This paper provides further, strong support for the extended prime-boost interval adopted in the UK.

“When the virus that causes Covid-19, SARS-CoV-2, emerged, and vaccines against it started to be developed, some in the scientific community suggested that it was a brand new, never-seen-before virus, and consequently we had no idea about vaccine efficacy or effectiveness beyond what the published phase I, II, and III trials told us. This was not entirely true, of course. The virus is closely related to other coronaviruses that we know a lot about; and we have been researching vaccines for decades (and other coronavirus vaccines for over a decade). While we had not yet observed precisely how vaccines against SARS-CoV-2 would work, we did already have grounds to make predictions.

“We knew, for example, that longer prime-boost intervals generate greater immunity. With the human papillomavirus (HPV) vaccine, for example, the guidance says that if the second dose (of the two-dose regime) is given before 6 months after the first dose, it is unlikely to generate good-enough immunity, and it should be repeated no less than 6 months after the first dose.2

“The prime-boost intervals used in the phase III trials were likely chosen because – at the time of the decision – we did not know how effective a single dose would be (it’s much better than expected), and we needed to get the most vulnerable in society protected as soon as possible. The trade-off was that it was likely to induce poorer immunity than a longer interval, increasing the likely need for additional boosters.

“When we realised, from the phase III trials, how effective a single dose was going to be, it meant we could extend the interval, thereby allowing more people to get the protection from a single dose more quickly. Many vaccinologists also predicted that it would enhance the quality of the immune response.

“This paper goes some way to showing just that, albeit only (so far) in vitro – looking at laboratory test results. (It has not demonstrated a difference in real world protection – cases, hospitalisations, critical care admissions, or deaths, though the authors cite a paper (their reference 18) which has shown greater real-world effectiveness for the AstraZeneca vaccine with an extended prime-boost interval).3

“The paper, by very well established scientists at respected institutions, PHE’s National Infection Service and the University of Birmingham, compares immune responses in people over the age of 80, who received their second dose of the Pfizer-BioNTech vaccine at an interval of either 3 or 12 weeks.

“There were 175 participants. This is, of course, far lower than you would see in a phase III vaccine trial; but the outcomes to be observed would be captured on all participants, and this number is sufficient for the findings to be robust. (In a phase III trial, by contrast, you need to have enough participants to accumulate enough outcomes – cases etc – to be able to compare the groups. If a vaccine is effective, and particularly when the disease is not very common, you might need thousands of participants to collect a handful of cases to compare.)

“It is not entirely clear to me why the study chose to focus on people over the age of 80. As the authors say, there is concern that immunosenescence may lead to lower vaccine efficacy, and in particular to a shorter duration of protection in older people. But it may also be that we have been able to accumulate and analyse more data in older people, because the vaccine programme prioritised this group, so they were vaccinated first. I would not be surprised to see further papers extending this analysis to younger age groups as more data accumulate.

“The study measured antibody responses 5-6 weeks after the first dose of vaccine, (2-3 weeks after the second dose for participants receiving the shorter, 3-week prime-boost interval, when the antibody response is likely to have peaked); and again about 8 weeks later (about 10 weeks after the second dose of those in the shorter interval group, and about 2-3 weeks after the second dose for those in the extended interval group). (They also looked at antibodies to parts of the virus that are not included in the vaccines, to assess whether participants had been exposed to and become immune to the virus itself. Those with such immunity showed an even higher antibody peak, as you would expect.)

“As many vaccinologists would have predicted, the peak antibody response was considerably higher (3.5 times as high) in participants in the extended prime-boost interval group.

“I expect that the participants will continue to be followed up, to observe whether and how quickly antibody levels will decline over time. The authors refer to recent studies showing that, from 43 days after vaccination, antibody levels drop by half every 52 days. Starting from a higher starting point in the extended-interval group is likely to mean that it be longer before antibody levels drop too low to provide protection.

“Measuring the antibody response is pretty straightforward, but measuring the cellular immune response is much harder. The authors also used a test sometimes referred to as an IGRA (interferon-gamma release assay) test to evaluate cellular immunity in vaccine recipients. Understanding measures of cellular immunity is not my area of expertise; but I would be surprised if an IGRA test provides a complete picture of the complex cellular immune responses to vaccination. In particular, part of the cellular response is the formation of memory B and T cells, which permit a very rapid antibody response, and T-cell response to future exposures. I am not clear that an IGRA test tells you about such responses. Another important effect – at least with other vaccines we have studied more (and can assume is likely to apply equally to Covid-19 vaccines) – is that a longer prime-boost interval can provide greater breadth of immunity, which might enhance cross-protection against vaccine variants; and I am not clear that an IGRA test helps us understand this.

“As such, the authors’ findings that those in the shorter prime-boost interval seemed to have stronger cellular immune response is puzzling. I am not clear what to make of it, but wonder if it is a genuine real-world effect, or perhaps an artefact of, say, the timing of the blood tests. As they say, “It will be of interest to assess the relative induction of long lived plasma cells and memory B cells following each regimen in order to assess potential cellular correlates of antibody response.” I expect to see more, interesting, findings in due course.”

References

1.            Parry H, Bruton R, Stephens C, et al. Extended interval BNT162b2 vaccination enhances peak antibody generation in older people [preprint]. National Infection Service & University of Birmingham; Received 13 May 2021 2021.

2.            Public Health England. Chapter 18a: Human Papillomavirus (HPV). Immunisation against infectious disease ("The Green Book") 2019; 1-19. Available at: https://www.gov.uk/government/publications/human-papillomavirus-hpv-the-green-book-chapter-18a. Accessed 16 Jul, 2019.

3.            Voysey M, Clemens SAC, Madhi SA, et al. Single Dose Administration, And The Influence Of The Timing Of The Booster Dose On Immunogenicity and Efficacy Of ChAdOx1 nCoV-19 (AZD1222) Vaccine. 2021; https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3777268. Accessed 03 Feb, 2021.

‍

The data in this preprint, from a sizeable study of older people, suggest that delaying the second dose of the Pfizer BioNtech COVID-19 vaccine from 3 weeks to 12 weeks has the advantage of substantially enhancing the antibody response at the cost of slightly reducing the cellular immune response. Nevertheless, both regimens induce significant antibody and cellular immune responses and, when taken together with the emerging clinical efficacy data, suggest that there is no detriment in delaying the second dose of the vaccine.

“It is true that delaying the second dose of the vaccine delays the peak of both the both cellular and antibody response, suggesting that immunity in the 12 week period between the two doses may be suboptimal but, again, the clinical efficacy data suggest that the first dose confers substantial protection against serious disease and death.

“Overall, these data add considerable support to the policy of delaying the second dose of COVID-19 vaccine when vaccine availability is limiting and the “at risk” population is large. Longer term follow up of this cohort will help us to understand which vaccine interval will be optimal in the future, once the immediate crisis is over.