B.1.1.7: How many doses of the COVID-19 vaccine are needed?
The question of how many doses are needed actually turns out not to be such a simple one. Both the Pfizer/BioNTech and Moderna are shown to be efficacious after two doses. So right now, regardless of what anyone says, everyone needs two doses of the vaccine, until such time as a single dose proves to provide highly efficacious (good at preventing disease) and durable (long-lasting) immunity. I fully support that study being performed, and I think Professors Tufekci and Mina make a good case for it here (though admittedly I am unsure that the NYT Opinions was the best place for the proposal, as some have pointed out). Immediately, I can see many potential advantages for a single-dose series for COVID-19 vaccines:
One dose is less reactogenic than two (the second dose tends to be worse in this regard).
You don’t have to worry about patient compliance- they don’t need to return for a second dose in 3-4 weeks depending on their vaccine.
Even if a single dose is less effective than a two-dose series, you may end up winning in terms of public health because you can immunize twice as many people. A less effective vaccine given to more people can be much more useful for public health than an extremely effective vaccine given to fewer people. This will depend on the specific efficacy of a single dose series.
Professor Vineet Menachery reported this PRNT comparing neutralization of wild type SARS-CoV-2 and N501Y variants. There isn’t any significant difference.
Needless to say, it’s a seductive option, though it is subject to the caveat that the vaccines elicit robust and durable immunity with a single dose. Unfortunately, real world variables complicate this question quite a bit.
You have by now probably heard of a new variant that has emerged of SARS-CoV-2 called B.1.1.7 (which is not itself a variant but rather a lineage in the phylogenetic tree; the variant was called VUI 202012/01 for variant under investigation (sometimes VOC- variant of concern), year 2020, month 12, variant 01, but I guess that name is long and annoying).
Thus far, we don’t know whether or not it’s more virulent (I expect there isn’t any significant difference in the virulence; some have suggested that because it lacks a premature STOP codon in ORF8 and this protein appears to be involved in subversion of interferon responses and reduces expression of MHC Class I it should be less virulent; indeed I was able to find one study that suggested that deletions in ORF8 were associated with milder disease, though it has significant limitations because of its sample size. However, ORF8 is extremely prone to recombination and deletions, and appears to be dispensable for the virus. SARS-CoV-2 has many different proteins to subvert innate immunity and reduced expression of MHC Class I would not be very valuable for a virus anymore because this is exactly how natural killer (NK) cells detect and kill virally infected cells. As it is, there isn’t a strong selection pressure against lethality for SARS-CoV-2 because it is able to spread so effectively despite the substantial mortality it causes. But I guess we can be optimistic in that respect until data tell us otherwise).
Rich. 2018. Clinical immunology: Principles and practice. 5th ed. London, England: Elsevier Health Sciences. Figure 15.3 Antibodies in general recognize conformational epitopes because they see proteins in 3 dimensional space. Sometimes, the conformational epitope is the same as the linear epitope. Because T cells recognize processed antigens, they recognize linear epitopes only. Thus modeling that examines only linear epitopes without considering how proteins folds does not fully account for the immune response
The big problem is whether or not this variant is more transmissible. Firstly, though a word on transmissibility: people have a tendency to look at the infection fatality ratio of any number of infectious diseases and decide it’s not a big deal because it’s relatively low. This is profoundly ignorant. Firstly, from the perspective of public health, even the most lethal infectious disease in the world isn’t all that important if it is terrible at spreading. You absolutely don’t want to be the person to get it, but in terms of lives claimed, it’s not going to be very impressive. On the other hand, a disease which is rarely fatal but spreads very easily is a huge problem, as eloquently explained by Professor Adam Kucharski in this thread. The general consensus among the genomic epidemiologists seems to be that this variant of SARS-CoV-2 is more transmissible; some individuals have argued that the data can also be explained by a founder effect i.e. someone had this particular variant and was part of a superspreading event which lead to this variant becoming dominant, but right now we can’t really wait for the data to unambiguously decide this and so the responsible thing to do is to behave as though this is a more transmissible variant and respond accordingly. In this case this means keeping up all the same public health precautions we have been, except maybe taking them even more seriously.
Walsh EE, Frenck RW Jr, Falsey AR, Kitchin N, Absalon J, Gurtman A, Lockhart S, Neuzil K, Mulligan MJ, Bailey R, et al. 2020. Safety and immunogenicity of two RNA-based Covid-19 vaccine candidates. N Engl J Med. 383(25):2439–2450. Figure 4
Looking at Table 1, you might see all those mutations in the spike protein and become concerned about what this means for the protection from vaccines or from infection; in particular, N501Y (an amino acid change from asparagine to tyrosine at position 501, which is located in the spike protein receptor binding domain targeted by antibodies) was a concern. Firstly though, it makes sense that the spike protein would mutate more than other parts of the virus. It is a major target of antibodies and SARS-CoV-2 is under selection pressure to become more transmissible (all viruses are) and antibodies get in the way of that happening. Here I can actually put your mind at ease. Firstly, a plaque reduction neutralization test (PRNT) compliments of Professor Vineet Menachery shows that there isn’t a significant difference in the ability of antisera produced from other variants to neutralize N501Y, which is excellent news. Following infection or vaccination you make many different kinds of antibodies, and in general it’s not easy for viruses to evolve to escape them (exceptions to this include influenza and HIV). On top of this, mRNA vaccines also produce excellent CD8 T cell responses (well, Pfizer’s did anyway- Moderna’s did not seem to but there was a very strong Th1 response from their vaccine which would be used to license the CD8 T cells so I don’t think that matters much), which means that hiding from antibodies is not enough for a virus. Serimmune examined the collection of mutations to see if there were significant differences in the targeted linear epitopes (the kind of sequences that T cells recognize; B cells recognize conformational epitopes, formed when proteins fold) and also did not find any substantial differences, but this was done in silico (i.e. through computer modeling), so that needs to be taken with a grain of salt (in particular, SERA, the method they used to run these simulations, is unable to model conformational epitopes so it’s not entirely clear how the sum of all of these variants affects responses). We are most likely years away from SARS-CoV-2 evolving to escape the protection conferred by vaccines or previous infection.
Also just as an interesting point, the variant has a unique quirk on some PCRs which can make surveillance easier:
The VOC includes a deletion of six nucleotides in the S gene, which results in loss of two amino acids at positions 69 and 70 (Δ69-70) and has been previously described by another group to cause S gene dropout in commercial assays (Bal, A. et al. medRxiv, 2020).
In other words, a PCR which shows a positive result but no amplification of the S gene likely reflects this particular variant.
Anderson EJ, Rouphael NG, Widge AT, Jackson LA, Roberts PC, Makhene M, Chappell JD, Denison MR, Stevens LJ, Pruijssers AJ, et al. 2020. Safety and immunogenicity of SARS-CoV-2 mRNA-1273 vaccine in older adults. N Engl J Med. 383(25):2427–2438. Figure 2
So, what does this have to do with how many doses of a vaccine we need? Well for one thing, the US’s vaccine rollout is not going at all according to plan. At this point we have done roughly one-tenth the number of vaccinations we had hoped to do. Given the potential implications of the emergence of the VOC, some have suggested that for now we should attempt to give as many people as possible dose 1 of the vaccine, and then give dose 2 in the coming months as the vaccine becomes available. Professor Akiko Iwasaki elegantly lays out the case for this in this thread (you will note that Professor Iwasaki strongly advocated for the importance of receiving both doses in the recommended interval initially). I think Professor Iwasaki makes a strong case, but I have some concerns.
We don’t know very much at all about the protection of these vaccines from a single dose. The data from the clinical trials available to us are very limited as there are so few events and almost all participants received a second dose. I think one of the worst things that could happen would be to learn that the response from the first dose is not sufficiently anamnestic to produce a boosting effect with the second dose of vaccine given months later, and the second dose will function just like the first prime with a third dose needed at least 3 weeks later as a boost to yield robust immunity. It’s important to acknowledge mitigating factors here. In general, for virus-like particles and live attenuated vaccines, even a single dose of vaccine does produce anamnestic responses. In terms of the immune response they produce, they are very similar to live attenuated vaccines, and thus I am inclined to think that the concern about a total loss of immunological memory in the months following a single dose is not likely. Mouse studies examining a single dose of SARS-CoV-2 mRNA vaccines do suggest long-lived anamnestic responses which is excellent. Nonetheless, it would be great to have the data.
The data above from Pfizer/BioNTech and Moderna’s vaccines show low titers before the boosting dose in the elderly (look at day 21 for Pfizer and day 29 for Moderna). I don’t like the idea of giving them a single dose, and making them wait months to receive the second, except perhaps in the case that they have already had COVID-19 before receiving the first dose.
At this point I don’t have any significant concerns about antibody-dependent enhancement in COVID-19 (although I guess I better not count my emerging variants before they hatch). Nonetheless, a single dose series will not perform as well as a prime-boost regimen as planned, and we know the titers would be lower. Subneutralizing titers in other diseases are shown directly to be the cause of antibody-dependent enhancement. Hence there is a very remote but distinct possibility that giving people just one dose will open them up to additional harm.
Perhaps the strongest argument comes from @notdred:
I think what bugs me about the single dose/delayed dose discussion is that we finally have a course of action that’s unambiguously proven and we feel compelled to mess with it to make up for all our other failures in controlling the pandemic
Looking at the totality of the evidence available to me, I think it is likely that a single dose of vaccine confers a nontrivial amount of protection and most likely is not going to be harmful, and I do think that the strategy of a single dose for as many people as possible can be used to great effect. Yet this is not what the data indicate to be the best course of action, but of course there haven’t been enough data generated for us to accurately make a judgment in that regard. On one hand, we could do as @notdred points out quite accurately to be what we know will work, and say that in the event a second dose appears superfluous, we acted with the best evidence available. On the other hand, that sort of rings hollow if we learn we could have saved many more lives than we actually did. So for now, I want this single dose clinical trial immediately, and for now, I think everyone should still receive their vaccines in the 2-dose series that has passed muster in clinical trials.
References
Anderson EJ, Rouphael NG, Widge AT, Jackson LA, Roberts PC, Makhene M, Chappell JD, Denison MR, Stevens LJ, Pruijssers AJ, et al. 2020. Safety and immunogenicity of SARS-CoV-2 mRNA-1273 vaccine in older adults. N Engl J Med. 383(25):2427–2438.
Florko N, Goldhill O, Branswell H, Boodman E, Cooney E. 2020 Dec 29. Frustration over vaccine rollout grows as new variant reported in U.S. Statnews.com. [accessed 2021 Jan 2]. https://www.statnews.com/2020/12/29/public-health-experts-grow-frustrated-with-pace-of-covid-19-vaccine-rollout/.
Flower TG, Buffalo CZ, Hooy RM, Allaire M, Ren X, Hurley JH. 2020. Structure of SARS-CoV-2 ORF8, a rapidly evolving coronavirus protein implicated in immune evasion. bioRxivorg. doi:10.1101/2020.08.27.270637. http://dx.doi.org/10.1101/2020.08.27.270637.
Lederer K, Castaño D, Gómez Atria D, Oguin TH 3rd, Wang S, Manzoni TB, Muramatsu H, Hogan MJ, Amanat F, Cherubin P, et al. 2020. SARS-CoV-2 mRNA vaccines foster potent antigen-specific germinal center responses associated with neutralizing antibody generation. Immunity. 53(6):1281-1295.e5.
Levin AT, Hanage WP, Owusu-Boaitey N, Cochran KB, Walsh SP, Meyerowitz-Katz G. 2020. Assessing the age specificity of infection fatality rates for COVID-19: systematic review, meta-analysis, and public policy implications. Eur J Epidemiol. 35(12):1123–1138.
Mahase E. 2020. Covid-19: What have we learnt about the new variant in the UK? BMJ. 371:m4944.
Plotkin SA, Orenstein W, Offit PA, Edwards KM. 2017. Plotkin’s Vaccines. 7th ed. Elsevier.
Preliminary genomic characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a novel set of spike mutations. 2020 Dec 18. Virological.org. [accessed 2021 Jan 2]. https://virological.org/t/preliminary-genomic-characterisation-of-an-emergent-sars-cov-2-lineage-in-the-uk-defined-by-a-novel-set-of-spike-mutations/563.
Public Health England. Investigation of novel SARS-COV-2 variant Variant of Concern 202012/01. Gov.uk. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/947048/Technical_Briefing_VOC_SH_NJL2_SH2.pdf.
Rich. 2018. Clinical immunology: Principles and practice. 5th ed. London, England: Elsevier Health Sciences.
Tufekci Z, Mina M. 2020 Dec 18. Can We Do Twice as Many Vaccinations as We Thought? NY Times. [accessed 2021 Jan 2]. https://www.nytimes.com/2020/12/18/opinion/coronavirus-vaccine-doses.html.
Walsh EE, Frenck RW Jr, Falsey AR, Kitchin N, Absalon J, Gurtman A, Lockhart S, Neuzil K, Mulligan MJ, Bailey R, et al. 2020. Safety and immunogenicity of two RNA-based Covid-19 vaccine candidates. N Engl J Med. 383(25):2439–2450.
Young BE, Fong S-W, Chan Y-H, Mak T-M, Ang LW, Anderson DE, Lee CY-P, Amrun SN, Lee B, Goh YS, et al. 2020. Effects of a major deletion in the SARS-CoV-2 genome on the severity of infection and the inflammatory response: an observational cohort study. Lancet. 396(10251):603–611.
Zhang Y, Zhang J, Chen Y, Luo B, Yuan Y, Huang F, Yang T, Yu F, Liu J, Liu B, et al. 2020. The ORF8 protein of SARS-CoV-2 mediates immune evasion through potently downregulating MHC-I. bioRxiv. doi:10.1101/2020.05.24.111823. https://www.biorxiv.org/content/10.1101/2020.05.24.111823v1.full.pdf.
Zinzula L. 2020. Lost in deletion: The enigmatic ORF8 protein of SARS-CoV-2. Biochem Biophys Res Commun. doi:10.1016/j.bbrc.2020.10.045. [accessed 2021 Jan 2]. http://dx.doi.org/10.1016/j.bbrc.2020.10.045.
