Covid-19 vaccines: to delay or not to delay second doses

The importance of successful vaccination strategies in controlling the covid-19 pandemic cannot be overstated and should be vigorously endorsed. Equally critical is that vaccines’ proven to be effective in a particular dosing schedule are not altered without solid scientific support or evidence. Two covid-19 vaccines, (from Pfizer/BioNTech and AstraZeneca/Oxford) recently approved for emergency use in the UK have a defined time scheduled two doses for use. Due to the accelerating pandemic and a desire to maximise the numbers in the population to receive a first dose vaccine, the Joint Committee on Vaccines and Immunization (JCVI) has proposed changing the dose schedules by considerably extending time to the second booster dose. The proposal has been supported by the UK Chief Medical Officers (CMOs) who indicated there are vaccine shortages across the UK. This is disputed by vaccine manufacturers.

The Pfizer/BioNTech vaccine is the first human mRNA vaccine evaluated in a phase 3 clinical trial. The study was a well-designed, prospective randomised controlled trial with blinded endpoint assessment. [1] It was well powered, reporting on 37,706 individuals at 152 sites in six countries who received vaccine or placebo and had median follow-up of 2 months. The overall results showed an efficacy of 95% reduction in covid-19 cases at least seven days after the second dose, which was delivered to a scheduled 21-day interval between the 1st and 2nd dose. Sub-group analyses confirmed similar vaccine efficacy among subgroups including patient age.

The JCVI advice and the CMO’s decision to delay the second dose to between 4-12 weeks is not based on data from the trial, but on an assumption of what would have happened if the second dose hadn’t been given at 21 days. While assumptions can be useful for generating a hypothesis, alone they are not a sufficient reason to alter a known effective dosing regimen. 

There are also no data on how long a first single dose of the mRNA generated immunogen (the virus Spike protein) induces a clinically effective immune response, comprising T cells, B cells, and their memory cells. [2] A crucial point is that the Pfizer vaccine uses mRNA.  Non-replicating mRNA (basic structure used in the two covid-19 vaccines- Pfizer/BioNTech and Moderna) after injection into the body becomes quickly degraded by extra and intracellular enzyme systems (RNAases). Modifications have been introduced, to improve the delivery and survival of the mRNA vaccines, including as in the Pfizer/BioNTech vaccine, enveloping the mRNA in lipid nanoparticles (LNP-mRNA). How well the LNP-mRNA survives after a 1st injection and induces durable human immune responses is unknown. [3-5] Animal models using LNP-mRNA vaccines against the virus show that ~28 days after a 1st injection, correlates of virus neutralising antibody production fall off markedly (suggesting limited survival and stimulation by the vaccine mRNA and its’ encoded Spike immunogen). However, the specific anti-viral immune response was strongly boosted with a 2nd injection of vaccine. [6] This raises concerns that extending a 2nd injection out to beyond 28 days could compromise vaccine efficacy. Human studies are urgently needed, ideally before pursuing the delayed 2nd dose strategy. The time interval for a mRNA booster may be very critical for getting the best sustained immune response. mRNA vaccines have never been used in late stage human trials before 2020 and the only data we have are from the Pfizer/BioNTech and the Moderna (another mRNA vaccine) studies, which use different vaccines at different concentrations. 

Furthermore, given that the Pfizer/BioNTech vaccine is scheduled to be delivered to 3-10 million vulnerable adults or health and social care workers exposed to high levels of risk, the JCVI/CMOs appear to be advocating what could prove to be a major change with attendant clinical risk (eg less efficacy than that generated by Pfizer/BioNTech study). The MHRA has approved and international expert bodies have advised using the existing study dosing schedule. The Centre for Disease Control in the USA has stated that for both mRNA vaccines (Pfizer/BioNTech & Moderna) “The second dose should be administered as close to the recommended interval as possible” – i.e. 21 days and 28 days respectively. [7] The World Health Organization has also advised that the Pfizer/BioNTech vaccine should be given “according to the following schedule: a single dose followed by a second dose 21 days later.” [8]

Maximising coverage with the 1st dose as intended by the CMOs could come at increased risk to already high risk/priority groups.  At a minimum, if the UK remains intent on pursuing this time extension to the second dose, the following should be implemented: 

Firstly, the second dose should continue to be provided at 21 days until the MHRA and/or JCVI make the data on which the JCVI recommendation is based publically available for independent scientific review. 

JCVI’s hypothesis that 50-60% efficacy with one dose is better than 95% efficacy for half the number of individuals with the MHRA approved dosing is, as far as we know, based on an assumption. However, if following a single dose there was a substantial decrease in efficacy during days 21-84, then the number of lives saved and hospital admissions avoided may be less than with the 21-day second dose schedule, for which we have efficacy results. Just as critical, the secondary immune response from a delayed dose at 84 days and any impact on the duration of immunity is unknown. 

Secondly, if the delay in the second dose is implemented then rigorous RCTs comparing the 21-day and delayed second dosing schedule should be conducted to rapidly ensure evidence-based future vaccination policy.    

The vaccine used in the Oxford/AstraZeneca study is a different type of vaccine (viral vector DNA) for which there are prior data from other similar vaccines. [8,9] In these circumstances, there is a valid argument to support delaying the second dose of this vaccine if there is not sufficient supply of the vaccine and the balance of public health risks warrants this. If JCVI’s advocacy for a dose delay has been based on unpublished research data and current modelling, such data should be immediately made available in the public domain for adequate peer reviewed scrutiny.  

John FR Robertson, Professor of Surgery & Consultant Surgeon, University of Nottingham. 

Herb F Sewell, Emeritus Professor of Immunology & Consultant immunologist, University of Nottingham.

Marcia Stewart, Social Care professional & emeritus academic, De Montfort University.

Denise Kendrick, Professor of Primary Care Research and General Practitioner, University of Nottingham.

Raymond M Agius, Emeritus Professor of Occupational Medicine, University of Manchester, Manchester.

Declaration of interest: All authors are current or intended recipients of covid-19 vaccines. HFS has served on UK Medicines Commission from 2002-2006. He has a doctor as a family member.

Acknowledgement: Professor Sheila M. Bird OBE FMedSci FRSE Formerly Programme Leader at MRC Biostatistics Unit, Cambridge Institute of Public Health; for her helpful comments and discussions about prospective RCTs and delayed covid-19 vaccine doses (see reference 10).


(1) Polak FP, Thomas SJ, Kitchin N et al Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine NEJM 2020 DOI: 10.1056/NEJMoa2034577

(2) Sewell HF, Agius RM, Kendrick D, Stewart M  Covid-19 vaccines: delivering protective immunity: Evidence supports both T and B cell responses to the three leading vaccines BMJ 2020;371:m4838 

(3) Pardi N., Hogan M.J., Porter F.W., Weissman D. mRNA vaccines—A new era in vaccinology. Nat. Rev. Drug Discov. 2018;17:261–279. doi: 10.1038/nrd.2017.243

(4) Xu S, Yang K, Li R, Zhang L. mRNA Vaccine Era-Mechanisms, Drug Platform and Clinical Prospection. Int J Mol Sci. 2020 Sep 9;21(18):6582. doi: 10.3390/ijms21186582. PMID: 32916818; PMCID: PMC7554980.

(5) Chung JY, Thone MN, Kwon YJ. COVID-19 vaccines: The status and perspectives in delivery points of view [published online ahead of print, 2020 Dec 21]. Adv Drug Deliv Rev. 2020;S0169-409X(20)30282-9. doi:10.1016/j.addr.2020.12.011

(6) Lederer K, Castaño D, Gómez Atria D, Oguin TH 3rd, Wang S, Manzoni TB, Muramatsu H, Hogan MJ, Amanat F, Cherubin P, Lundgreen KA, Tam YK, Fan SHY, Eisenlohr LC, Maillard I, Weissman D, Bates P, Krammer F, Sempowski GD, Pardi N, Locci M. SARS-CoV-2 mRNA Vaccines Foster Potent Antigen-Specific Germinal Center Responses Associated with Neutralizing Antibody Generation. Immunity. 2020 Dec 15;53(6):1281-1295.e5. doi: 10.1016/j.immuni.2020.11.009. Epub 2020 Nov 21. PMID: 33296685; PMCID: PMC7680029.

(7) Site accessed 1st January 2021 last reviewed: December 30, 2020 Content source: National Center for Immunization and Respiratory Diseases)


(9) Voysey M, Clemens SAC, Madhri SA et al Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK Lancet 2020

(10) Bird S M, MHRA’s Public Assessment Report Authorisations for Temporary Supply (under regulation 174 of the Human Medicines Regulation); recommendations by Joint Committee for Vaccination and Immunization.