key: cord-0859664-6xgy5u8b
authors: Sherman, Amy C.; Rouphael, Nadine; Baden, Lindsey R.
title: COVID-19 Vaccine Trials (and Tribulations): How to improve the process of clinical trials in a pandemic
date: 2022-04-18
journal: Clin Infect Dis
DOI: 10.1093/cid/ciac301
sha: c8aabf74749cf0fd0e4132c6f3c8d5746c12e4fd
doc_id: 859664
cord_uid: 6xgy5u8b

Vaccine clinical trials have been essential to develop effective SARS-CoV-2 vaccines. The challenges of supply chain disruptions, infection control, study designs, and participant factors that affect trial procedures are reviewed, with specific solutions to streamline the clinical trial process.

A C C E P T E D M A N U S C R I P T

Although RCTs are designed to be the simplest path to demonstrate efficacy of a new drug 12 product, the development from the pre-clinical stage to licensure is lengthy, with each phase (1-4) 13 conventionally progressing in a stepwise fashion and requiring months to years to complete each phase. 14 Clearly this laborious and protracted process is misaligned with the pandemic induced urgency to rapidly 15 develop a preventative agent. At the beginning of the COVID-19 pandemic, even the pre-clinical data 16 were intrinsically limited since there were no validated SARS-CoV-2 animal models. To address this issue, 17 preclinical and toxicology data from related vaccines (SARS-CoV and MERS-CoV candidates) were used 18 to expedite much of the pre-clinical vaccine development work [10] . The clinical trials were further 19 designed to have overlapping phases. Indeed, manufacturers began large-scale production of vaccines 20 prior to the accumulation of data and results from the phase III trials, which was financially risky since 21 the vaccine product may have failed for either safety or efficacy reasons. 22

Design considerations for the primary endpoints of the phase 1-3 studies were especially 1 challenging. At a time when little was known about SARS-CoV-2 pathogenesis and disease, even less was 2 known about appropriate laboratory assays, with no validated tools to measure clinical outcomes. 3 Furthermore, study endpoints had to be defined based on limited data, with reliance on home 4 questionnaires (via Smartphone applications), home oxygen monitoring, and sampling kits that could be 5 sent by mail. Identifying a reliable and practical molecular target for primary end point analysis was 6 problematic, with several types of assays in development and insufficient data initially to know the 7 validity of a given test. In addition, the testing materials and reagents had to be both scalable and 8 available for all the clinical trial sites to maintain consistency across study procedures. How should a 9 clinical trial be designed with meaningful endpoints when knowledge is simultaneously building and 10 shifting the targets? As researchers grappled with this task, heterogeneous targets and different assays 11 were selected across different vaccine candidate trials, which later limited the ability to compare 12 outcome measures across vaccine products. 13

Recruitment and enrollment of participants, especially for phase 3 vaccine trials that require 14 30,000-40,000 participants, is often difficult even under the best circumstances. Unlike volunteers who 15 suffer from disease and seek trials for therapeutic benefits, volunteers for vaccine trials are healthy 16 individuals who may not have a direct health benefit from the study product (or may receive placebo) 17 and are exposed to potential harm from the product [11] . Identifying high-risk and willing participants 18 was problematic due to many of the issues described above, including staffing issues (resulting in 19 decreased recruitment efforts) and ability to maintain COVID-19 infection control practices, as well as 20 ensuring safe transportation and study flow for the participants. In addition, with misinformation 21 equally as viral as SARS-CoV-2 transmission and an overflow of data and publications (followed by an 22 unprecedented number of retractions and redactions) highlighted in the media [12] , overall trust in 23 science among the public was low. This directly impacted the trials and outreach efforts. Many of the 24

communities most affected by COVID-19 had concerns about COVID-19 vaccines and the trial process, 1 which limited participation [13, 14] . Another criticism of the trials involved the exclusion of special 2 populations, such as pregnant women, immunocompromised individuals, and children. Paradoxically, 3 these vulnerable populations are traditionally protected from phase 3 vaccine trials, although if the 4 vaccines are efficacious, these individuals eventually receive the vaccine without significant data to 5 support their use in these populations. Clinical trials for the pediatric population were initiated in March 6 2021 after the initial efficacy and safety signal in adults, with special considerations for dosing and side 7 effects [15] . However, for other special populations such as pregnant women and immunocompromised 8 individuals, the vaccines were mostly studied using observational cohort studies or real-world 9 evaluations, which often have many confounding variables. 10

Study procedures had to constantly adapt to the vicissitudes of the pandemic itself. The 11 protocols underwent multiple revisions to address new data, align with product and endpoint kit 12 availability, and react to shifting realities of the pathogen with new circulating variants. Amid these 13 adaptations, the participant's perceptions and expectations had to be considered as well. For example, 14 as seen in the mRNA-1273 and BNT162b2 trials, an early efficacy signal was seen less than six months 15 into the trial. At the core, trial participants must be protected and allowed access to the product if 16 shown to be efficacious; this is essential to maintain trust and transparency with the study participants 17 and the community. However, how should new information and knowledge be incorporated into the 18 study design, in a rationale and ethical manner, while maintaining scientific rigor? For most of the phase 19 3 efficacy trials, an unblinding phase was introduced mid-trial when this efficacy signal was confirmed 20 and the vaccine granted EUA by the FDA. This allowed for participants to remain in the study and have 21 additional data collected, although at the cost of losing the placebo-controlled value of the study [16] . 22

The EUA approval of the mRNA-1273 and BNT162b2 vaccines in December 2020 further influenced the 23

other vaccine candidate trials, manufactured by Janssen and Novavax. Many participants left these 1 studies in favor of obtaining an EUA approved vaccine, thus compromising the results from those trials. 2

The challenges affecting the COVID-19 vaccine clinical trials offers a unique opportunity to 4 reflect upon the clinical trial process. Many solutions were developed in "real-time" and reactionary to 5 immediate demands. However, a thoughtful exploration of these challenges can yield novel ideas and 6 approaches for future conduct of trials that will expedite regulatory and administrative affairs, engage a 7 more diverse range of participants, streamline the study visits, encourage data sharing and 8 transparency, and allow for flexible trial designs (Table 1) . 9 and Institutional Review Boards) to ensure prompt institutional approval on RCTs. Establishing this 23

infrastructure not only would allow for rapid initiation of trials, but also would allow for consistency 1 across vaccine candidate protocols and management. 2

Flexible and adaptive trial designs are critical to success, especially in the context of a new 4 pathogen. As described, overlapping the preclinical, phase 1, 2, and 3 was necessary to expedite 5 discovery and clinical testing, with some studies even incorporating all three phases into one protocol 6 (e.g., NCT04368728). However, future study designs should carefully determine appropriate transition 7 points; for example, determine the level and quality of data required to advance from phase I to phase 2 8 and 3. Traditionally, phase 2 has served to further refine and optimize the dosing and schedule. With a 9 more fluid and adaptive model, this important step could be better integrated into phase 1 10 investigations. Designing each phase with earlier interim analyses and ongoing safety evaluations 11 throughout may aid in acquiring relevant data in an expedited fashion. 12

Furthermore, while RCTs are the current gold standard, alternatives to placebo-controlled trials 13 should be considered. Human challenge studies [18], non-inferiority studies, and immunogenicity 14 studies (once correlates of protection (CoP) have been established) are more efficient and less costly. 15

Indeed, early analysis and identification of a correlate of protection is essential to allow for rapid 16 iteration and allow bridging of immunogenicity data, which would be useful to evaluate efficacy in 17 populations that were not strictly studied in the trials, determine efficacy quickly (and at less cost 18 compared to RCTs) of novel vaccine candidates, and investigate efficacy of vaccines against new variants 19 as they emerge [19] . 20

The regulatory processes as they relate to early data from clinical trials is also an important 21 component. The FDA's role is to evaluate the full body of evidence and determine safety and efficacy for 22 a novel product; full approval and licensure should not be expedited and thorough evaluation is 23 necessary to maintain this important safeguard. However, with the expeditious nature of the vaccine 24 A C C E P T E D M A N U S C R I P T clinical trials, the FDA utilized the EUA mechanism to rapidly enable availability of COVID-19 vaccines to 1 the public. By definition, the EUA allows for authorizations based on limited evidence, with 2 consideration of direct risks and benefits in the context of a public health emergency [20] . While the 3 EUA was overall beneficial and allowed for rapid distribution of COVID-19 vaccines earlier in the 4 pandemic, further refinements and criteria for EUAs should be addressed, with improved transparency 5

[21]. In addition, independent reviews by experts (e.g., the Vaccines and Related Biological Products 6 Advisory Committee) are essential to address decisions regarding the EUA and to provide a platform for 7 open communication with scientists and the public allowing increased trust and confidence in the 8 process by the public. Perhaps further expansion or increased frequency of these independent review 9 committees would aid the FDA in being able to iteratively reflect upon emerging data and reassess 10 interventions as pandemic conditions change (e.g, in response to variants of concern or waning of 11 immunity and considerations to deploy booster immunizations). Educating the public and clinicians on 12 the differences between EUA and full licensure is also warranted, since confusion about terminology can 13 lead to further misunderstandings and mistrust of vaccines and the regulatory processes that govern 14 approval [22] . 15

Visits should be streamlined favoring online consenting, telemedicine visits, use of Smartphone 17 apps for safety monitoring, and provision of home thermometers, oximeters, and testing kits for 18 participant's self-monitoring. Utilizing these technologies will not only reduce in-person visits and 19 directly address many of the logistical barriers described above, but also will use more cost-effective and 20 efficient tools, with the benefit of more fully engaging the participants [23] . However, with technological 21 adaptations, researchers should also evaluate strategies that are inclusive for elderly populations and 22 people with low health literacy who may not be as facile in e-Health tools. Other approaches are 23 warranted as well, which can both reduce the burden of on-site visits and expand access to trial 24

participation. For example, hiring personnel for home visits or partnering with home health agencies to 1 conduct research visits at alternative locations to the research site are important strategies to pursue. 2

Recruitment should reflect the population we serve. While central registries can be very 4 beneficial, involving local leaders as well as neighboring healthcare systems are crucial strategies for 5 success. By local outreach efforts to communities about vaccines in particular and research in general, a 6 more successful partnership can be developed and nourished. Strategies to improve communication 7

with the public is also essential to combat widespread misinformation. While sharing of information has 8 been conducted mainly through press release, timely manuscript write-up, transparent review process 9 and prompt publishing of papers (e.g utilizing pre-print servers, online posting after peer review or 10 reporting results on clinicaltrials.gov) should be prioritized. Tackling The speed and successes of the COVID-19 vaccine trials have been remarkable and have altered 2 the course of the pandemic. Clinical trialists and scientists have proven that the traditionally slow 3 machinery of RCTs is not necessarily warranted, and a safe and rigorous process can be achieved in a 4 more efficient manner. However, further refinement and novel strategies are necessary. New pathogens 5 will continue to emerge, and pandemics will continue to plague our global ecosystem. Investing in 6 preparedness is essential. Continuous investment in global research and training the next generation of 7

vaccinologists (e.g., through nationally funded vaccinology T32 training grants) is a priority. 8 Furthermore, with the COVID-19 pandemic particularly affecting under-represented minorities, we must 9 also focus on increasing diversity of our research staff and faculty and promote continuous training in 10 cultural competency. Through these actions, we will ensure safe, collaborative, and inclusive clinical trial 11 processes that will continue to advance our scientific knowledge. 

Estimating the cost of vaccine development against 2 epidemic infectious diseases: a cost minimisation study

Risk in Vaccine 4 Research and Development Quantified

Clinical features of patients infected with 2019 novel coronavirus in 6

COVID-19 Vaccine

Spikevax and Moderna COVID-19 Vaccine

COVID-19 Vaccination Program at One Year: How Many Deaths and Hospitalizations 17 Were Averted? Available at

Extraordinary diseases require extraordinary solutions

Contributing factors to personal protective equipment shortages 23 during the COVID-19 pandemic

SARS-CoV-2 vaccines in development

More than the money: A review of the literature examining healthy 26 volunteer motivations

An alarming retraction rate for scientific publications on Coronavirus 28 Disease 2019 (COVID-19)

COVID-19 Related Medical Mistrust, Health Impacts, and 30 Potential Vaccine Hesitancy Among Black Americans Living With HIV

Medical Mistrust and Perspectives on COVID-19 Vaccine Trial Participation and Vaccine Uptake in the 34 US

Evaluation of the BNT162b2 Covid-19 Vaccine in 36 Children 5 to 11 Years of Age

A strategic approach to COVID-19 vaccine R&D

SARS-CoV-2 Human Challenge 1

Studies -Establishing the Model during an Evolving Pandemic

Vaccine-Induced Severe Acute Respiratory Syndrome 4

Coronavirus 2 Antibody Response and the Path to Accelerating Development (Determining a Correlate 5 of Protection)

During Pandemics: Double-edged Swords

Communicating Effectively About Emergency Use 12 Authorization and Vaccines in the COVID-19 Pandemic

Using e-technologies in clinical 14 trials

SARS-CoV-2 mRNA Vaccine Attitudes as Expressed in U.S. 16 FDA Public Commentary: Need for a Public-Private Partnership in a Learning Immunization System. 17 Frontiers in Public Health 2021