Fail-Fast in Respiratory Syncytial Virus Vaccine Development

Despite of progress made for the past decades in the field of respiratory syncytial virus (RSV), we are still lacking a safe and effective RSV vaccine. In the absence of a correlate of protection to RSV, vaccine developers are working in the dark, and therefore often find out that their product is insufficiently effective (1). For example, Novavax developed a nanoparticle-based RSV vaccine for older adults and went through a large phase 2b trial, only to find out it was not effective in a large phase 3 trial (2). Novavax got back on its feet and is currently unblinding their next phase 3 trial, now in pregnant women (3). To accelerate RSV vaccine development, developers adopted the fail-fast approach.

Fail-fast systems were designed to immediate report failure to stop product development, rather than continue developing a product that likely will never be good enough. The fail-fast approach has been adopted by the pharmaceutical industry and became the mantra of many start-up companies not only to prevent wasting efforts, but also to create a healthy society in which entrepreneurs can fail, learn and improve (4). However, the fail-fast strategy has its own challenges, which are illustrated in this issue of the Journal by Ascough and colleagues (pp. 481–492) (5). In their article, they describe the results of a phase 1 study of a novel needle-free RSV vaccine: SynGEM. The vaccine is based on a stable prefusion F antigen of the virus and uses a bacterial-like particle as an immune-enhancing carrier. The vaccine had already been shown to protect against infection in mice and cotton rats on RSV challenge (6). Now, healthy adult volunteers were vaccinated using a prime-boost approach. The authors used conventional methods to measure induced antibodies, including nasal IgA levels, serum neutralization, and palivizumab competing assays. The vaccine was safe and induced a twofold rise in specific antibodies, including when participants were seropositive at the start of the trial. RSV-specific mucosal IgA concentrations against RSV were variable, with the strongest increase in individuals with low preexistent levels of mucosal antibodies. Furthermore, the vaccine induced an increase in concentration of circulating RSV-specific B cells determined by ELISpot. However, despite the use of a subunit vaccine based on the stabilized pre-F protein, there was no induction of neutralizing antibodies. Unfortunately, these results were not convincing enough to grant funding for a human challenge trial.

This study is important and well performed, given the challenges of doing this with limited resources of an early-phase clinical trial. The study sets an example that collaboration between a relatively small biotech company, such as Mucosis, and the highly experienced RSV research team at Imperial College has the potential of developing vaccines with worldwide impact. Unfortunately, the study did not reach the endpoint threshold, and SynGEM was withheld from proceeding to next-phase trials. Was this a rushed decision?

To answer this question, we should think about RSV vaccine development in general. The development of an RSV vaccine is hampered by several problems. First, there is no optimal animal model. Mice can be infected with human RSV, but they lack many of the clinical characteristics of RSV bronchiolitis in children. Cotton rats are often used to develop therapeutics but do not reliably predict efficacy of antivirals or vaccines against RSV. In addition, the formalin-inactivated RSV vaccine, which caused vaccine-augmented disease, led to great caution taking an RSV vaccine into seronegative infants (7). Most important, we lack serological markers of protection to RSV. There is evidence that suggests mucosal IgA is protective against RSV, and neutralizing antibodies against RSV-pre-F show high neutralizing activity (8, 9). Still, this does not guarantee that the vaccine protects against RSV infection. Until we have established correlates of protection to RSV, we are left with trial-and-error approaches, such as those used for SynGEM in this article. Considering the risk that vaccines might still fail during late-stage clinical trials, manufacturers are vexed on how to move forward with clinical development. This uncertainty may well have contributed to the premature ending of potentially safe and effective RSV vaccines, which would be detrimental to the development of a working vaccine. A possible way to negate these adverse effects, and a more secure option in the fail-fast approach, would be to use the human challenge model.

In the case of SynGEM, the decision to discontinue vaccine development resulted in discontinuation of funding and the bankruptcy of Mucosis. Could this public–private partnership have used a human challenge study to add value to the decision whether or not to continue the clinical development of SynGEM? The human challenge model has limitations, including the use of a single viral strain, absence of the target population (infants), and relatively mild disease severity. Nevertheless, human challenge studies can quickly provide proof of concept of efficacy of novel RSV therapeutics, as is acknowledged by the World Health Organization and regulators (10–13). The authors are part of the publicly funded prestigious HIC-Vac network, which is hosted at Imperial College, London, which can perform a human challenge with SynGEM in healthy volunteers. As Virtuvax has now taken over Mucosis’ vaccine technology, this may still be a viable option to get the answer to the critical question whether or not immunity by SynGEM could be protective.

In conclusion, the authors have provided compelling evidence that bacterial-like particle–based vaccines may have a future, in particular for RSV infection. However, the conclusions of this study leave us with a dilemma. On the one hand, the vaccine may deserve a second chance in a human challenge study to define clinical protection. On the other, after the fail-fast culture, we should be bold enough to terminate a program without hesitation, as this allows us to move on and develop an even better program to fight one of the most deadly diseases during infancy.