Traditional approaches for vaccine discovery and development are being reimagined and accelerated – leveraging technologies in novel ways to streamline workflows and forging collaborations. Dr. Sanjay Garg and Dr. Anis H Khimani, discuss key learnings and best practices during COVID-19 vaccine research and how we can take these forward to pathogens yet unknown.
Typically, it takes five-18 years and costs $200-$500 million (£149-374 million) to develop a vaccine (Kis, 2018). It is a lengthy, complex process where few candidates proceed from early stages of discovery to approval and manufacturing (Lurie, 2020).
To ensure safety throughout research and development, vaccine developers have historically employed a sequential process, marked by frequent pauses for data analysis and quality improvement (Lurie, 2020).
However, traditional timelines and workflows needed to change when the Severe Acute Respiratory Syndrome – Coronavirus 2 (SARs-CoV-2) virus began spreading in late 2019 and early 2020. With millions of people worldwide contracting coronavirus starting in late 2019 (COVID-19), vaccine developers were faced with the enormous challenge of producing a safe, effective vaccine in months, not years, to help stop the global footprint of the deadly disease.
In fact, the World Health Organization (WHO) released a research roadmap that established a global imperative to accelerate research as “a moral obligation to learn as much as possible, as quickly as possible” (World Health Organization, 2020).
What we are participating in today – as scientists and innovators -– is changing vaccine development now, and in the future through collaboration, not competition. Stakeholders are sharing the common goal of countering the virus by providing long-term immunity.
In short, everyone is against COVID-19 and we cannot cut corners on safety, immunogenicity, or compliance.
In support of these herculean scientific efforts, technology and solution providers have accelerated and configured their platforms with virus research workflow protocols and related applications development to arm academic and industry researchers with the tools they need.
Technologies, including synthetic biology, informatics and artificial intelligence, are facilitating various application areas, resulting in rapid identification and characterisation of the viral genome, study of the viral life cycle, disease progression, diagnostics, as well as anti-viral therapeutics and vaccine development. Furthermore, advances in bioanalytical and physical characterisation and automation and digitisation are supporting rapid process development for anti-viral biologics.
Together, the fresh thinking in our R&D mandates and technological innovations are ushering in a new age of vaccine discovery and development.
A new paradigm in vaccine development
The slow, linear, methodical process of the past is being reconstructed. In the sprint to protect the public’s safety from COVID-19, clinical trial phases overlap instead of occurring sequentially. In the New England Journal of Medicine Coalition for Epidemic Preparedness Innovations (CEPI) advisor Nicole Lurie, M.D., and colleagues plotted the difference between traditional vaccine development and a pandemic paradigm (Lurie, 2020). In the new paradigm, preclinical studies may occur in tandem with first-in-human studies. Phase I may intersect with Phase II. And Phase II may crossover into Phase III.
When initial results show promise, decisions are being made for mass distribution, manufacturing large quantities in parallel with Phase III studies, as well as continuing to emphasise the importance of conducting post-marketing Phase IV studies to continue data gathering and analysis.
To make the most of each day of research, developers must harness all the technology available to increase speed without reducing safety.
Recent developments in predicting vaccine designs via the use of reverse vaccinology (RV) have emerged (Moxon et al., 2019). This approach leverages bioinformatics to perform the analysis of the genome of the pathogen, for example. One such web-based bioinformatics tool, Vaxign, applies machine learning tools to predict putative vaccine candidates with improved confidence (He et al., 2010), (Vaxign-ML; Ong et al., 2020). Furthermore, artificial intelligence (AI) approaches can expedite many aspects of clinical research, adaptive clinical trials are leveraging insights from accumulating data to make prospective changes in trial design (Harrer, 2019) and machine learning can help streamline efficiency in patient recruitment, enrollment, and monitoring (Harrer, 2019).
mRNA, protein and virus-based approaches to vaccine development- opportunities & challenges
As of 2 December 2020, 51 COVID-19 vaccine candidates were in clinical evaluation, several having reached the critical clinical milestone, and another 163 in preclinical evaluation (World Health Organization update, Dec, 02, 2020). The COVID-19 vaccine candidates include mRNA, recombinant protein and virus-based approaches (van Riel, 2020).
Recombinant protein or virus-like particles mimic the virus’s structure, but lack of a viral genome prevents them from replicating, improving the safety profile compared with live viral vaccines (van Riel, 2020). While the virus-like particle vaccine has proven successful over time in protecting against smallpox and HPV, they take longer to produce (van Riel, 2020). They must be grown under strict conditions with extensive safety testing.
Newer viral-based approaches offer the advantage of speed but have yet to be time-tested in humans. Viral vector vaccines carry DNA containing virus-encoded antigens into the cells to stimulate an immune response (Corey, 2020). Two of three vaccine candidates in Phase III clinical trials in the United States that use this approach (World Health Organization, 2020) are: Ad.26.COV2.S developed by Janssen Pharmaceutical Companies of Johnson and Johnson; and, AZD1222 developed by the University of Oxford and AstraZeneca. Details around different clinical studies can be accessed at (https://clinicaltrials.gov/).
Other next-generation vaccines use nucleic acid, both DNA and RNA. These types of vaccines are easy to design once a viral genome has been sequenced and easy to produce but are not without challenges. DNA-based vaccines, when administered intramuscularly can show reduced immunogenicity (van Riel, 2020) (Brisse, 2020). Therefore, alternative delivery systems are under review using electroporation (Corey, 2020) (Brisse, 2020).
RNA-based vaccines show higher immunogenicity than DNA vaccines. RNA vaccines can be designed in many ways to activate an immune response. But prior to COVID-19, no RNA-based vaccine has ever been approved for use in humans. Messenger RNA (mRNA) vaccines use lipid nanoparticles to protect and deliver the molecule (Corey, 2020). Of the multiple mRNA COVID-19 vaccines under investigation, two have entered Phase 3 clinical trials in the USA (van Riel, 2020) (World Health Organization, 2020): mRNA-1273 developed by Moderna, Inc.; and, the National Institutes of Allergy and Infectious Disease (NIAID) and BNT162b2 developed by BioNTech/Pfizer with the latter having received emergency use authorization in the UK already.
As and if these newer vaccine approaches help in containing the spread of COVID-19, it represents a milestone achievement in terms of speed and technology.
Science-based vaccine discovery
To shorten the timelines of clinical research, we need to have a good science base for vaccines.
A rational approach to vaccine development can increase speed and reduce costs by concentrating on the most prolonged vaccine development phase—discovery and target validation (Brisse, 2020). One of the strategies to minimise attrition rates in trials from Phase-I to Phase-III is improving the quality of target identity and validation upstream.
To identify the right target, qualifying its relevance to the disease and determining its antigenicity for vaccine design is key. More recent molecular screening approaches utilising RNA-based and CRISPR technologies have advanced target identification. In addition, systems thinking can help researchers understand the mechanisms underlying pathogenesis and characterise the effect on the immune system (Dhillon, 2020).
In addition, researchers used to study virology, viral pathogenesis, and immunology separately. Now, all those aspects are considered together. A holistic perspective of immune response combines genomics, transcriptomics, and proteomics—backed up by artificial intelligence and bioinformatics. New technology with multiplexing infrastructure is needed to look into how the virus is interacting and tracking with the immune system at the cellular level.
High-throughput technology can accelerate immune profiling and vaccine development (Oberg, 2011). Transcriptomics and proteomics provide insights on the biomarkers of response to a vaccine. Traditional histology with advancements in multiplex imaging for surveillance of humoral and cellular responses identifies and quantifies subsets of cells involved in the immune response like antigen-specific B cells and T cells (van Riel, 2020). Recent advances in flow cytometry, such as imaging flow cytometry and multi-parameter flow cytometry, offer even greater capabilities. Flow cytometry data can be integrated with gene expression, biomarker profiles, serum cytokine levels, virus titers, and ELISPOT data to identify individuals who will and will not respond to a vaccine (Pezeshki, 2019). RNA sequencing reveals immune cell development and regulatory networks to predict immune function (Cotugno, 2019).
To explore the mechanisms leading to a successful vaccine, proteomics supports the study of immune response effectors like proteins, antibodies, and cytokines. Traditional methods of ELISA and Western blot limit the number of proteins that can be studied simultaneously. However, higher throughput bead- based and time-resolved fluorescence (TRF / TR-FRET) assays, as well as mass spectrometry-based proteomics deliver the precision, throughput, as well as sensitivity for T cell profiling and antigenic B cell activity (Cotugno, 2019).
All these approaches create large amounts of multiparametric data. To combine and analyse the data, bioinformatics experts need to be added to the team to visualise and interpret data. Simultaneously, artificial intelligence can develop models to analyse multi-assay data, predict immune response, and validate data.
In addition to the science itself, companies must be thinking of how science, process development, and supply chain interact as the multidisciplinary nature of future vaccine development teams also includes supply, manufacturing, and logistics. For example, mRNA vaccines have to be formulated and transported in a frozen state. Hence, in parallel with the science of vaccine development, the clinical healthcare ecosystem must also build infrastructure that includes freezer farms and onsite storage to support distribution and availability to even the most remote communities.
Collaboration at every level
Leading manufacturers are collaborating and licensing technology from early phase biotech companies, including Sanofi Pasteur and GlaxoSmithKline, combining technology in the development of a recombinant protein vaccine candidate funded by BARDA.
Unprecedented levels of collaboration are occurring at every level. At pharmaceutical and biotechnology companies, virtual cross-functional teams with representation from senior leadership, research and development, manufacturing, regulatory, informatics, and marketing collaborate in real-time with broad access to information.
Stronger going forward
Immunogenic vaccines are fundamental to public health. Partnerships with government, industry, and academia are needed. We can’t do it alone. No one can claim it all. It is about collaboration.
To strengthen preparedness for future pandemics, we need science-focused vaccine development programs in academia and industry, with a broader perspective. We have to take a wider view and stop looking at only the obvious pathogens. COVID-19 has shown us that we need to look into other, riskier pathogens with the potential to be pandemic in nature.
As early as March of 2020, the WHO outlined a blueprint to develop cross-cutting research and development preparedness for disease “X,” a currently unknown pathogen capable of causing another international epidemic (World Health Organization, 2020). In its call-to-action, the WHO stated that preparedness requires an understanding of immunity and pathophysiology with serological testing and assays that monitor response to treatment and prognostic markers. Furthermore, the global health agency prioritised access to reagents such as virus isolates, panels of clinical samples, research reagents, and quality control reagents (World Health Organization, 2020).
While current scaled-up innovation focuses on restructuring and accelerating workflows for vaccine development for COVID-19, it is also intrinsically about creating a new era in vaccine development.
Disclaimer: The information in this white paper reflects Dr. Sanjay Garg’s views on this topic and should not be taken as representing Sanofi Pasteur.
Volume 22, Issue 1 – Winter 2020/21
About the Authors:
Dr. Sanjay Garg, Senior Expert Scientist and Platform Head, Vendor Management R&D Global Operations, Sanofi Pasteur.As a vaccine expert, Dr. Sanjay Garg, Senior Expert Scientist and Platform Head, Vendor Management R&D Global Operations, Sanofi Pasteur, has studied vaccine development for two decades. He followed his academic research training at Emory University, Atlanta, GA, with experience at the Centers for Disease Control and Prevention, also in Atlanta.
Dr. Anis H. Khimani, leads strategy and market segment development at PerkinElmer.Dr. Anis H. Khimani, who leads strategy and market segment development at PerkinElmer has prior academic and research background in molecular biology, viral pathogenesis and vaccine studies from Harvard Medical School and Dana-Farber Cancer Institute, followed by research and development in genomics and assay development at PerkinElmer. Subsequently, his focus has been in product management and business development within the Life Sciences segment of the company.
References
Brisse, M. V. (2020, September 30). Emerging concepts and technologies in vaccine development. Frontiers in Immunology. Retrieved from https://www.frontiersin.org/articles/10.3389/fimmu.2020.583077/full
Corey, L. M. (2020, May 11). A strategic approach to COVID-19 vaccine R&D. Science, 948-950. Retrieved from http://science.sciencemag.org/content/368/6494/948
Cotugno, N. R.-T. (2019). OMIC technologies and vaccine development: From the identification of vulnerable individuals to the formulation of invulnerable vaccines. Journal of Immunology Research, 2019, Article ID 8732191. Retrieved from https://www.hindawi.com/journals/jir/2019/8732191/
Dhillon, B. M. (2020, July 30). Systems biology approaches to understanding the human immune system. Frontiers in Immunity, 11, 1683. Retrieved from https://www.frontiersin.org/articles/10.3389/fimmu.2020.01683/full
Harrer, S. A. (2019, August). Artificial intelligence for clinical trial design. Trends in Pharmaceutical Sciences, 40(8), 577-591. Retrieved from https://doi.org/10.1016/j.tips.2019.05.005
He, Y., Xiang, Z., Mobley, HLT (2010). Vaxign: the first web-based vaccine design program for reverse vaccinology and applications for vaccine development. J. Biomed. Biotechnol.2010:1-15
Kis, Z. S. (2018). Emerging technologies for low-cost rapid vaccine manufacture. Biotechnology Journal. Retrieved from https://onlinelibrary.wiley.com/doi/full/10.1002/biot.201800376
Lurie, N. S. (2020, May 21). Developing Covid-19 vaccines at pandemic speed. The New England Journal of Medicine, 1969-1973. Retrieved from https://www.nejm.org/doi/full/10.1056/NEJMp2005630
Moxon, R., Reche, P.A., and Rappuoli, R., (2019). Reverse Vaccinology (editorial). Front. Immunol., 10, 2776. Retrieved from https://www.frontiersin.org/articles/10.3389/fimmu.2019.02776/full
Oberg, A. K. (2011). Systems biology approaches to new vaccine development. Current Opinion in Immunology, 23(3), 436-443. Retrieved from https://www.sciencedirect.com/science/article/abs/pii/S0952791511000409?via%3Dihub
Ong E., Wang, H., Wong, M. U., Seetharaman, M., Valdez, N., He Y. (2020). Vaxign-ML: supervised machine learning reverse vaccinology model for improved prediction of bacterial protective antigens. Bioinformatics.36:3185-91
Pezeshki, A. O. (2019). The role of systems biology approaches in determining molecular signatures for the development of more effective vaccines. Expert Review of Vaccines, 18(3), 253-267. Retrieved from https://www.tandfonline.com/doi/abs/10.1080/14760584.2019.1575208?journalCode=ierv20
U.S. Health and Human Services. (2020, May 21). Trump Administration’s Operation Warp Speed accelerates AstraZeneca COVID-19 vaccine to be available beginning in October.Retrieved October 21, 2020, from HHS.gov: https://www.hhs.gov/about/news/2020/05/21/trump-administration-accelerates-astrazeneca-covid-19-vaccine-to-be-available-beginning-in-october.html
van Riel, D. d. (2020, August). Next-generation vaccine platforms for COVID-19. Nature Materials, 19, 810-820. Retrieved from https://doi.org/10.1038/s41563-020-0746-0
Vaxign-ML. Retrieved from http://www.violinet.org/vaxign/
World Health Organization. (2020). A coordinated global research roadmap: 2019 novel coronavirus.Geneva: R&D Blueprint. Retrieved from https://www.who.int/blueprint/priority-diseases/key-action/Coronavirus_Roadmap_V9.pdf?ua=1
World Health Organization. (2020, October 19). Draft landscape of COVID-19 candidate vaccines.Retrieved December 02, 2020, from Who.it: https://www.who.int/publications/m/item/draft-landscape-of-covid-19-candidate-vaccines
