically restricted nature of NiV outbreaks makes conventional phase III efficacy trials difficult. This limitation increases the need for validation in animal models, immune correlates of protection, and standardized neutralization assays. Overall, we conclude that NiV vaccine development has entered a clinically relevant phase, as seen in many different platforms. A safe, broadly effective, rapid, and long-term antibody response would represent a major advance in preventing future NiV outbreaks and reducing the risk of wider epidemic or pandemic spread.

TABLE 4. Detailed comparative overview of experimentally validated NiV vaccine platforms, summarizing antigen composition, animal models, immune responses, protective efficacy, duration of immunity, and storage requirements.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. During the preparation of this manuscript, ChatGPT (OpenAI) and Claude Opus (Anthropic) were used solely to assist with language editing, grammar correction, improvement of readability, and refinement of writing style. These tools were not used for literature retrieval, reference generation, data analysis, interpretation of scientific findings, or formulation of conclusions. All scientific content, interpretation, literature verification, and reference checking were independently performed and verified by the authors. The authors accept full responsibility for the accuracy, integrity, and originality of the manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Zeeshan Ahmad, The University of Manchester, United Kingdom

Reviewed by: Hasanain Odhar, Al-Zahrawi University College, Iraq

Maryam Ehsasatvatan, University of Tabriz, Iran