De-risking vaccine development: lessons, challenges, and prospects

Estimates for the decade leading up to 2030 show that over approximately 50 million deaths could be averted through vaccinations against infectious diseases¹. Globally, the demand for vaccines is growing due to several key drivers. First, new vaccines are needed to counter the increasing prevalence of potentially pandemic viral pathogens, as highlighted by the recent SARS-CoV-2 pandemic or the current m-pox outbreaks. This situation is compounded by the continuing emergence of antimicrobial resistance (AMR) which has been linked to 5 million deaths globally in 2019². This unprecedented emergence led the US Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO) to formulate a list of first-line targets mostly involved in nosocomial infections, the ‘ESKAPE’ pathogens—Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species³. Effective vaccines are also needed against residual unmet needs such as malaria, tuberculosis, and HIV/AIDS, and against pathogens associated with high impact and case fatality rates but with a locally restricted incidence, such as Marburg virus. Second, the rising vaccine demand is driven by global population growth and demographic shifts. Particularly in high- and middle-income countries (HICs and MICs, respectively) there is a steady increase in the population of older adults (OAs), which will make up nearly a quarter of the global population by 2050. Indeed, while the incidence and severity of infectious disease symptoms increases with age, only few vaccines are currently specifically targeted at the OA population⁴. In parallel, pediatric vaccine demands continue to grow, driven in part by the demographics of African countries. This has been a driver for GAVI’s aim to vaccinate a further 300 million children between 2021 and 2025⁵. Third, across the age spectrum there is a growing awareness of the value of preventive healthcare and meeting the varying vaccine needs across a person’s life-span⁶. The mindset shift toward an increased focus on adult vaccination is illustrated by the WHO’s decision to make global life-course vaccination a strategic priority in their Immunization Agenda 2030, aiming to mirror the successful delivery of pediatric vaccines⁷. However, achieving optimal vaccine coverage is increasingly impeded by vaccine hesitancy, which has been listed in 2019 among the WHO’s top 10 global health threats^8,9.

The health economic (HE) gain of vaccines will be mostly felt in low- and middle-income countries (LMICs), due to their higher disease burden and less developed medical infrastructures. Indeed, with a return of investment of ~$20 per $1 spent on vaccination against 10 diseases prevalent in LMICs—which can be more than doubled when broader HE and social benefits are considered¹⁰— the impact of global vaccination cannot be underestimated. Yet, the standard approach to prophylactic vaccine development struggles to address the medical needs, with as root causes the long timelines from discovery/preclinical phases to licensure, spanning up to 15 years, and exceedingly high expenditures, with Phase 3 development accounting for ~70% of the total costs^11,12,13 (Fig. 1). As R&D productivity is a function of the development time and costs for both successful and unsuccessful candidates, the price tag for developing one vaccine, adjusted for the costs of those failing in R&D, can amount to roughly $1 billion^11,12,13. Throughout the process there is a high risk of failure, due to limitations posed by immunology/epidemiology (owing to specificities of the disease, target population, or availability of validated correlates of protection [CoPs] or of reactogenicity/safety biomarkers), technology (platforms and tools, production scale-up), execution, market (size, competitive landscape, strategic fit), regulatory stringency, and funding¹². In addition, prelicensure clinical trials can detect the more common adverse events but can be limited in their capacity to detect certain rare adverse events, which may lead to a vaccine’s retraction from the market^14,15. All in all, this state of affairs amounts to a slim (~10%¹²) probability of market entry for any candidate reaching clinical stages. While the number of new pharmaceuticals remained stable over the last three decades, the associated R&D costs have multiplied and became for many manufacturers increasingly difficult to afford. As a consequence, the number of investing vaccine manufacturers and overall investments have dwindled¹⁶. This imbalanced situation resulted in a mismatch between the investments and expected turn-over of a product (the ‘productivity gap’¹²).

Marking an abrupt break with the traditional development norms, the SARS-CoV-2 pandemic represented a cornerstone in vaccine R&D and sparked new opportunities for innovation. This watershed is underscored by the COVID-19 vaccines’ extraordinarily fast (9 months) progress from Phase 1 development start through regulatory approval, or the 3-month strain-switch development recently seen for certain mRNA COVID-19 vaccines^17,18,19. The high pace was underpinned by the efficiency and scalability of nucleic-acid-based platforms (causing a surge in particularly mRNA vaccines), seamless clinical trial phases supported by real-world evidence, strong regulatory prioritization, and, importantly, the scientific/technological collaborations fostered by effective public-private partnerships (PPPs)¹⁷. Unfortunately, the post-pandemic drive for innovation has been manifested unevenly across disease areas and vaccine archetypes²⁰. Indeed, while cutting-edge innovation aided the development of vaccines against seasonal diseases such as COVID-19, medium-level innovation was observed for persisting threats or potential outbreaks, such as respiratory syncytial virus (RSV) and chikungunya virus (CHIKV), respectively, while innovation lagged behind for neglected tropical diseases (NTDs)—for which very few approved vaccines exist despite incentivizing initiatives²¹—and nosocomial pathogens. This situation has created a global vaccine inequity and profound public health challenges.

Besides public funding, de-risking and accelerating vaccine development remains crucial to close the productivity gap and address these disparities. By taking effective early-stage decisions and, if needed, by swiftly terminating the likely unsuccessful projects, fewer but more promising programs advance to the expensive later stages such that the proof-of-concept (PoC) is reached faster and at lower cost. This ‘quick-win, fast-fail’ paradigm requires switching a success-focused mindset toward a stronger focus on decreasing technical uncertainty before progressing to late-stage development²². Here, we illustrate on the basis of real-world examples how frontloading the risk in early development stages (preclinical and early clinical phases) can be leveraged to produce more innovative, cost-effective candidates with increased market potential. Moreover, by controlling R&D costs, these early de-risking strategies can incentivize an increased focus on the more challenging vaccine targets, to ultimately attain a more equitable and sustainable global access to life-saving vaccines.

Apart from clinical safety, which is beyond the scope of this article, clinical efficacy remains the first-choice endpoint to demonstrate vaccine performance and support licensure. However, demonstrating vaccine efficacy is in many cases unfeasible. This can be due to either a low or unpredictable disease incidence, an uncertain epidemiology (e.g. for Zika, Chikungunya, or mpox), or to the unattainability of adequate trial sample sizes as a result of already existing vaccines or standards of care. In these cases, an accepted CoP (an immune marker that mechanistically predicts protection against a specified clinical disease endpoint) can provide an alternative pathway to licensure. A CoP can also de-risk and reduce the size of late-stage clinical studies by supporting dose/regimen selections, pivotal data generation, and immuno-bridging between populations²³. CoPs can be identified by analyzing immune response data from protected versus susceptible participants in effectiveness or immuno-epidemiological studies using a standardized and validated assay^24,25,26, or through extrapolation from human or animal challenge studies^27,28,29.

For example, in the case of meningococcal vaccine development, initial evidence came from seroepidemiological studies conducted in the United States³⁰. These studies showed that the proportion of individuals with serum bactericidal activity (by serum bactericidal antibody [SBA] assay with exogenous complement) to meningococci of serogroups A, B, and C (MenA, MenB and MenC, respectively) was inversely related to disease incidence. Furthermore, during an outbreak of MenC disease in military recruits, the vast majority of cases occurred in individuals with low SBA titers against MenC (i.e. <1:4)³⁰. Based on this evidence, this serological CoP served as basis for licensure of MenC conjugate vaccines, in the absence of efficacy data³⁰. Following widespread use, clear SBA-based CoPs have been derived and the threshold extended to all meningococcal vaccines, including protein-based MenB vaccine^31,32. The SBA assay was used throughout the entire vaccine development cycles for both polysaccharide conjugate vaccines targeting serogroups A, C, W, Y, and X, and protein-based MenB vaccines: in the preclinical screening of candidates for advancement to clinical studies, and in the clinical phases, in turn, to select the optimal formulation, generate an early PoC, and (when complemented with Phase 3 effectiveness data), to serve as basis for licensure³³.

Furthermore, for pneumococcal vaccine development, an aggregate CoP (a serotype-specific IgG level of 0.20 μg/mL²³) was established from a vaccine efficacy (VE) study of a heptavalent CRM₁₉₇ pneumococcal conjugate vaccine (PCV7) in infants, and used as basis for licensure of the 13-valent vaccine in the US. Based on trials in different settings and populations, the CoP for infants was then updated to 0.35 μg/mL²³. This consensus threshold was incorporated into the WHO guidance for licensure of higher-valency pneumococcal vaccines, which is informed by a candidate’s head-to-head performance versus PCV7.

Monoclonal antibodies have been used as substitutes of mechanistic CoPs for certain viral vaccines, e.g. the first approved maternal RSV vaccine, a prefusion F (PreF) antigen-based vaccine; reviewed in refs. ^23,34. A trial in at-risk infants, evaluating an RSV fusion (F) protein-targeting monoclonal antibody (palivizumab), defined a neutralizing antibody (nAb) titer that was associated with protecting the infants against severe disease. This threshold was then used to assess maternal vaccine responses in a Phase 1 study evaluating the abovementioned PreF-based candidate in pregnant women, followed by modeling to identify the fold-rise in maternal nAb titers needed to confer protection to their infants. PoC was achieved when a post-hoc analysis of a Phase 2b study testing the unadjuvanted candidate revealed high-level ( ~ 90%) VE against severe RSV-associated lower respiratory tract disease, consistent with the nAb titers exceeding the palivizumab threshold in infants³⁵. In 2023, the maternal vaccine was granted approval and market authorization from the FDA and EMA, respectively³⁴.

Unfortunately, very few vaccines can rely on an accepted CoP, and the currently known CoPs are predominantly serological benchmarks, thus disregarding any contributions from mucosal and/or cell-mediated immunity (CMI)^23,29. Compared to CoPs, biomarkers encompass a wider range of indicators including immune responses present pre- or post-vaccination, biological processes or disease states, and can be based on laboratory tests for a pre-specified target response such as CMI, or on mechanistic links to protection or reactogenicity^23,36,37. For example, several biomarkers have linked innate signatures or the presence of a specific B cell subset at baseline, with reactogenicity^38,39,40.

While having a weaker statistical basis than CoPs, biomarkers may enable de-risking by supporting PoC generation in early clinical phases and allowing efficient connections of preclinical and clinical data³⁷. A case in point is the development of a live-attenuated vaccine against Chikungunya^41,42 which has an unpredictable epidemiology with disease cases and outbreaks affecting over 100 countries. The combination of human seroepidemiological data with data from non-human primate (NHP) challenge studies allowed defining a surrogate which, in turn, propelled the vaccine’s clinical development up to Phase 3 studies and licensure (Fig. 2)^43,44,45.

Particularly when biomarkers are supported by mechanistic preclinical data, they can support a rapid advancement to Phase 1/2 studies, when de-risking is still a relatively quick ‘win’. Then, the marker may serve to validate antigens/adjuvants and platforms, and support early human PoC generation. However, broader use of biomarker or CoP data still faces several roadblocks, such as restrictive guidance on the level of evidence required for (regulatory) decision-making, and insufficient standardization of data-collection tools and assays²³. Especially the latter issue has become more pressing given the increased use of ‘omics’ technologies and multi-parameter assessments for biomarker identification^23,46,47.

CHIMs represent another vital de-risking tool for obtaining an immunological association with protection for a given read-out²⁷. Indeed, a CHIM study was instrumental in securing the FDA license for the inactivated oral cholera vaccine⁴⁸. CHIM data also enabled the early clinical de-risking of another bacterial vaccine, namely the Generalized Modules for Membrane Antigens (GMMA)-based vaccine against shigellosis, a leading bacterial cause of diarrhea in young children (Fig. 3, left). A monovalent S. sonnei vaccine candidate based on engineered outer membrane vesicles exposing Shigella’s O-antigen (O-Ag) was tested in a Phase 2b CHIM study⁴⁹. Although the candidate failed to demonstrate clinical efficacy in that study, the CHIM data indicated a link between anti-lipopolysaccharide antibody levels and protection against infection⁵⁰. These results prompted a return to the design stage, and suggested that a higher amount of O-Ag per vaccine dose may be needed to achieve adequate protection. Subsequent development focused on a new GMMA candidate harboring a ten-fold higher amount of S. sonnei O-Ag (relative to protein and lipid A). The antigen was included in a four-component vaccine also containing GMMA from three S. flexneri serotypes (1b, 2a and 3a), all of which are highly prevalent in LMICs^51,52. Preclinical results indicated that the humoral immunogenicity of the OAg-enriched candidate was significantly higher as compared to the original monovalent vaccine⁵², and this result was used as PoC to progress the new candidate into clinical development⁵³.

The development of RSV vaccines for OA represents another example of the importance of CHIM studies for de-risking late-stage clinical evaluations (Fig. 3, right). Results from a failed Phase 3 study combined with data from CHIM trials provided preliminary clinical evidence and a certain level of confidence in the optimal antigen conformation and localization of neutralizing epitopes^54,55,56,57. This data facilitated the decision-making process on at-risk investments during Phase 2, and enhanced the probability of success of pivotal Phase 3 trials^34,58,59,60.

Of note, CHIM studies have not proven equally useful across pathogens and populations^61,62. For example, the RSV CHIM studies in young adults have been useful for the PoC demonstrations of the vaccines for OA, but the data were not fully generalizable to the OA target population (or, for that matter, to the pediatric target population for RSV vaccines). For example, when comparing the RSV CHIMS in young adults with the RSV trials in OA, differences existed in the main study endpoints (i.e., upper vs. lower respiratory tract infections, respectively), and in the level of immunocompetence of the respective study populations. Moreover, CHIM data generated in HICs may not be generalizable to LMICs due to variations in nutrition, genetics, and co-morbidities^61,63. Thus, the selection of an ad-hoc trial population can have far-reaching results for development timelines, as detailed below.

Especially for low-incidence diseases, Phase 3 studies can be hampered and made more complex by the large sample size needed for statistical VE demonstration in the general population, in particular when VE is the only possible endpoint. An opportunity to generate an early-stage PoC is to seek protection in an at-risk setting or in populations of at-risk individuals (‘enriched’ populations), and then extend the findings to a more general population. A typical example is the development of a vaccine against Staphylococcus aureus. This ‘ESKAPE’ pathogen³ is associated with skin and soft tissue infections and invasive/pulmonary disease in healthcare, community and military settings, and is globally the second leading pathogen for AMR-associated deaths². Despite several attempts, all Staphylococcus aureus vaccine programs failed to show efficacy in large Phase 3 trials, despite promising immunogenicity results in preclinical and early clinical stages. This is likely caused by the lack of reliable animal models, the pathogen’s multiple immune evasion strategies, and a still unclear mechanism of protection. The inability to achieve early clinical PoC and the need for substantial investments to fund large efficacy studies ultimately resulted in a disinvestment by manufacturers in new Staphylococcus aureus vaccine programs. In this context, ‘reverse vaccine development’ represents a new paradigm⁶⁴ in which the typical order of studies is reversed, by seeking the demonstration of efficacy (PoC) early in the process instead of in Phase 3, as well as in a high attack-rate population which may differ from the actual vaccine target population. This approach was applied to the Phase 2 evaluation of a five-component candidate Staphylococcus aureus vaccine performed in an ad hoc population of adults with Staphylococcus aureus-associated ongoing skin and soft tissue infections (NCT04420221). Though ultimately unsuccessful, the clinical development of this candidate has been lean ( ~ 600 subjects in total), reasonably inexpensive, and drastically shortened compared to classical development times. The data from the enriched population thus allowed the program’s advancement from preclinical stages to clinical futility (i.e., demonstrated to be unlikely to meet its goal of demonstrating VE) within only 4 years^64,65. Remarkably, this condensed timeline was achieved despite the concurrent COVID-19 pandemic.

Another striking example is the Phase 3 trial of a COVID-19 vaccine in Brazil, conducted for the first time in a high-risk population of healthcare professionals taking care of COVID-19 patients^66,67. In this case-driven trial, the expected force of infection was significantly higher in the enriched population than in the general population. This allowed using a smaller sample size for the pivotal Phase 3 study of the vaccine, which subsequently became part of the first wave of COVID-19 vaccines approved for the WHO Emergency Use listing. A risk of following this approach is that a very high force of infection can translate into a lower efficacy rate in the enriched versus the general population^68,69. The origin of VE variations in different settings may be of a multifactorial nature, including contributions from coinfections, nutritional status, and/or social/economic factors, amongst others⁶⁹.

Besides a short-tracked manufacturing phase and maximized regulatory prioritization, early pandemic vaccine development was also streamlined by adaptive trial designs and smart trial protocols. Master protocols such as the WHO’s SOLIDARITY protocol allowed evaluating several candidates in diverse populations simultaneously, and included shared control arms⁷⁰. In such adaptive designs, disease rates in a candidate’s arm were compared with concurrently randomized arms of other candidates or controls. This informed a vaccine’s addition or removal from the trial, and resulted in faster advancement of the lead candidates. These protocols also adopted a seamless sequence of partially overlapping Phase 1, 2, and 3 trials¹⁷, if possible following a Bayesian design, as applied in a cross-industry collaboration⁷¹. Thus, while no shortcuts were taken in the clinical research of COVID-19 vaccines and trial sizes were not reduced¹⁷, these designs allowed the progression of VE and safety evaluations in a faster, more ethical manner, and in more participants as compared to traditional trials.

The increased use of interim data as a basis for regulatory review was another game-changer for pandemic vaccine R&D strategies. Indeed, in the decade preceding the pandemic, clinical development timelines for an FDA-approved vaccine amounted to 8 years, spread across seven trials of which at least two were pivotal VE trials⁷². While this could potentially be shortened in cases of a fast-track designation or rolling data submissions for expedited approval (e.g., FDA’s Emergency Use Authorization [EUA] or EMA’s Conditional Marketing Authorization [CMA]) as was applied to A/H1N1 pandemic vaccines, the more drastic accelerations did not occur until the pandemic. Then, regulators facilitated timelines of ~3 months for BLA approval, or 3 weeks for EUA/CMA applications based on interim Phase 3 results^73,74,75,76, which were also used for the approval of real-world-evidence and Phase 4 studies. These achievements highlight the merit of adopting a strategic approach with a strong focus on early regulatory contacts.

Across all phases, artificial intelligence/machine learning (AI/ML) combined with high-dimensional data generation is increasingly recognized as the new frontier for fast-tracking and de-risking vaccine R&D^77,78,79 (Fig. 4). This particularly applies when AI/ML is supported by innovative statistical approaches to optimize early and interim analyses of vaccine R&D data. Indeed, Bayesian statistics (which inherently incorporate uncertainty) and predictive modeling can significantly enhance such data analyses by providing a robust framework for managing uncertainty and integrating diverse information sources. These approaches allow incorporating prior knowledge—e.g. insights from previous studies, expert opinions, or biological data—into the analysis, and such historical data, underpinned by a deep immunological understanding, can then help guide the analyses of new data. Predictive models, especially those built using ML techniques, can forecast outcomes based on early and interim data. Especially Bayesian predictive models are powerful, as they quantify uncertainty in predictions, enabling more informed decisions on whether to continue, modify, or halt a trial. By deeply informing the decision-making process, these methods can lead to faster and safer vaccine development.

Finally, FDA’s ‘animal rule’ represents a last resort for pathogens for which testing on humans is impractical or unethical, as it allows extrapolating human efficacy solely from well-characterized animal efficacy data. Currently this approach has only been adopted for licensed anthrax vaccines for post-exposure prophylaxis.

To fulfill the growing demand for next-generation vaccines, including those against emerging pathogens and new indications, a multipronged de-risking approach is needed that can shift vaccine development towards a more streamlined and lower-risk process. While all development phases benefit from expanded use of digital technologies, particularly the clinical development stages can be de-risked by enhanced biomarker and CoP definitions, smart study protocols with streamlined (if possible CHIM-based) designs, and, if feasible, early alignment with regulators. The latter ensures that early on, integrated evidence plans are fit for purpose, and can support the vaccine’s licensure pathway as well as any subsequent recommendation stages. If a classical approach is clinically unfeasible, allocating the necessary resources in industry to address an otherwise unmet medical need will necessitate early assurance of the vaccine’s eligibility for the appropriate regulatory pathways, including guidance on how to navigate the prequalification. Such pathways can include FDA’s Accelerated Approval and LPAD (‘Limited Population Pathway for Antibacterial and Antifungal Drugs’) processes, which can be further streamlined by employing biomarkers, alternative correlates beyond the classical mechanistic CoPs, and/or simplified clinical development plans. If possible, another key strategy is improving manufacturability by selecting flexible platforms/technologies in the design phase, while applying the learnings from the pandemic era in relation to manufacture consistency across batches and sites in CMC processes^17,80. Not only does this heighten pandemic preparedness, it will also facilitate market authorization and national immunization program (NIP) inclusion, as manufacturing efficiency, antigen adaptability, and the potential for combination vaccines are among the core criteria in regulatory decision-making⁸¹.

Deploying sustainable manufacturing capacity and delivery infrastructures in LMICs is also critical for tackling the vaccine inequity persisting across communities, sexes, and countries. A case in point is the global COVID-19 vaccine inequity. Indeed, only 60% and 20% of the number of vaccine doses given per 100 people in HICs (with all doses counted individually) were administered in LMICs or LICs, respectively (Aug 2024 data⁸²), and LMICs were also the most likely to be affected by pandemic-related cancellations of other vaccinations⁸³. While equitable vaccine access is increasingly considered in decision-making by HICs and supranational organizations^84,85 (e.g. WHO’s Health Equity Assessment Toolkit), stronger commitment is required globally to ensure it is consistently used as criterion in regulatory reviews. These developments call for enhanced building of manufacturing and regulatory capacity, to ensure that clinical trials are globally authorized and conducted in a timely and equitable manner⁸⁶. Currently, several international regulatory collaborations are dedicated to addressing this need⁸⁷, including the African Vaccine Regulatory Forum and the International Coalition of Medicines Regulatory Authorities, for example.

Creating more incentives to develop vaccines with a low commercial value will also enhance vaccine equity, by allowing industry to better balance their portfolios between such high-risk projects and expected ‘blockbusters’ that sustain their R&D budgets. These budgets can also be nurtured by faster funding of the industry through public procurement/reimbursement (which can take 6 years from regulatory recommendation, or longer, in the EU⁸⁸), to be accomplished by earlier negotiations across all NIP stakeholders including the industry. Overall, this situation calls for greater alignment and more collaboration across borders. This can be leveraged by strong PPPs, which have historically been instrumental in addressing two key bottlenecks for the industry: funding, and scientific (academic) support e.g. for experimental medicine studies exploring disease mechanisms. PPPs have supported critical higher-risk vaccine projects already in the pre-pandemic era—e.g. the world’s first malaria vaccine (RTS,S/AS01), the M72/AS01 tuberculosis vaccine, and Ebola and CHIKV vaccines—and the COVID-19 pandemic has propelled many new collaborations between industry and non-industry partners (academics, governments, and supranational organizations). A prime example is ‘Operation Warp Speed’, which provided a paradigm for efficient PPP collaboration. Other COVID-19 vaccine-related initiatives include the Beyond COVID-19 Monitoring Excellence (BeCOME) or the Safety Platform for Emergency Vaccines (SPEAC; Coalition for Epidemic Preparedness Innovations–Brighton Collaboration), focusing on post-marketing monitoring or standardization of safety reporting, respectively^89,90. As the returns on investment for society can also be substantial for more challenging projects⁹¹, fostering globally-funded PPPs remains imperative to sustain the momentum for increased innovation after the pandemic. There is an urgent need to robustly fund and optimize adult-targeting NIPs to establish global life-course vaccination⁹², as adult immunization programs are considered highly effective investments for governments and healthcare providers, with returns to society of up to 19 times their investment⁸⁵. This holds true in the post-COVID-19 era, as the investments in the associated infrastructures made during the pandemic did not result in inclusion of adult vaccinations in routine immunization schedules⁸⁵, and vaccination coverage among adults remains mostly inadequate⁹³.

Crucially, the urgent need to develop new vaccines should be balanced with ensuring their safety. According to the stringent protocols enforced by regulatory agencies, careful surveillance and evaluation of potential vaccine-related adverse events are maintained throughout the vaccine’s development phases, as well as post licensure. This is supported by ongoing research efforts to understand the mechanisms of action underlying CoPs as well as safety biomarkers, though such benchmarks are unlikely to replace actual pharmacovigilance studies. All such efforts call for ongoing transparency, e.g. by publishing safety testing protocols and evaluations, to maintain public confidence in vaccines and address vaccine hesitancy.

To conclude, the key elements of the ‘toolkit’ for expedited and de-risked vaccine R&D include the optimized identification and use of CoPs, biomarkers, CHIMs and AI/ML technologies, and robust support from PPPs and cross-industry collaborations. Collectively, these measures will significantly contribute to the global efforts to close the R&D productivity gap, in the path towards a more equitable global vaccine access and improved public health.

1. Carter, A. et al. Modeling the impact of vaccination for the immunization Agenda 2030: deaths averted due to vaccination against 14 pathogens in 194 countries from 2021 to 2030. Vaccine 42, S28–S37 (2024).

   CAS  Google Scholar 

2. Antimicrobial Resistance Collaborators. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet 399, 629–655 (2022).

   Google Scholar 

3. De Oliveira, D. M. P. et al. Antimicrobial resistance in ESKAPE pathogens. Clin. Microbiol. Rev. 33, e00181–00119 (2020).

   Google Scholar 

4. Pollard, A. J. & Bijker, E. M. A guide to vaccinology: from basic principles to new developments. Nat. Rev. Immunol. 21, 83–100 (2021).

   CAS  Google Scholar 

5. Gavi—the Vaccine Alliance. Global immunisation in 2023: 7 things you need to know. https://www.gavi.org/vaccineswork/global-immunisation-2023-7-things-you-need-know (2024).

6. Jones, C. H., Jenkins, M. P., Adam Williams, B., Welch, V. L. & True, J. M. Exploring the future adult vaccine landscape-crowded schedules and new dynamics. NPJ Vaccines 9, 27 (2024).

   Google Scholar 

7. World Health Organization. Immunization agenda 2030: a global strategy to leave no one behind. https://www.who.int/teams/immunization-vaccines-and-biologicals/strategies/ia2030 (2020).

8. Acharya, S., Aechtner, T., Dhir, S. & Venaik, S. Vaccine hesitancy: a structured review from a behavioral perspective (2015–2022). Psychol. Health Med. 30, 119–147 (2025).

   Google Scholar 

9. World Health Organization. Ten Threats to Global Health in 2019. https://www.who.int/news-room/spotlight/ten-threats-to-global-health-in-2019 (2019).

10. Sim, S. Y., Watts, E., Constenla, D., Brenzel, L. & Patenaude, B. N. Return on investment from immunization against 10 pathogens in 94 low- and middle-income countries, 2011-30. Health Aff. 39, 1343–1353 (2020).

    Google Scholar 

11. Han, S. Clinical vaccine development. Clin. Exp. Vaccin. Res. 4, 46–53 (2015).

    Google Scholar 

12. Pronker, E. S., Weenen, T. C., Commandeur, H., Claassen, E. H. & Osterhaus, A. D. Risk in vaccine research and development quantified. PLoS ONE 8, e57755 (2013).

    CAS  Google Scholar 

13. Gouglas, D. et al. Estimating the cost of vaccine development against epidemic infectious diseases: a cost minimisation study. Lancet Glob. Health 6, e1386–e1396 (2018).

    Google Scholar 

14. Salmon, D. A., Orenstein, W. A., Plotkin, S. A. & Chen, R. T. Funding postauthorization vaccine-safety science. N. Engl. J. Med. 391, 102–105 (2024).

    Google Scholar 

15. Knipe, D. M., Levy, O., Fitzgerald, K. A. & Mühlberger, E. Ensuring vaccine safety. Science 370, 1274–1275 (2020).

    CAS  Google Scholar 

16. World Health Organization. WHO releases first data on global vaccine market since COVID-19. https://www.who.int/news/item/09-11-2022-who-releases-first-data-on-global-vaccine-market-since-covid-19 (2022).

17. Kuter, B. J., Offit, P. A. & Poland, G. A. The development of COVID-19 vaccines in the United States: why and how so fast? Vaccine 39, 2491–2495 (2021).

18. Excler, J. L., Saville, M., Berkley, S. & Kim, J. H. Vaccine development for emerging infectious diseases. Nat. Med. 27, 591–600 (2021).

    CAS  Google Scholar 

19. U. S. Food and Drug Administration. Updated COVID-19 Vaccines for Use in the United States Beginning in Fall 2024. https://www.fda.gov/vaccines-blood-biologics/updated-covid-19-vaccines-use-united-states-beginning-fall-2024 (2024).

20. Sabow, A. et al. Beyond the pandemic: the next chapter of innovation in vaccines. McKinsey & Company. https://www.mckinsey.com/industries/life-sciences/our-insights/beyond-the-pandemic-the-next-chapter-of-innovation-in-vaccines (2024).

21. Mukherjee, S. The United States Food and Drug Administration (FDA) regulatory response to combat neglected tropical diseases (NTDs): a review. PLoS Negl. Trop. Dis. 17, e0011010 (2023).

    Google Scholar 

22. Owens, P. K. et al. A decade of innovation in pharmaceutical R&D: the Chorus model. Nat. Rev. Drug Discov. 14, 17–28 (2015).

    CAS  Google Scholar 

23. King, D. F. et al. Realising the potential of correlates of protection for vaccine development, licensure and use: short summary. NPJ Vaccines 9, 82 (2024).

    Google Scholar 

24. White, C. J. et al. Modified cases of chickenpox after varicella vaccination: correlation of protection with antibody response. Pediatr. Infect. Dis. 11, 19–22 (1992).

    CAS  Google Scholar 

25. Gilbert, P. B. et al. Immune correlates analysis of the mRNA-1273 COVID-19 vaccine efficacy clinical trial. Science 375, 43–50 (2022).

    CAS  Google Scholar 

26. The FUTURE II Study Group. Quadrivalent Vaccine against Human Papillomavirus to Prevent High-Grade Cervical Lesions. NEJM 356, 1915–1927 (2007).

    Google Scholar 

27. Laurens, M. B. Controlled human infection studies accelerate vaccine development. J. Infect. Dis. 231, 1112–1116 (2025).

28. Little, S. F. et al. Defining a serological correlate of protection in rabbits for a recombinant anthrax vaccine. Vaccine 22, 422–430 (2004).

    CAS  Google Scholar 

29. Plotkin, S. A. Recent updates on correlates of vaccine-induced protection. Front. Immunol. 13, 1081107 (2022).

    CAS  Google Scholar 

30. Goldschneider, I., Gotschlich, E. C. & Artenstein, M. S. Human immunity to the meningococcus. I. The role of humoral antibodies. J. Exp. Med. 129, 1307–1326 (1969).

    CAS  Google Scholar 

31. Andrews, N., Borrow, R. & Miller, E. Validation of serological correlate of protection for meningococcal C conjugate vaccine by using efficacy estimates from postlicensure surveillance in England. Clin. Diagn. Lab. Immunol. 10, 780–786 (2003).

    Google Scholar 

32. Borrow, R. et al. Methods to evaluate the performance of a multicomponent meningococcal serogroup B vaccine. mSphere 10, e0089824 (2025).

    Google Scholar 

33. Pizza, M., Bekkat-Berkani, R. & Rappuoli, R. Vaccines against meningococcal diseases. Microorganisms 8, 1521 (2020).

    CAS  Google Scholar 

34. Ruckwardt, T. J. The road to approved vaccines for respiratory syncytial virus. NPJ Vaccines 8, 138 (2023).

    Google Scholar 

35. Simões, E. A. F. et al. Prefusion F protein-based respiratory syncytial virus immunization in pregnancy. N. Engl. J. Med. 386, 1615–1626 (2022).

    Google Scholar 

36. Weiner, J. et al. Characterization of potential biomarkers of reactogenicity of licensed antiviral vaccines: randomized controlled clinical trials conducted by the BIOVACSAFE consortium. Sci. Rep. 9, 20362 (2019).

    CAS  Google Scholar 

37. Van Tilbeurgh, M. et al. Predictive markers of immunogenicity and efficacy for human vaccines. Vaccines 9, 579 (2021).

    Google Scholar 

38. Pullen, R. H. et al. A predictive model of vaccine reactogenicity using data from an in vitro human innate immunity assay system. J. Immunol. 212, 904–916 (2024).

    CAS  Google Scholar 

39. Burny, W. et al. Inflammatory parameters associated with systemic reactogenicity following vaccination with adjuvanted hepatitis B vaccines in humans. Vaccine 37, 2004–2015 (2019).

    CAS  Google Scholar 

40. Sobolev, O. et al. Adjuvanted influenza-H1N1 vaccination reveals lymphoid signatures of age-dependent early responses and of clinical adverse events. Nat. Immunol. 17, 204–213 (2016).

    CAS  Google Scholar 

41. Cherian, N. et al. Strategic considerations on developing a CHIKV vaccine and ensuring equitable access for countries in need. NPJ Vaccines 8, 123 (2023).

    Google Scholar 

42. Principi, N. & Esposito, S. Development of vaccines against emerging mosquito-vectored arbovirus infections. Vaccines 12, 87 (2024).

    CAS  Google Scholar 

43. Yoon, I. K. et al. High rate of subclinical chikungunya virus infection and association of neutralizing antibody with protection in a prospective cohort in the Philippines. PLoS Negl. Trop. Dis. 9, e0003764 (2015).

    Google Scholar 

44. Roques, P. et al. Effectiveness of CHIKV vaccine VLA1553 demonstrated by passive transfer of human sera. JCI Insight 7, e160173 (2022).

    Google Scholar 

45. U.S. Food and Drug Administration. FDA and CDC recommend pause in use of Ixchiq (Chikungunya Vaccine, Live) in individuals 60 years of age and older while postmarketing safety reports are investigated: FDA Safety Communication. https://www.fda.gov/safety/medical-product-safety-information/fda-and-cdc-recommend-pause-use-ixchiq-chikungunya-vaccine-live-individuals-60-years-age-and-older (2025).

46. Raeven, R. H. M., van Riet, E., Meiring, H. D., Metz, B. & Kersten, G. F. A. Systems vaccinology and big data in the vaccine development chain. Immunology 156, 33–46 (2019).

    CAS  Google Scholar 

47. Nemes, E. et al. The quest for vaccine-induced immune correlates of protection against tuberculosis. Vaccin. Insights 1, 165–181 (2022).

    Google Scholar 

48. Chen, W. H. et al. Single-dose Live Oral Cholera Vaccine CVD 103-HgR protects against human experimental infection with Vibrio cholerae O1 El Tor. Clin. Infect. Dis. 62, 1329–1335 (2016).

    CAS  Google Scholar 

49. Frenck, R. W. Jr. et al. Efficacy, safety, and immunogenicity of the Shigella sonnei 1790GAHB GMMA candidate vaccine: results from a phase 2b randomized, placebo-controlled challenge study in adults. EClinicalMedicine 39, 101076 (2021).

    Google Scholar 

50. Conti, V. et al. Putative correlates of protection against shigellosis assessing immunomarkers across responses to S. sonnei investigational vaccine. NPJ Vaccines 9, 56 (2024).

    CAS  Google Scholar 

51. Micoli, F., Nakakana, U. N. & Berlanda Scorza, F. Towards a four-component GMMA-based vaccine against Shigella. Vaccines 10, 328 (2022).

    CAS  Google Scholar 

52. Rossi, O. et al. A next-generation GMMA-based vaccine candidate to fight shigellosis. NPJ Vaccines 8, 130 (2023).

    CAS  Google Scholar 

53. Leroux-Roels, I. et al. Safety and immunogenicity of a 4-component generalized modules for membrane antigens Shigella vaccine in healthy European adults: randomized, phase 1/2 study. J. Infect. Dis. 230, e971–e984 (2024).

    CAS  Google Scholar 

54. Novavax. Novavax announces topline RSV F vaccine data from two clinical trials in older adults. https://ir.novavax.com/2016-09-25-Novavax-Announces-Topline-RSV-F-Vaccine-Data-from-Two-Clinical-Trials-in-Older-Adults (2016).

55. Sadoff, J. et al. Prevention of respiratory syncytial virus infection in healthy adults by a single immunization of Ad26.RSV.preF in a human challenge study. J. Infect. Dis. 226, 396–406 (2022).

    CAS  Google Scholar 

56. Schmoele-Thoma, B. et al. Vaccine efficacy in adults in a respiratory syncytial virus challenge study. N. Engl. J. Med. 386, 2377–2386 (2022).

    CAS  Google Scholar 

57. Schmoele-Thoma, B. et al. Phase 1/2, first-in-human study of the safety, tolerability, and immunogenicity of an RSV prefusion F-based subunit vaccine candidate. Open Forum Infect. Dis. 6, S970 (2019).

    Google Scholar 

58. Ricco, M. et al. Efficacy of respiratory syncytial virus vaccination to prevent lower respiratory tract illness in older adults: a systematic review and meta-analysis of randomized controlled trials. Vaccines 12, 500 (2024).

    CAS  Google Scholar 

59. Falsey, A. R. et al. Efficacy and safety of an Ad26.RSV.preF-RSV preF protein vaccine in older adults. N. Engl. J. Med. 388, 609–620 (2023).

    CAS  Google Scholar 

60. Schwarz, T. F. et al. Immunogenicity and safety following 1 dose of AS01_E-adjuvanted respiratory syncytial virus prefusion F protein vaccine in older adults: a phase 3 trial. J. Infect. Dis. 230, e102–e110 (2024).

    CAS  Google Scholar 

61. Cavaleri, M. et al. Fourth controlled human infection model (CHIM) meeting, CHIM regulatory issues, May 24, 2023. Biologicals 85, 101745 (2024).

    Google Scholar 

62. Abo, Y. N. et al. Strategic and scientific contributions of human challenge trials for vaccine development: facts versus fantasy. Lancet Infect. Dis. 23, e533–e546 (2023).

    Google Scholar 

63. Msusa, K. P., Rogalski-Salter, T., Mandi, H. & Clemens, R. Critical success factors for conducting human challenge trials for vaccine development in low- and middle-income countries. Vaccine 40, 1261–1270 (2022).

    Google Scholar 

64. Bagnoli, F., Galgani, I., Vadivelu, V. K. & Phogat, S. Reverse development of vaccines against antimicrobial-resistant pathogens. NPJ Vaccines 9, 71 (2024).

    Google Scholar 

65. Sauvat, L. et al. Vaccines and monoclonal antibodies to prevent healthcare-associated bacterial infections. Clin. Microbiol. Rev. 37, e00160–00122 (2024).

    Google Scholar 

66. Palacios, R. et al. Double-blind, randomized, placebo-controlled Phase III clinical trial to evaluate the efficacy and safety of treating healthcare professionals with the adsorbed COVID-19 (inactivated) vaccine manufactured by Sinovac – PROFISCOV: a structured summary of a study protocol for a randomised controlled trial. Trials 21, 853 (2020).

    CAS  Google Scholar 

67. Palacios, R. et al. Efficacy and safety of a COVID-19 inactivated vaccine in healthcare professionals in Brazil: the PROFISCOV study. https://doi.org/10.2139/ssrn.3822780 (2021).

68. World Health Organization. Background document on the inactivated vaccine Sinovac-CoronaVac against COVID-19: background document to the WHO Interim recommendations for use of the inactivated COVID-19 vaccine, CoronaVac, developed by Sinovac. World Health Organization. https://iris.who.int/handle/10665/341455 (2021).

69. Kaslow, D. C. Force of infection: a determinant of vaccine efficacy? NPJ Vaccines 6, 51 (2021).

70. Liu, M., Li, Q., Lin, J., Lin, Y. & Hoffman, E. Innovative trial designs and analyses for vaccine clinical development. Contemp. Clin. Trials 100, 106225 (2021).

    Google Scholar 

71. Senn, S. The design and analysis of vaccine trials for COVID-19 for the purpose of estimating efficacy. Pharm. Stat. 21, 790–807 (2022).

    Google Scholar 

72. Puthumana, J., Egilman, A. C., Zhang, A. D., Schwartz, J. L. & Ross, J. S. Speed, evidence, and safety characteristics of vaccine approvals by the US Food and Drug Administration. JAMA Intern. Med. 181, 559–560 (2021).

    Google Scholar 

73. U.S. Food and Drug Administration. BLA Clinical Review Memo, August 23, 2021 – COMIRNATY. https://www.fda.gov/media/152256/download (2021).

74. Wong, J. C., Lao, C. T., Yousif, M. M. & Luga, J. M. Fast tracking-vaccine safety, efficacy, and lessons learned: a narrative review. Vaccines10, 1256 (2022).

    CAS  Google Scholar 

75. Kalinke, U. et al. Clinical development and approval of COVID-19 vaccines. Expert Rev. Vaccines 21, 609–619 (2022).

    CAS  Google Scholar 

76. Joffe, S. et al. Data and safety monitoring of COVID-19 vaccine clinical trials. J. Infect. Dis. 224, 1995–2000 (2021).

    CAS  Google Scholar 

77. Wong, F., de la Fuente-Nunez, C. & Collins, J. J. Leveraging artificial intelligence in the fight against infectious diseases. Science 381, 164–170 (2023).

    CAS  Google Scholar 

78. Jin, M., Li, Q. & Kaur, A. Bayesian design for pediatric clinical trials with binary endpoints when borrowing historical information of treatment effect. Ther. Innov. Regul. Sci. 55, 360–369 (2021).

    Google Scholar 

79. Olawade, D. B. et al. Leveraging artificial intelligence in vaccine development: A narrative review. J. Microbiol. Methods 224, 106998 (2024).

    CAS  Google Scholar 

80. Warne, N. et al. Delivering 3 billion doses of Comirnaty in 2021. Nat. Biotechnol. 41, 183–188 (2023).

    CAS  Google Scholar 

81. Buck, P. O. et al. New vaccine platforms-Novel dimensions of economic and societal value and their measurement. Vaccines12, 234 (2024).

    Google Scholar 

82. Mathieu, E. et al. Coronavirus Pandemic (COVID-19). https://ourworldindata.org/coronavirus (2020).

83. Ferranna, M. Causes and costs of global COVID-19 vaccine inequity. Semin. Immunopathol. 45, 469–480 (2024).

    Google Scholar 

84. Ismail, S. J. et al. A framework for the systematic consideration of ethics, equity, feasibility, and acceptability in vaccine program recommendations. Vaccine 38, 5861–5876 (2020).

    CAS  Google Scholar 

85. El Banhawi, H. et al. Socio-economic value of adult immunisation programmes. Office of Health Economics Contract Research Report. https://www.ohe.org/publications/the-socio-economic-value-of-adult-immunisation-programmes/ (2024).

86. Wellcome Trust. COVID-19 vaccines: the factors that enabled unprecedented timelines for clinical development and regulatory authorisation. https://cms.wellcome.org/sites/default/files/2022-03/unprecedented-timelines-covid19-vaccine-clinical-development-authorisation.pdf (2022).

87. Cavaleri, M. et al. A roadmap for fostering timely regulatory and ethics approvals of international clinical trials in support of global health research systems. Lancet Glob. Health 13, e769–e777 (2025).

    CAS  Google Scholar 

88. Laigle, V. et al. Vaccine market access pathways in the EU27 and the United Kingdom-analysis and recommendations for improvements. Vaccine 39, 5706–5718 (2021).

    Google Scholar 

89. Brighton Collaboration. Safety Platform for Emergency vACcines (SPEAC). https://brightoncollaboration.us/speac/ (2022).

90. Bauchau, V. et al. Multi-stakeholder call to action for the future of vaccine post-marketing monitoring: proceedings from the first beyond COVID-19 monitoring excellence (BeCOME) conference. Drug Saf. 48, 577–585 (2025).

    Google Scholar 

91. Tortorice, D., Rappuoli, R. & Bloom, D. E. The economic case for scaling up health research and development: lessons from the COVID-19 pandemic. Proc. Natl. Acad. Sci. USA 121, e2321978121 (2024).

    CAS  Google Scholar 

92. Doherty, T. M., Di Pasquale, A., Finnegan, G., Lele, J. & Philip, R. K. Sustaining the momentum for adult vaccination post-COVID-19 to leverage the global uptake of life-course immunisation: a scoping review and call to action. Int. J. Infect. Dis. 142, 106963 (2024).

    Google Scholar 

93. Doherty, T. M., Ecarnot, F., Gaillat, J. & Privor-Dumm, L. Nonstructural barriers to adult vaccination. Hum. Vaccin. Immunother. 20, 2334475 (2024).

    Google Scholar 

Download references

Author notes

1. Vega Masignani

   Present address: CARB-X, Boston University School of Law, Boston, MA, USA

Authors and Affiliations

1. GSK, Siena, Italy

   Vega Masignani & Ricardo Palacios

2. GSK, Wavre, Belgium

   Marie-Thérèse Martin & Fabian Tibaldi

3. GSK, Rockville, MD, USA

   V. Kumaran Vadivelu

Contributions

All authors contributed towards conception, drafting, and editing of the manuscript, reviewed the draft critically for important intellectual content, approved the final manuscript before it was submitted by the corresponding author, and agree to be accountable for all aspects of the work.

Corresponding author

Correspondence to Ricardo Palacios.

Competing interests

All authors are or were employees of the GSK and hold financial equities in GSK.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions