The defining concept of Reverse Vaccine Development is the evaluation of efficacy early in clinical development with the goal of identifying a proof-of-concept correlate of protection. The exploration of the correlate must be based on efficacy and immunogenicity endpoints selected to have robust readouts. Further, when the strain variability of the pathogen is significant and the vaccine targets only some strains (e.g., targeting variable polysaccharide capsule or non-conserved protein antigens), microbiology endpoints are needed. Finally, for the study to be feasible, a relatively small, high attack-rate population must be available for study, which need not be the population intended for licensure. Importantly, a correlate of protection identified in this population (training set) needs to be confirmed in a holdout set or second matched population (test set).

While the study population can be a natural high attack-rate population, e.g., the patients suffering from S. aureus SSTI discussed above, a controlled human infection model (CHIM; aka, human challenge studies)12 can also be used as the study population. As compared to a naturally infected population, a CHIM has advantages and disadvantages. The advantages include the ability to control the population being infected regarding their naturally acquired immunity, co-infections, genetics, microbiome, nutrition, and environment, as well as the ability to control the timing, route, and/or dose of the infection and the infecting microorganism so that no disease is caused or the disease is self-limiting or can be controlled with cures or treatments13. The knowledge of the time of the infection allows a detailed characterization of the post-infection time course of the host immune response. The ability to control the pathogen dosage enables the study of pathogen dosage effects on the clinical and immune responses14. Importantly, the ability to collect pre-exposure samples provides a background to understand the correlate of protection, and whether it may differ between symptomatic primary infection and symptomatic re-infection. Because of these benefits, there has been an increase in calls for CHIMs in areas where study volunteers may have had prior exposure to the pathogen being studied and other pathogens15,16,17. Additionally, as compared to trials of vaccines in high attack-rate populations, CHIMs can be smaller, shorter, less expensive, and expose fewer participants to experimental vaccines13. Furthermore, CHIMs can be valuable for identifying lead vaccine candidates to test in larger studies, and thereby accelerate vaccine development to realize public health benefit sooner, strengthening the ethical rationale against the risk of harming participants18.

However, CHIMs have disadvantages, as previously described by Abo et al. 12 and summarized here: (1) The challenge strain may poorly represent naturally circulating pathogen strains, so the efficacy and correlate of protection of the vaccine may not extrapolate to field settings; (2) For safety reasons, challenge models use disease of low severity that might not apply to disease of greater severity in natural populations; (3) A high pathogen inoculum to achieve a high attack rate can overwhelm a vaccine’s protection, and low attack rates in placebo arms have affected efficacy rate determinations; and, (4) As compared to the target population, CHIM participants can differ in comorbidities, immunity, age, immunogenetics, and microbiome, which can influence a vaccine’s efficacy and correlate of protection. For a correlate of protection identified in a CHIM to be used to guide vaccine development, it needs to be validated in a field trial, typically a phase 2b study19. CHIMs that overcome these potential difficulties with participants, pathogens, and diseases, as with studies of natural high attack-rate subpopulations, offer the potential to de-risk and accelerate late-stage vaccine development.