Globally, it is estimated that respiratory syncytial virus (RSV) causes 33 million new episodes of acute lower respiratory tract infection (LRTI) in children <5 years of age and ≈120,000 deaths annually. In infants, RSV represents the leading cause of hospitalization worldwide and the second commonest cause of mortality in low- and middle-income countries.1,2 RSV also causes significant disease in immunocompromised hosts and the elderly and has been associated with the development of asthma.3 The increasingly recognized burden of RSV disease has made the development of a vaccine(s) a global health priority. The World Health Organization recently released a roadmap to facilitate the development and implementation of vaccines and monoclonal antibodies (mAbs) and estimated that RSV vaccination will be available in the next 5–10 years.4 This review summarizes the strategies and challenges associated with RSV vaccine development and the vaccine candidates undergoing clinical evaluation, with a focus on those geared toward the pediatric population.
THE STRUCTURE OF RSV
RSV has a negative sense nonsegmented RNA genome that encodes 11 proteins: 3 are nonstructural (NS1/NS2—that counteract interferon responses—and M2-2), and 8 are structural proteins. Of those 8 proteins, 5 are internal [N, P, M, M2-1, L]), and 3 are embedded in the virion membrane: the small hydrophobic (SH), fusion (F) and attachment (G) glycoproteins. RSV G and F carry antigenic determinants that elicit neutralizing antibodies. However, F is the preferred target for vaccine, mAb and antiviral development because it plays an essential role in host cell viral entry, is highly conserved within and among RSV A and B subtypes and because of its 6 antigenic sites that elicit the production of high-potency neutralizing antibodies (≥90% of neutralizing antibodies are directed against this protein).5 Most of the G protein is covered in glycans, leaving the central conserved domain available for neutralizing antibody binding. Except for this domain, G is not well conserved and it is recognized by few neutralizing antibodies, which has reduced enthusiasm for it as a vaccine target.
Our understanding of the F protein in its 2 conformations, prefusion (pre-F) and postfusion (post-F), has revolutionized the field of RSV biology. Pre-F, the active form of F on the virion, is metastable and switches unpredictably to the stable post-F conformation that once it is folded cannot return to the pre-F form. Antibodies that bind to pre-F are more efficient at neutralizing RSV than those against post-F. As examples, antibodies against site ϕ, a pre-F-specific epitope, are 150 times more potent than palivizumab that binds to site-II, present in both F conformations, while antibodies against site I, exclusively present in post-F, show weak or no neutralization.5 In addition, non-neutralizing antibodies to F, G and also SH, may inhibit infection by complement-mediated neutralization or antibody-dependent cell-mediated cytotoxicity. Furthermore, all viral antigens have the potential to induce protection by T-cell-mediated immunity.
CHALLENGES FOR RSV VACCINE DEVELOPMENT
Despite the burden associated with RSV, and after 60 years of active research, there is no licensed vaccine due in part of our incomplete understanding of the pathogenesis of the disease. In general, primary RSV infections are more severe; however, reinfections are common throughout life as immunity is neither complete nor long-lasting. The ideal vaccine should induce a more durable and improved immune response than natural infection.
Legacy of the Formalin-inactivated Vaccine
RSV vaccine development has been hindered after the safety concerns of the first RSV vaccine that was developed in the 1960s. The formalin-inactivated-whole virus alum-precipitated vaccine, which recent evidence indicating that it was directed against post-F, was associated in naive infants, but not older children, with enhanced RSV disease (ERD) and 2 deaths upon subsequent exposure to natural RSV. The mechanisms of ERD are not well understood, but it appears that an excess of non-neutralizing antibodies coupled with a skewed T-helper 2 (Th2) immune response, and complement deposition in the lungs contributed to its development. This is a critical aspect that is being considered for the development of inactivated vaccines, and strategies to assess safety risks according to the different vaccine platforms in the infant population are required.
There are different age groups that will benefit from RSV vaccines, and these might require different approaches: young RSV-naive infants (<4–6 months), children >6 months and the elderly. Vaccination of older children (2–5 years of age) may also limit transmission, as older siblings frequently introduce RSV into the household.
Infants <4–6 Months
This age group has an immature/developing immune system characterized by low expression of interferon, abundance of regulatory T cells with tolerogenic reactivity and a limited B-cell repertoire because of inefficient generation of somatic hypermutations. All these factors are associated with a poor response to foreign antigens and the generation of high-affinity matured antibodies. In addition, the presence of maternal antibodies may interfere with vaccine immunogenicity. Young infants represent the main target population because the peak of severe RSV disease occurs in the first 2–3 months of life. This age group would likely benefit from maternal vaccination or neutralizing mAbs administered at birth.
The main goal of maternal vaccination is to boost neutralizing RSV titers and thereby transplacental antibody transfer. However, the optimal timing for vaccination (2nd or 3rd trimester) and the durability of protection in the infant need to be defined. This coupled with the high prevalence of hypergammaglobulinemia in low- and middle-income countries, associated with HIV or malaria, which impairs transplacental antibody transfer, suggest the need for high maternal antibody titers to compete for transfer. Nevertheless, RSV antibody transfer through breast-feeding (IgG > IgA) may complement the maternal vaccination strategy.6 Vaccinating pregnant women could be questioned if it exclusively benefits the infant and not the mother. The limited data available in pregnant women are mostly derived from influenza surveillance studies with rates of RSV infection varying from 0.2% to 13%, which likely underestimates the real incidence of RSV during pregnancy. The concerns regarding adverse fetal outcomes are relatively low, because this would not be the first time the mother’s immune system encounters RSV antigens and the safety profile of other vaccines used in pregnancy, such as tetanus, diptheria and acellular pertussis (Tdap) or influenza, is excellent. A number of RSV maternal vaccines are currently in clinical development (Table 1).
Older Infants and Children
Based on the experience of the formalin-inactivated-RSV vaccine, within this age group those who are naive at the time of vaccination might be at risk of ERD with protein vaccines. This target population would likely benefit most from live-attenuated or vectored vaccines.
On the other side of the spectrum, the immunosenescence of adults >65 years of age and the presence of additional comorbidities may compromise vaccine responses and the ability to assess efficacy. This population might benefit most from adjuvanted vaccines.
The ideal vaccine should be able to prevent severe disease and limit transmission, but the lack of a standard definition of severe disease or precise markers to assess severity in infants has been a barrier for vaccine development. Clinical endpoints that define a successful vaccine might be different depending on the target population. Hospitalization and other endpoints that capture the inpatient/outpatient burden of the disease, such as a reduction in medically significant visits for RSV infection, should be considered.7 Developing composite endpoints that include a combination of viral (and possibly bacterial) factors, clinical parameters, and fast turn-around point of care biomarkers could help with patient classification and to standardize definitions.8 Also, long-term follow-up is recommended, as studies suggest that interventions reducing the acute burden of RSV disease may also impact the development of recurrent wheezing/asthma.9
Immune Correlates of Protection
Serum neutralizing antibodies (IgG against pre-F > post-F and G) represent the main surrogate of protection, as shown by the effectiveness of immunoprophylaxis with anti-F mAb (palivizumab) in high risk infants. However, a standardized protective threshold has not been defined yet. Newer systems biology approaches are helping to define the optimal correlates of protection, which are complex and depend on multiple factors, rather than a single cutoff value in antibody assays, and will need to be adjusted to each target population. Other cocorrelates of protection may include, F-specific epitope antibodies, mucosal IgA, interferon responses, antibody-dependent cell-mediated cytotoxicity and cell-mediated immunity. In addition, a balanced Th1/Th2 immune response, indicated by a high IgG2a/IgG1 ratio, is desirable.
The lack of an ideal animal model has also slowed down RSV vaccine development. Human challenge models mostly reproduce upper but not LRTI, limiting the generalizability of the results or the ability to assess the impact of vaccines on disease severity. There are also gaps in RSV epidemiology with lack of accurate information defining the temporal and geographic patterns of RSV circulation in inpatients/outpatients, across different age groups or RSV-associated mortality. Implementing robust multiplex polymerase chain reaction-based surveillance platforms could help to assess the impact of interventions on the burden of RSV disease, to identify possible escape mutants, or the contribution of other respiratory viruses causing RSV-like illnesses.
The most effective approach to protect young infants and children from severe RSV infection may be a combined strategy using passive and active immunization: either maternal vaccination with stabilized pre-F or virus-like particles containing the F protein or mAb against pre-F at birth; followed by pediatric active immunization with a live vaccine, either attenuated RSV or the pre-F protein expressed from a virus vector. There are 39 vaccines candidates under development (http://www.path.org); of those 19 are undergoing clinical trials (Table 1).10
The recombinant adjuvanted RSV post-F nanoparticle vaccine is the most advanced vaccine in clinical development. Results from a phase-3 clinical trial that enrolled 4636 pregnant women on the third trimester demonstrated a decrease in RSV hospitalizations in the offspring; however, the study did not meet the primary endpoint defined as prevention of medically significant RSV LRTI. The potential approval of this vaccine is being evaluated. It also aims to target elderly individuals and children >6 months to 5 years of age.
These vaccines consist of purified, adjuvanted proteins and use stabilized pre-F as the main antigen with promising results. They are mainly directed at pregnant women or the elderly because of the risk of ERD in RSV-naive infants. Other subunit vaccines in clinical or preclinical stages are using SH or G as main vaccine antigens.
There are 5 vector-based vaccines in clinical development. The first 4 use adenovirus as a vector, while the other uses a modified vaccinia Ankara virus. Two of them are intended for use in pediatric seronegative patients. All of these vaccines express RSV F (pre-F > post-F depending on the vaccine) and 2 of them also express other viral antigens (N, M2 or G proteins).
(LAVs) represent an attractive alternative for older infants and young children. LAVs are administered intranasally and are able to elicit broad innate, humoral and cellular responses and replicate in the respiratory tract despite the presence of maternal antibodies. Importantly, these vaccines have not been associated with ERD and are considered safer in infants. The use of reverse genetics has made possible to incorporate different mutations in the viral genome, making LAV sufficiently immunogenic and, except for rhinorrhea, not associated with adverse events. There are 6 intranasal LAVs undergoing phase-1 clinical trials; 4 are using attenuated RSV, one Sendai virus as a backbone expressing RSV F and the last one is a chimeric vaccine using bacille Calmette-Guerin (BCG). The BCG/RSV vaccine is the only LAV intended to be administered systemically (subdermal) and in newborns.
mAbs are also being evaluated for the prevention of RSV LRTI in young infants. Of those, suptavumab (REGN-2222), that targeted the pre-F-specific site V, has been discontinued from the market after it failed to prevent serious RSV LRTI in premature infants (primary endpoint). During the study, RSV type B was the predominant circulating strain and developed escape mutations that conferred resistance to this mAb. MK-1654 is an extended half-life mAb currently undergoing phase-I clinical trials and it is directed against antigenic site-IV (present the pre-F and post-F forms). Nirsevimab (MEDI8897) is a highly potent human neutralizing IgG1Κ targeting the pre-F-specific antigenic site ϕ. It also has an extended half-life because of modifications in the FC region using YTE technology. MEDI8897 is entering phase-3 clinical trials with the intent to provide passive immunization for prevention of severe RSV LRTI to all infants (preterm and full term), using a fixed, once per season intramuscular dose.
Over the past decade, there have been significant advances in our knowledge of RSV molecular and structural biology and in the understanding of the human immune response to RSV. Despite the barriers, there are several opportunities for RSV vaccine development to protect the most vulnerable populations. The increasing interest of academic, industry and international bodies, such as the World Health Organization or Bill & Melinda Gates Foundation, is helping to move the field forward, promoting the implementation of surveillance platforms and standardization of clinical definitions, assays and surrogate markers of protection.
1. Shi T, McAllister DA, O’Brien KL, et al; RSV Global Epidemiology Network. Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study. Lancet. 2017;390:946–958.
2. Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380:2095–2128.
3. Falsey AR, Hennessey PA, Formica MA, et al. Respiratory syncytial virus infection in elderly and high-risk adults. N Engl J Med. 2005;352:1749–1759.
4. Vekemans J, Moorthy V, Giersing B, et al. Respiratory syncytial virus vaccine research and development: World Health Organization technological roadmap and preferred product characteristics. Vaccine. 2018;S0264410X:3136431366.
5. Ngwuta JO, Chen M, Modjarrad K, et al. Prefusion F-specific antibodies determine the magnitude of RSV neutralizing activity in human sera. Sci Transl Med. 2015;7:309ra162.
6. Mazur NI, Horsley NM, Englund JA, et al. Breast milk prefusion F immunoglobulin G as a correlate of protection against respiratory syncytial virus acute respiratory illness. J Infect Dis. 2019;219:59–67.
7. Villafana T, Falloon J, Griffin MP, et al. Passive and active immunization against respiratory syncytial virus for the young and old. Expert Rev Vaccines
8. Mejias A, Dimo B, Suarez NM, et al. Whole blood gene expression profiles to assess pathogenesis and disease severity in infants
with respiratory syncytial virus infection. PLoS Med. 2013;10:e1001549.
9. Blanken MO, Rovers MM, Molenaar JM, et al; Dutch RSV Neonatal Network. Respiratory syncytial virus and recurrent wheeze in healthy preterm infants
. N Engl J Med. 2013;368:1791–1799.
10. Mazur NI, Higgins D, Nunes MC, et al; Respiratory Syncytial Virus Network (ReSViNET) Foundation. The respiratory syncytial virus vaccine landscape: lessons from the graveyard and promising candidates. Lancet Infect Dis. 2018;18:e295–e311.