How do you go about making a vaccine? We received this very pertinent question from a sci-fi author a few weeks ago, and thought it was worth a new short blog series! This month we cover the early R&D stages of vaccine discovery, while in the next blog, we’ll talk about the clinical stages of development, and the critical role of adjuvants. As always, stay tuned!

While the discovery of the first (and very successful!) vaccines preceded a more detailed knowledge of pathogen biology and immunity, today vaccine discovery would be inconceivable without either. Also, as vaccines are administered to healthy individuals and children, their safety profiles must be even higher than for conventional drugs, which are generally administered to individuals that are already ill. This requires detailed knowledge of at least three factors: the pathogen, the population affected, and the type of immune response required to control the pathogen. Ultimately, vaccines work by eliciting a suitable immune response, so it makes sense that this is the first aspect to consider.

1) Immunity

Upon the first encounter with a new pathogen, innate immunity – the first arm of the immune response – is activated. This is a potent and non-specific response to certain pathogen signatures (known as pathogen-associated molecular patterns, or PAMPs), e.g., double-stranded RNA that characterises the replication stage of RNA viruses. Innate immunity can rely on an arsenal of weapons to reduce viral replication, mostly orchestrated by the interferon response and by specialised cells, such as dendritic cells,  macrophages, or Natural Killer cells that sense the environment, destroy the pathogen, and present what remains after the attack to the more specialised cells that constitute the adaptive immunity. These cells, comprising B and T cells, are responsible for a pathogen-specific response based on selective recognition of one specific pathogen, which triggers the production of antibodies or a cytotoxic response, respectively. Having undergone the necessary training the first time around, during a second encounter with the same pathogen, the adaptive response will activate more rapidly and defeat infection in a highly selective manner, right from the start. This time, the host will barely notice. The goal of a vaccine is to mimic the first infection in a far more controlled manner but maintaining the ability to train the adaptive immunity as the real infection would. This not only implies a fine balance between safety and reactogenicity but also means that a vaccine will need to induce the type of immune response that the target pathogen is more sensitive to, whether this is antibody-mediated or T cell-mediated. Therefore, one of the first steps is to determine whether a vaccine will need to induce neutralising antibodies or a cytotoxic (T cell-mediated) response. These so-called “correlatives of protection” can be hard to determine, and they need to be studied both at a population level and, if available, in animal models. More recently we have also become aware of the crucial importance of many other parameters, including the role of specific subsets of T cells, as well as the magnitude of release of different groups of chemokines and cytokines. While modulating some of these factors in a vaccine formulation remains difficult, it is critical to understand the role they play during infection and in a vaccination regime, to make sure they don’t compromise vaccine effectiveness or exacerbate the disease. Once the type of immunity required to control infection has been determined, the next step is to decide which elements of the pathogen will be able to induce such a response. This is when pathogen biology becomes important.

2) The pathogen

The first vaccines were made by attenuated or inactivated pathogens. This makes a lot of sense: if infection with the live pathogen confers protection to survivors, it makes sense to let the immune response decide for itself how to handle the virus, and simply administer a form of the pathogen that is not going to be as dangerous as the circulating form. Viruses can be attenuated by serial passaging in a cell type (or animal) from a different species. In the attempt to evolve in this new host, the virus will lose some of its virulence in the original target species but will remain replication competent enough to alert and activate the immune response. Alternatively, viruses can be completely killed (inactivated) before administration. The immune response will see the full pathogen, but as this is unable to replicate, inactivate pathogens have also been associated with a weaker immune activation. The risks of both these strategies have been experienced during the course of history: batches of poorly inactivated vaccines can cause the same harm as the pathogen they are meant to cure, while attenuated vaccines that are still able to replicate can be harmful to immune-compromised individuals and could even spread through the population. Sometimes this is a good thing, but it can be difficult to control. Because of the need for increased safety, alternative methods have been developed over the years. Nowadays, selected viral genes can be isolated, and recombinantly expressed protein can be used as vaccine immunogens. The response induced by individual proteins tends to be weaker than when the protein is expressed in the context of a full replicating pathogen; however, this approach is considerably safer, and novel administration systems consisting of adjuvants (more about this in the second blog of this series), virus-like particles, or virus-based expression systems have all been shown to stimulate a broad response in a much more controlled and safer manner. Choosing to express a single protein (or even just a section of that protein) means that its antigenic properties and its ability to induce a sufficient amount of the correct immune response (e.g., sufficiently high titres of neutralizing antibodies) must have been established beforehand. This can be achieved experimentally by testing the immune response elicited in vivo, or in some cases, through computer programmes able to predict the immunogenic properties of a protein sequence or conformation. Affinity, binding kinetics, antigen exposure, and virus neutralization are all properties that need to be carefully quantified in vitro.

3) Epidemiology

In-depth knowledge of the pathogen is not limited to its molecular biology but needs to encompass the existence of different strains, serogroups, and possible mutants. Where the selected immunogen is not conserved, the most common or the most relevant strains have to be selected. The route of infection also has to be taken into consideration: a virus that infects via the respiratory route might require different immunity and, therefore, different administration than a virus that infects via the gut, or by entering the bloodstream. Other important criteria to consider are the geographical target area, as well as the demographic of the most affected populations. Is a specific group (children, pregnant women, or older individuals) more severely affected than others? Is the vaccine designed for populations in developing countries, or in areas which lack refrigeration, or that are difficult to reach? The earlier these factors are taken into consideration, the higher the chances that an effective vaccine would reach the clinical stages of development.

Preparing for the unexpected

So far, it looks like a lot of work, but fairly straightforward: by thoroughly investigating pathogen, immunity, and epidemiology, and harnessing all the newly developed methodologies, it looks like there may be a solution for every pathogen out there. Unfortunately, this is hardly the case. The first obstacle is that viruses mutate, and some of them mutate too fast to keep up with. We have become quite good at making vaccines against flu, but these vaccines still need to be made once per year. A vaccine against HIV is still elusive, specifically because of the very high mutation rate that would render a vaccine completely obsolete in no time. There are also viruses that play havoc with our immune response, or that have evolved to use it to their advantage. Non-neutralising concentrations of antibodies have been shown to facilitate dengue virus access to the very cells that support its replication, exacerbating infection in individuals exposed to the virus for the second time. Our still limited knowledge of immunity and especially of immune regulatory mechanisms also precludes our ability to exactly predict vaccine effectiveness, particularly in understudied age groups, populations, or immune-compromised patients. Even with these limitations, however, the field has made huge progress, ensuring that effective and safe vaccines are available against an impressive number of almost forgotten horrible diseases. Also, many more candidates against new and emerging viruses are at various stages of clinical trials. In the next blogs, we will explore the steps that take vaccines from bench to humans, and we will dig deeper into the unsung heroes of vaccinations: adjuvants.