Current State and Challenges in Developing Respiratory Syncytial Virus Vaccines

Respiratory syncytial virus (RSV) is the main cause of acute respiratory tract infections in infants and it also induces significant disease in the elderly. The clinical course may be severe, especially in high-risk populations (infants and elderly), with a large number of deaths in developing countries and of intensive care hospitalizations worldwide. To date, prevention strategies against RSV infection is based on hygienic measures and passive immunization with humanized monoclonal antibodies, limited to selected high-risk children due to their high costs. The development of a safe and effective vaccine is a global health need and an important objective of research in this field. A growing number of RSV vaccine candidates in different formats (particle-based vaccines, vector-based vaccines, subunit vaccines and live-attenuated vaccines) are being developed and are now at different stages, many of them already being in the clinical stage. While waiting for commercially available safe and effective vaccines, immune prophylaxis in selected groups of high-risk populations is still mandatory. This review summarizes the state-of-the-art of the RSV vaccine research and its implications for clinical practice, focusing on the characteristics of the vaccines that reached the clinical stage of development.


Epidemiology
Respiratory syncytial virus (RSV) is often responsible for severe seasonal respiratory disease in infants and elderly people, and it causes a great burden on health systems worldwide. RSV clinical presentation in children involves respiratory infections ranging from mild upper to severe lower respiratory tract infections (LRTI), including pneumonia or bronchiolitis [1][2][3][4]; in fact, it was estimated that RSV may cause up to 22% of all severe acute LRTI in young children. A recent systematic review [3] reported that worldwide in 2015, RSV caused 33.1 million episodes of LRTI, resulting in nearly 3.2 million hospitalizations and 59,600 in-hospital deaths in children younger than 5 years. In infants younger than 6 months, RSV caused 1.4 million hospital admissions and 27,300 in-hospital deaths [3]. The estimated overall RSV mortality in 2015 was 118,200 deaths in infants younger than 5 years. Moreover, RSV is responsible for pulmonary morbidity and hospitalizations also in the elderly and high-risk adults [5], causing more than 17,000 deaths for underlying pneumonia and circulatory complications [6]. Risk factors for severe RSV infections in paediatric populations that may require hospitalization and Intensive Care Unit admission include prematurity, chronic lung disease of Table 1. Summary of the RSV vaccines and monoclonal antibodies in clinical development by target population. Only the most advanced trial for a specific target group is reported. RSV: respiratory syncytial virus; NIAID: National Institute of Allergy and Infectious Diseases; mAb: monoclonal antibody.

Particle-Based Vaccines
Particle-based vaccines are synthesized by self-assembling nanoscopic particles that expose multiple copies of a selected antigen on their surface and mimic the native virions [23]. Thanks to the high copy number of the selected antigen and the immune-boosting properties of the particle matrix, these vaccines elicit strong humoral and cellular immune responses [23]. Moreover, the lack of the viral genome required for replication make them safe. To date, two nanoparticle-based RSV vaccines have been tested in clinical trials: the RSV F nanoparticle vaccine, without or with an aluminium adjuvant, and SynGEM ( Table 2). The RSV F nanoparticle vaccine, developed by Novavax, is composed of recombinant F-proteins, which have the post-F morphology and are formulated with polysorbare 80 [24]. The conformation of the F proteins is a singly cleaved pre-fusogenic form [25,26]. This vaccine candidate is being evaluated in women of childbearing age, pre-school children (2-6 years old) and the elderly (≥60 years old). In Phase I clinical trials, it has proven to be well tolerated and highly immunogenic in all the target populations [27][28][29][30]. Subsequent clinical trials have been conducted in pregnant women and the elderly. ResVax is a maternal RSV F nanoparticle vaccine with an aluminium phosphate adjuvant. It is being developed to protect infants from RSV disease via maternal immunization. ResVax has been shown to be safe in a Phase II trial enrolling 50 healthy third-trimester pregnant women and to elicit RSV neutralizing antibodies and palivizumab-competing antibodies that are efficiently transferred to the infants [31]. This maternal vaccine has been the object of the PREPARE trial, a Phase III multi-country, randomized, placebo-controlled trial evaluating the vaccine efficacy against RSV-LRTI in infants from birth to 90-180 days of life [32]. From December 2015 to March 2019, 4363 pregnant women with expected delivery near the beginning of the RSV season were enrolled in the study. Women were randomized to receive a single intramuscular dose of the vaccine (120 µg RSV-F protein adsorbed to 0.4 mg aluminium) or a placebo between 28 and 36 weeks of pregnancy. ResVax failed to meet the primary outcome of prevention of medically significant LRTI. However, it showed 44% efficacy in reducing RSV-LRTI hospitalization. Moreover, it demonstrated 39.4% efficacy in reducing RSV-specific medically significant LRTI and 58.8% efficacy in reducing RSV-related severe hypoxemia in young infants (<3 months of age). In addition, pneumonia was 49.4% less common in infants of the vaccinees than the placebo recipients [32]. According to these results, ResVax is the first RSV vaccine to show efficacy in a Phase III clinical trials, even if it did not meet the desired primary outcome. Because of the failure to achieve the primary outcome, according to the Food and Drug Administration (FDA) and the European Medicines Agency (EMA) recommendations, Novavax will conduct an additional Phase III clinical trial to confirm the efficacy of ResVax [33]. Regarding older adults, the non-adjuvanted RSV F nanoparticle vaccine failed to demonstrate efficacy in a 2015 Phase III clinical trial (RESOLVE trial). This study enrolled 11,850 adults ≥60 years of age randomized to receive 135 µg of the vaccine via intramuscular injection or placebo. The vaccine did not reduce the incidence of RSV-positive moderate to severe LRTI, nor the incidence of all RSV-symptomatic respiratory diseases [34]. However, the vaccine was associated with a 61% reduction in hospitalizations due to exacerbations of chronic obstructive pulmonary disease (COPD). On the basis of this result, Novavax is planning to start an efficacy trial to evaluate the COPD exacerbations as a prospective endpoint [35]. Following the unsatisfactory results of the RESOLVE trial, in 2017, Novavax conducted a second Phase II clinical trial in the elderly. The aim was to evaluate the safety and immunogenicity of single-or two-dose regimens of the RSV F vaccine with and without adjuvants (aluminium phosphate or Matrix-M1). All formulations and regimens were well tolerated. The study showed that both adjuvants increased the magnitude, duration and quality of the immune response versus the non-adjuvanted RSV F vaccine [35]. These data support the inclusion of the RSV F vaccine adjuvant formulations in future elderly trials. Another nanoparticle-based RSV vaccine that has been tested in clinical trials is SynGEM, developed by Mucosis. SynGEM is a mucosal vaccine containing the RSV F protein in the prefusion conformation bound to a bacterium-like particle (BLP) derived from Lactococcus lactis. BLP has the role to present the vaccine antigen in a more natural conformation and to boost the immune system of the virus [36]. The safety and tolerability of SynGEM have been evaluated in a Phase I trial enrolling 48 healthy volunteers [37]. The participants were randomly assigned to receive intranasal SynGEM (low or high dose) or a placebo according to a prime-boost schedule with the boost vaccination at 28 days after the prime. SynGEM was generally well tolerated and induce persistent antibody responses in healthy adults. High-dose recipients (350 mg RSV-F protein, 5 mg BLP) achieved plateau responses in antibody titres after the first dose. SynGEM induced bursts of plasmablast activity and mucosal IgA and elicit systemic RSV-specific antibodies (non-neutralizing and palivizumab-like antibodies) for at least 6 months. However, no detectable RSV neutralizing antibodies (F protein site ∅-specific antibodies) were found, despite preclinical data that had shown protection [38]. Future studies are needed to optimize the immunogenicity of SynGEM. In 2017, Mucosis declared bankruptcy, thus any further development of the vaccine may be planned by other companies. The SynGEM technology platform is now made available by the private company Virtutax [39].  Particle-based vaccines are promising both for young infants through immunization of the pregnant mothers and for the elderly. ResVax has already reached Phase III clinical trials for pregnant women, and hopefully new trials will soon allow its marketing approval. On the other hand, for elderly people, the use of RSV F vaccine adjuvant formulations in future trials may increase the magnitude, duration and quality of the induced immune response.

Vector-Based Vaccines
Vector-based vaccines use a carrier vector to deliver RSV antigens and induce an immune response against RSV components exploiting the adjuvant effect of the vector. Due to the chimeric nature of the vectors, there is no risk of reversion to wild-type RSV and of ERD. However, the presence of pre-existing anti-vector immunity or its potential development may challenge the clinical use of these vaccines. To date, eight RSV vector-based vaccines have been tested in clinical trials ( Table 3). The MVA-BN-RSV vaccine has been developed by Bavarian Nordic and it is based on a non-replicating modified vaccinia Ankara (MVA) virus, previously used as a vaccine against smallpox [43,44]. This vaccine candidate displays the RSV surface proteins F and G (for both A and B subtypes) and two internal RSV proteins N and M2. A Phase I study has shown that it can induce broad cellular and humoral immune responses against RSV [45]. A subsequent Phase II trial, evaluating the safety and immunogenicity of the vaccine in 420 older adults (≥55 years), showed that it is well tolerated and lead to a persistent immune response for at least 6 months with the possibility to boost it at 12 months [46]. On the basis of these results, a Phase III trial in the elderly has been planned for initiation in 2021 [47]. ReiThera developed two novel, recombinant, viral-vectored vaccine candidates for RSV, PanAd3-RSV and MVA-RSV. Both vaccines use RSV F, N and M2 proteins as antigens, delivered, respectively, by a replication-defective simian Adenovirus (PanAd3) and MVA vector. PanAd3-RSV and MVA-RSV, given in different combinations and by different routes of administration (PanAd3 either intramuscularly or intranasally, MVA intramuscularly), were well tolerated and immunogenic in healthy adults despite pre-existing natural immunity to RSV [48]. Similar results occurred in the elderly, who presented both humoral and cellular vaccine responses in a Phase I study [49]. The development of these vaccines has been halted since 2015. Three other virus-vectored vaccines use an Adenovirus vector to deliver RSV antigens. VXA-RSV-f, developed by Vaxart, is a molecularly adjuvanted Adenovirus serotype 5 based RSV vaccine encoding the F protein. It is being developed for older adults. Oral delivery of this vaccine has been shown to induce a strong humoral response against RSV in cotton rats [50]. Following these pre-clinical results, VXA-RSV-f was tested orally in healthy adults in a Phase I trial. The study ended in September 2017, but the results have not yet been published [51]. Janssen developed Ad26.RSV.Pre-F vaccine for the elderly and children, based on the human Adenovirus strain 26 vector expressing the F protein stabilized in the pre-fusion conformation [52]. This vaccine candidate demonstrated superior efficacy to the previously tested Ad26. RSV.FA2 that displayed post-F as an antigen. A Phase II study evaluated the co-administration of a seasonal influenza vaccine with the Ad26.RSV.preF vaccine in 180 healthy elderly people, showing an acceptable safety profile and no evidence of interference in immune response [53]. Two other Phase II trials are actually ongoing in the elderly. The first one aims to demonstrate the efficacy of the Ad26.RSV.Pre-F vaccine in the prevention of reverse transcriptase-polymerase chain reaction and confirmed the RSV-LRTI when compared to the placebo [54]. The second study evaluates the safety and immunogenicity of the Ad26.RSV.preF and/or RSV pre-F protein combinations [55]. Regarding children, a Phase I/IIa study evaluating the safety and tolerability of two doses given one month apart of Ad26.RSV.preF in adults and RSV-seropositive toddlers ended in April 2020. The results of the study have not yet been published [56]. Moreover, a Phase I/IIa trial started in January 2019 is evaluating whether an intramuscular regimen of thee doses of Ad26.RSV.preF is safe, well tolerated and immunogenic in RSV-seronegative toddlers [57]. ChAd155-RSV, developed by GlaxoSmithKline, is another Adenovirus-based vaccine currently in clinical development. This vaccine is intended for children and uses a viral vector chimpanzee-Adenovirus-155, encoding RSV F, N and M2-1 proteins [58]. In adults previously naturally exposed to RSV, ChAd155-RSV delivered intramuscularly was found to be well tolerated and to elicit specific humoral and cellular immune responses [58]. Phase II clinical trials in seropositive toddlers aged 12 to 23 months [59], and in likely seronegative infants aged 6 and 7 months [60], are still ongoing. Two vector based-vaccines use the parainfluenza virus (PIV) to display RSV antigens, with the aim to induce immunity against both viruses: MEDI-534, developed by MedImmune, and the recombinant Sendai virus vectored RSV (SeVRSV), developed by the National Institute of Allergy and Infectious Diseases (NIAID). MEDI-534 is based on a modified bovine PIV3, expressing the human PIV3 fusion, the human PIV3 hemagglutinin-neuraminidase and the RSV F proteins [61]. In a Phase I trial enrolling seropositive children, this live attenuated intranasal vaccine was safe but minimally immunogenic [62]. When tested in seronegative infants, who are the target population for this vaccine, MEDI-534 was well tolerated and induced an immune response against RSV in about 50% of the subjects and against hPIV3 in all cases [63]. The low immunogenicity against RSV is probably due to a decreased expression of the virus secondary to genetic changes involving the F protein [64]. Therefore, further studies are warranted to reach the genetic stabilization of MEDI-534 and increase the RSV immune response. The other PIV-based vaccine is the SeVRSV vaccine. SeVRSV is a replication-competent Sendai virus, a murine PIV1 strain, that carries the RSV F gene produced by reverse genetics technology. This vaccine has been reported to induce a humoral response and protect the lower respiratory tract from RSV in African green monkeys [65]. A Phase I trial to evaluate the safety and immunogenicity of intranasal SeVRSV in healthy adults has been completed in 2019 [66]. The results have not yet been published.
Vector-based vaccines are potentially good candidates for the paediatric population, because there is no risk of reversion to wild-type RSV and of ERD. These vaccines are still in an initial stage concerning their evaluation. The eight vaccines that have been tested in clinical trials are based on MVA, Adenovirus, bovine PIV or Sendai Virus vectors. The MVA-BN-RSV vaccine is the only one that has passed Phase 2 clinical trial, and a Phase 3 trial in the elderly has been planned for initiation in 2021.

Subunit Vaccines
Subunit vaccines are created with RSV protein fragments. They are poorly immunogenic due to their non-replicating nature and their limited components; therefore, booster administration and adjuvants are often necessary to make them effective [70]. This type of vaccine primarily induce a CD4 T cell response [71], with higher risk for vaccine-ERD in seronegative infants [20,21]. The adjuvants' activity is essential for the creation of neutralizing antibodies and a protective response, thanks to Toll-like receptors stimulation and B-cell affinity maturation, which also help prevent ERD [70]. Nowadays, subunit vaccines are under development only for pregnant women and older people that have already had a previous RSV infection and that are not a risk of developing ERD [14] ( Table 4). The most appropriate protein to design subunit vaccines is the F protein, mostly in prefusion conformation [71]. The postfusion conformation lacks important antigenic sites [72] and this could be the reason for the failure of some candidate vaccines [73]. NIAID is developing VRC-RSVRGP084-00-VP, a prefusion F-based RSV vaccine. The protein contained in this vaccine candidate is DS-Cav1, a secreted variant of F glycoprotein that has been stabilized in the prefusion native conformation thanks to protein engineering [74]. The Phase I trial of VRC 317 has recently been concluded (April 2020), and the results are expected to be published soon. During the trial, this vaccine has been tested in healthy adults with or without an aluminium adjuvant, with the aim of monitoring safety, tolerability and immunogenicity [75]. Preliminary results that were published in 2019 are promising: DS-Cav1 is able to increase the RSV neutralizing activity in serum from 7-to 15-fold, increasing the pre-F-specific antibodies [76]. Another vaccine candidate based on the prefusion F protein is being developed by Pfizer. There are two ongoing Phase I/II trials (with and without adjuvant) who enlist healthy adults of various age groups. Participants from certain arms of the studies will concomitantly receive a seasonal inactivated influenza vaccine. These trials will be completed at the end of 2020 and in 2021 [71,77,78]. Meanwhile, the Phase IIb trial, concluded in December 2019, recruited a group of healthy non-pregnant women (18-49 years old) to which the Pfizer's vaccine candidate was given together with diphtheria/tetanus/pertussis vaccine [79]. There is a last Pfizer's trial in progress: a Phase IIb study currently in progress will evaluate the safety, tolerability and immunogenicity of the stabilized prefusion F subunit vaccine in pregnant participants and will assess the safety and characteristics of the transplacentally transferred antibodies in their infants [80]. DepoVax (DPX)-RSV is a candidate based on the ectodomain of the SH protein (She), presented with a novel lipid-based formulation (DepoVax) that ensures a prolonged exposure of the antigen. Phase I results have shown a good safety profile and a sustained serum IgG response (lasting more than a year) [81]. Phase II has not yet started [82]. G glycoprotein was also used to create a subunit vaccine candidate, BARS13: the Phase I study has involved healthy adult volunteers, but the results have not yet been published [82,83]. Unfortunately, the vaccine called GSK3003891A, after two Phase I trials and three Phase II studies that have established its safety and immunogenicity, is no longer under development because of the instability of the pre-F antigen during manufacturing [14,71,82]. However, GSK is working on two new candidates, in either a prefusion or an undisclosed conformation: GSK3888550A, with pregnant women as the target population, and GSK3844766A, designed for older adults [82]. With regard to the first vaccine, only a Phase I trial has been concluded, but the results have not yet been published; a Phase II trial is assessing the safety and immune response in healthy pregnant women and in infants born to vaccinated mothers, and will be concluded in 2021 [84][85][86]. MEDI-7510, a subunit vaccine developed by MedImmune, has been proven to be immunogenic but it did not protect older adults from RSV illness in a Phase IIb trial, so studies to develop it are no longer in progress [87].
Subunit vaccines can be considered a safe choice: they do not contain live viruses that could return to a virulent state and they cannot induce an exceeding immune response. However, owing to concerns of ERD associated with protein-based vaccines, they are potentially good candidates only for pregnant women and the elderly. Observing the concluded trials and the studies in progress, induction of protective immunity and obtaining F protein conformation stability remain the major unsolved problems for creating an effective subunit vaccine.

Live-Attenuated and Chimeric Vaccines
Live-attenuated RSV vaccines (RSV-LAVs) are produced with versions of RSV that are able to replicate but have been modified to discourage severe disease. They can be created by traditional techniques (i.e., temperature or chemical sensitivity) or, thanks to an improved understanding of the RSV viral genome, by reverse genetics to create an attenuated replication-competent vaccine [91]. ERD has not been observed with RSV-LAVs or replicating vaccine vectors. For this reason, these candidates can be considered safe for naive-RSV infants [20,92]. Furthermore, RSV-LAVs have other benefits: the ability to replicate in the respiratory tract despite the presence of maternal antibodies, the capacity to promote both a humoral and cellular immune response and the possibility to be administered as nasal drops, which are less invasive and better tolerated in children [93,94] (Table 5). Most vaccines are in Phase I and no candidates of this type have progressed beyond Phase II clinical trials. Anyway, the replicative nature of RSV-LAVs and their major safety compared to other types of vaccine make them an attractive strategy for seronegative infants. A promising strategy involves deletion of the RNA synthesis regulatory protein M2-2, resulting in increased viral RNA gene transcription and antigen expression but decreased genome replication. Two candidate vaccines, MEDI/∆M2-2 and LID/∆M2-2, have originally been evaluated. Both induced strong RSV-neutralizing antibody responses; however, the LID/∆M2-2 vaccine has been considered more effective, as it confers a slight increase in replication [95,96]. Although LID/∆M2-2 was well tolerated in the Phase I study, the higher replication might make it poorly tolerated when administrated to a larger population. Other candidates containing the M2-2 protein mutations have been tested. LID/cp/∆M2-2, designed with the insertion of a set of five defined point mutations originally derived from serial cold passage, resulted in an over-attenuated vaccine that had low infectivity and low-titre antibodies in only a fraction of the participants, thus being not suitable for further development [97]. Another vaccine constructed by the addition of the genetically stabilized mutation 1030s, LID/∆M2-2/1030s, was investigated in a Phase I study [98]. This candidate was more attenuated than the parent vaccine with vaccine-induced titres of serum RSV-neutralizing antibodies, essentially equivalent to a primary RSV infection [98]. These characteristics make this candidate very attractive for further investigations. D46/NS2/N/∆M2-2-HindIII was built on the LID backbone but with several modifications that are expected to generate a phenotype similar to MEDI/∆M2-2. This vaccine has been demonstrated to have a greater attenuation than LID/∆M2-2, but considerably higher peak viral titres than MEDI/∆M2-2 [99]. The pre-fusion RSV protein is a structural protein that plays an important role as a natural immunogen. Stobart et al. identified a chimeric RSV strain with enhanced pre-fusion antigen levels, thermostability and immunogenicity, despite heavy attenuation in the airways of cotton rats generating a promising RSV-LAV candidate [100]. The results of this Phase I clinical trial have not yet been published. MEDI-559, a candidate that contains five attenuating mutations involving the nucleoprotein and fusion, the large polymerase M2-1 and M2-2 proteins, was a promising candidate. However, after administration, it appeared to be genetically unstable, leading to a partial loss of its phenotype: it was more biologically active and immunogenic in vitro than in vivo, and showed a higher risk of LRTI [101]. RSVcps2 represents the stabilized version of the MEDI-559 vaccine. In a Phase I clinical trial, RSVcps2 was well tolerated and moderately immunogenic [102]. Another promising strategy involves deletion of the non-structural (NS) proteins 1 and 2, which modulate the immune response to promote transforming growth factor-β-mediated cell cycle arrest and viral replication [103]. In fact, it has been seen that deletion of the NS2 gene diminished RSV replication in chimpanzees [104]. The increased interferon response to infection may enhance the adaptive immune response, as has been demonstrated in calves [105]. NS2 also functions as a pathogenicity factor, promoting epithelial cell shedding in vitro and in a hamster model, potentially contributing to small airway disease [106]. Karron et al. conducted a stepwise Phase I evaluation of RSV/∆NS2/∆1313/I1314L, demonstrating that it is attenuated yet immunogenic in RSV seronegative children [107]. Other RSV-LAVs that include NS2 or NS1 gene modification, such as RSV 6120/∆NS2/1030s and RSV 6120/∆NS1, are under investigation. The results have not yet been published [108][109][110][111]. LAV candidates are considered safe for clinical evaluation in children because these vaccines are not expected to cause ERD. RSV-LAVs are able to generate a robust immune response and, thanks to their safety and nature, can be administered also to infants. Another advantage is that this type of vaccine is administered intranasally, avoiding the use of needles. Even if most clinical trials are only in Phase I, RSV-LAVs represent promising candidates worth of further investigation.
The only chimeric vaccine candidate is rBCG-N-hRSV [112]. It consists of the bacillus Calmette-Guerin (BCG) vaccine expressing the nucleoprotein (N) of RSV. It is delivered via a BCG strain and induces a Th1 and a humoral response. It is the only LAV that combines protection against two respiratory pathogens, Mycobacterium tuberculosis and RSV. Furthermore, it could be safe for administration to newborn babies. Céspedes et al. demonstrated that this vaccine is safe, showing no side effects in mice. A Phase I clinical trial on adults has been conducted, even if the results have not been reported yet [113].

Monoclonal Antibodies
In parallel with the development of a vaccine, passive immunoprophylaxis has also been studied as an alternative approach to RSV prevention. Before the year 2000, studies had already shown that intravenous immunoglobulins against RSV prevent severe RSV respiratory disease, and in 1996, RespiGam (Massachusetts Public Health Biologic Laboratories, and MedImmune) was approved by the FDA for use in high-risk groups [116] ( Table 6). The new millennium has seen the development of humanized mAb directed against the F and G surface glycoproteins encoded by RSV. The first and most studied mAb is palivizumab (MEDI-493, Synagis, MedImmune, Inc., Gaithersburg, MD). It is a humanized IgG1 mAb that binds to the F-RSV surface glycoprotein and is currently the only mAb licensed for the RSV infection in infants and children, whose indications are now restricted to premature birth, chronic lung disease and hemodynamically significant congenital heart disease [17]. Despite its great effectiveness, palivizumab has important limitations, such as needing to be administered by intramuscular injection once a month during the RSV season and a high cost, which make this approach unfeasible for healthy infants. Consequently, more recent research has pointed to the development of antibodies with a higher potency or with an extended serum half-life [116,117]. A first attempt has been done with motavizumab (MEDI-524, MedImmune). This is an RSV-specific mAb developed by remodelling the heavy and light chains of palivizumab, so that an approximately 70-fold higher affinity for the F protein of RSV and 20-fold higher potency than palivizumab were obtained [118]. Many trials have been conducted in infants, comparing motavizumab to the placebo or to palivizumab, and they have shown that motavizumab has a pharmacokinetic profile similar to palivizumab [118]. Given the likelihood that both products could be available concurrently for commercial use, and that children could receive both agents during the same RSV season, a specific trial was conducted in which infants received, sequentially, motavizumab and palivizumab, demonstrating similar results in terms of efficacy and safety [119]. Motavizumab showed comparable results to palivizumab also when tested in 1236 children with hemodynamically significant congenital heart disease [120]. Nevertheless, in a very large Phase III, randomized, double-blind, palivizumab-controlled study enrolling a total of 6635 children, a large number of cutaneous adverse events was reported in the motavizumab cohort, and the FDA decided not to approve its license [116,121]. Eventually, the effect of motavizumab on RSV viral load has also been studied, with an apparent antiviral activity that could not be confirmed in subsequent trials [122,123]. Another tested product is suptavumab, also called REGN2222, an IgG1 monoclonal antibody directed against the RSV F glycoprotein, conceived for prophylaxis in infants who do not meet the criteria for palivizumab use. In 2015, a first-in-human study, on 132 healthy adults who received intravenously or intramuscularly REGN2222 compared to the placebo, demonstrated that suptavumab was well tolerated at all evaluated doses without serious adverse events or dose-limiting toxicities [124]. More recently, a Phase III trial enrolling preterm infants (<35 weeks and 6 days of gestational age) with less than 6 months of age was conducted. In August 2017, the Sponsor Agency announced that suptavumab did not meet its primary endpoint of preventing medically attended RSV infections in infants, so this mAb was discontinued [125]. A new possibility is represented by MEDI8897, also known as nirsevimab, a mAb that targets antigenic site ∅ on the pre-F conformation of RSV, and whose half-life has been increased three-fold with a triple amino acid substitutions, reaching the possibility of protection against RSV for an entire season with a single intramuscular injection [117]. Many randomized, double-blind, placebo-controlled clinical trials, conducted first in healthy adults and then in healthy preterm infants, have shown a favourable safety profile and a mean half-life of about 80-120 days for nirsevimab, so that it could be proposed as a once-per-RSV-season prophylactic agent [126,127]. In another trial, comparison between 969 infants who received nirsevimab and 484 infants who received placebo revealed a significantly lower RSV-LRTI incidence and hospitalization rate for MEDI-8897 vs. the placebo [19]. Given these promising data, a Phase III study is currently enrolling healthy late preterm and term infants, to determine the efficacy for MEDI-8897 in this population who would not be eligible to receive RSV prophylaxis [128]. Moreover, in July 2019, a Phase II/III randomized, double-blind, palivizumab-controlled study was started to evaluate the safety of MEDI-8897 compared to palivizumab in high-risk children [129]. Finally, nirsevimab will be tested in a Phase II, open-label, uncontrolled, single-dose study to evaluate its efficacy in immunocompromised Japanese children aged ≤2 years [130]. Lastly, MK-1654 is another studied mAb, binding to site IV of the F glycoprotein, with an extended half-life. A first double-blind, Phase I study involving 152 healthy adults has shown an apparent half-life of 70-85 days with a safety profile similar to the placebo [131,132]. Subsequently, a new trial was conducted, enrolling 80 healthy adults to determine if a single intravenous dose of MK-1654 might decrease the viral RSV load compared to the placebo. This Phase IIa double-blind, randomized, placebo-controlled study ended in March 2020, and the results are not available yet [133]. Another double-blind, randomized, placebo-controlled, single ascending dose study enrolling 180 healthy pre-term or full-term (29 weeks of gestational age or higher) infants receiving one of four dose levels of MK-1654 intramuscularly started in September 2018 and it is still recruiting [134].
The use of passive prophylaxis is an alternative to active vaccination. The success of palivizumab and the challenges related to the development of an effective RSV vaccine have spurred new research in this field. However, the high cost of passive prophylaxis is still limiting its use and calls for the development of a cost-effective vaccine.

Conclusions
Development of an effective vaccine to protect high-risk groups from severe RSV infections is of critical importance, but still challenging. Different antigens and vaccine formats should be considered for different target populations (children, the elderly and pregnant women). To date, the formats that are being evaluated are particle-based vaccines, vector-based vaccines, subunit vaccines and LAVs. Currently, the only vaccine that has reached Phase III clinical trials is a maternal RSV F nanoparticle vaccine, which showed efficacy in reducing hospitalization and RSV-LRTI in young infants, but did not meet the desired primary outcome, so future trials are needed to confirm its efficacy. The same vaccine with adjuvants was safe and immunogenic in the elderly, and future clinical trials will evaluate its efficacy. Only one vector-based vaccine has passed a Phase II clinical trial, and a Phase III trial in the elderly has been planned for initiation in 2021. Subunit vaccines are potentially good candidates for pregnant women and the elderly; induction of protective immunity and obtaining F protein conformation stability remain the major unsolved problems for creating an effective subunit vaccine. Most clinical trials for RSV-LAVs are only in Phase I, but these vaccines represent promising candidates worthy of further investigation as they are able to generate a robust immune response and, thanks to their safety and nature, can be administered also to infants. While waiting for commercially available safe and effective vaccines, immune prophylaxis in selected groups of high-risk populations is still mandatory. Monoclonal antibodies with a better cost-effectiveness ratio than palivizumab, such as nirsevimab or MK-1654, are the subjects of clinical trials. In young infants, combining the use of passive immunization via maternal vaccination or mAbs, followed by paediatric active immunization, may be effective to prevent severe RSV infection. Research on this topic is still of utmost importance.