A few months before the first COVID-19 vaccines received Emergency Use Authorization (EUA) in late 2020, a global vaccine safety expert cautioned the rushed circumstances and said it is essential to “get [safety monitoring] right” by “intensively” and “robustly” scrutinising adverse events following the experimental rollout.
As this expert stated, “Deploying any new vaccine based on data from truncated or expedited clinical trials into a population without a functioning safety monitoring system in place is reckless and irresponsible given the tools that are available.”
Moreover, she added, any investments needed to beef up safety monitoring would be “inexpensive in comparison” to the massive funding allocated to COVID-19 vaccine development and scale-up.
Vaccines against COVID-19 are being developed at speeds not previously achieved. With this unprecedented effort comes challenges for post-marketing safety monitoring and challenges for vaccine safety communication. To deploy these new vaccines fast across diverse populations, it is vital that robust pharmacovigilance and active surveillance systems are in place. Not all countries have the capability or resources to undertake adequate surveillance and will rely on data from those who can.
The tools exist to assess COVID-19 vaccines as they are deployed such as surveillance systems, administrative data and case definitions for adverse events of special interest. However, stitching these all together and using them effectively requires investment and collaboration. This paper provides a high-level overview of some of the facets of modern vaccine safety assessment and how they are, or can be, applied to COVID-19 vaccines.
An unprecedented commitment to developing and producing vaccines against SARS-CoV-2 in record time is underway and new candidates are entering clinical testing almost weekly. The speed at which development is unfolding has led to widespread concern among both health professionals and the public that vital steps may be skipped, in particular the assessment of safety.
Vaccine development has traditionally been a long process taking an average of 10–15 years. The vaccine with the shortest timeline from antigen discovery to licensure is the mumps vaccine, which took 4 years. It is the high financial cost, particularly high-risk advanced clinical development, coupled with the investment in production facilities that has, in part, hampered nimble vaccine responses to emerging infectious diseases. However, recent developments in technology along with unprecedented collaboration and investment mean we may be able to escape the barriers of the past .
As well as the speed at which vaccine candidates were advanced, where possible the clinical development and regulatory phases are occurring alongside each other rather than sequentially . This means that while all steps are adhered to, their timing can be expedited.
The desperate need for an Ebola vaccine galvanised us, and in less than 12 months, 12 clinical trials ran the gamut from a “first in man” dosing study to a phase III efficacy trial .
This was achieved through successful collaborations and running these stages in parallel [4, 5]. However, while the pre-licensure clinical programme was executed in record time, fragile settings are often ill equipped for post-licensure safety surveillance.
Pre-Clinical: Assessment in Animal Models
Potential vaccine candidates need to be assessed in suitable animals for safety, immunogenicity and efficacy under challenge. Translating data from any single animal to humans can be problematic as the disease may not mimic human infection accurately, and therefore fail to predict vaccine effects, positive or negative.
The models for assessing SARS-CoV-2 vaccines include mouse (transgenic for the human ACE2 receptor), hamster, ferret and non-human primates, depending on what question is being asked .
Studies of earlier SARS vaccines in animals identified two potential safety issues; antibody-dependent enhancement and cellular immunopathology.
These have not been observed in human studies but flag potential responses for close examination and highlight the importance of selecting vaccine approaches and adjuvants that drive desirable responses .
Pre-Licensure: Assessment in Humans
Clinical studies in humans generally follow three phases. Phase I with 10’s (~ 30 to 50) of healthy volunteers assesses the safety, immunogenicity and dose ranging; phase II progresses to 100’s of volunteers and assesses safety and immunogenicity; phase III includes 10,000’s of volunteers and assesses safety and efficacy. Phase III is usually placebo-controlled studies and while safety continues to be studied efficacy will be assessed.
Normally, these phases progress sequentially after careful assessment of results of each stage before moving to the next. In the case of COVID-19 vaccines, as with recent Ebola vaccines, these stages can be expedited without skipping anything thanks to investment and collaboration .
Each trial will have an independent drug safety monitoring board and ideally this group will have diverse expertise, including a biostatistician. For COVID-19 vaccines, it is recommended that there be persons with expertise in rare disease epidemiology. A meta-drug safety monitoring board has been established to ensure that high-level expertise is available to support all drug safety monitoring boards [8, 9].
Challenges and Solutions for the Safe and Responsible Deployment of COVID-19 Vaccines
Too few countries have high functioning pharmacovigilance systems, and far fewer are able to undertake robust signal verification and post-licensure studies on safety. These countries will need to rely on data generated by those who do have the capability, perhaps placing some further ethical obligations on those countries who can, rather than rely on the predominant data contributions from Europe and the USA.
Adverse events will coincide temporally with vaccine administration . Prior to the use of COVID-19 vaccines, it is important to understand the background rates of conditions that may be temporally associated with vaccine administration to be able to assess observed rates vs the expected rates . For most events, these rates are unknown and to further complicate matters the rates of many events, such as multiple sclerosis, vary by sex and geography [32, 33]. Developing background rates for COVID-19 vaccine AESIs for as many populations as possible is a matter of urgency.
Deploying any new vaccine based on data from expedited clinical trials into a population without a functioning safety monitoring system in place is reckless and irresponsible given the tools that are available. While there are international collaborations aimed at supporting coordinated efforts in COVID-19 vaccine safety assessments, vaccine nationalism and a lack of a globally coordinated vaccine safety effort could limit the potential in this space. Furthermore, deployment of vaccines before the successful completion of robust clinical programmes could threaten not only public confidence in COVID-19 vaccines but also immunisation programmes in general.
While the clinical testing of COVID-19 vaccines can be done robustly and assessment by regulatory agencies can be stringent, the vaccines are likely to be used under emergency conditions and the follow-up time from the trials will be minimal. Under such conditions, it is vital that the products are monitored (in near real time) for rare adverse events until risks can be either quantified or excluded (see Box for a case study).
Only a few countries have the capability to conduct this monitoring  and even fewer are prepared with systems at the ready and baseline rates of AESIs established. There is an urgency to support as many sites as possible to prepare in collaboration with each other to actively monitor COVID-19 vaccines as they are deployed using common protocols so that data may be pooled, and rare events assessed in diverse populations.
We have the tools to intensively monitor the safety of COVID-19 vaccines. While billions are being spent on the development and scale manufacturing of vaccines that have yet to demonstrate efficacy, with the exception of the European Union, there is limited investment in the post-licensure phase yet, which is inexpensive in comparison. Failure to assess these vaccines for safety to our full ability is wrong. As we well know from extensive experience, vaccine safety issues can threaten not only the success of any COVID-19 vaccine programme but also routine immunisation programmes. It is vital we get this right and we have the tools and the expertise to do so and to do it well.
1. Brende B, Farrar J, Gashumba D, Moedas C, Mundel T, Shiozaki Y, et al. CEPI: a new global R&D organisation for epidemic preparedness and response. Lancet. 2017;389(10066):233–235. doi: 10.1016/S0140-6736(17)30131-9. [PMC free article] [PubMed] [CrossRef] [Google Scholar] 2. Lurie N, Saville M, Hatchett R, Halton J. Developing Covid-19 vaccines at pandemic speed. N Engl J Med. 2020;382(21):1969–1973. doi: 10.1056/NEJMp2005630. [PubMed] [CrossRef] [Google Scholar] 3. Henao-Restrepo AM, Longini IM, Egger M, Dean NE, Edmunds WJ, Camacho A, et al. Efficacy and effectiveness of an rVSV-vectored vaccine expressing Ebola surface glycoprotein: interim results from the Guinea ring vaccination cluster-randomised trial. Lancet. 2015;386(9996):857–866. doi: 10.1016/S0140-6736(15)61117-5. [PubMed] [CrossRef] [Google Scholar] 4. Wolf J, Bruno S, Eichberg M, Jannat R, Rudo S, VanRheenen S, et al. Applying lessons from the Ebola vaccine experience for SARS-CoV-2 and other epidemic pathogens. NPJ Vaccines. 2020;5:51. doi: 10.1038/s41541-020-0204-7. [PMC free article] [PubMed] [CrossRef] [Google Scholar] 5. Henao-Restrepo AM, Preziosi M-P, Wood D, Moorthy V, Kieny MP. On a path to accelerate access to Ebola vaccines: the WHO’s research and development efforts during the 2014–2016 Ebola epidemic in West Africa. Curr Opin Virol. 2016;17:138–144. doi: 10.1016/j.coviro.2016.03.008. [PMC free article][PubMed] [CrossRef] [Google Scholar] 6. Lakdawala SS, Menachery VD. The search for a COVID-19 animal model. Science. 2020;368(6494):942–943. doi: 10.1126/science.abc6141. [PubMed] [CrossRef] [Google Scholar] 7. Lambert P-H, Ambrosino DM, Andersen SR, Baric RS, Black SB, Chen RT, et al. Consensus summary report for CEPI/BC March 12-13, 2020 meeting: assessment of risk of disease enhancement with COVID-19 vaccines. Vaccine. 2020;38(31):4783–4791. doi: 10.1016/j.vaccine.2020.05.064. [PMC free article][PubMed] [CrossRef] [Google Scholar] 8. Petkova E, Antman EM, Troxel AB. Pooling data from individual clinical trials in the COVID-19 era. JAMA. 2020;324(6):543–545. doi: 10.1001/jama.2020.13042. [PubMed] [CrossRef] [Google Scholar] 9. Brighton Collaboration. SPEAC. 2020. https://brightoncollaboration.us/speac/. Accessed 17 Aug 2020.
30. Siegrist C-A, Lewis EM, Eskola J, Evans SJ, Black SB. Human papilloma virus immunization in adolescent and young adults: a cohort study to illustrate what events might be mistaken for adverse reactions. Pediatr Infect Dis J. 2007;26(11):979–984. doi: 10.1097/INF.0b013e318149dfea. [PubMed] [CrossRef] [Google Scholar]
31. Black S, Eskola J, Siegrist C-A, Halsey N, MacDonald N, Law B, et al. Importance of background rates of disease in assessment of vaccine safety during mass immunisation with pandemic H1N1 influenza vaccines. Lancet. 2009;374(9707):2115–2122. doi: 10.1016/S0140-6736(09)61877-8. [PMC free article][PubMed] [CrossRef] [Google Scholar]
34. Donahue JG, Kieke BA, Lewis EM, Weintraub ES, Hanson KE, McClure DL, et al. Near real-time surveillance to assess the safety of the 9-valent human papillomavirus vaccine. Pediatrics. 2019;144(6):e20191808. doi: 10.1542/peds.2019-1808. [PMC free article] [PubMed] [CrossRef] [Google Scholar]