Do you remember the spring? We were said to be ‘following the science’, but the scientific advice was kept secret. The minutes of Sage meetings were not published; even the ad hoc membership of the government’s advisory group was uncertain. Now that the minutes are available, they make grim reading. A vaccine would not arrive in any useful timeframe; there was no evidence to support the use of masks or other simple measures; the impending catastrophe would have to be endured as best we could. ‘Flattening the curve’ and building up a degree of immunity while shielding the vulnerable would have to see us through. We were told we were ‘well prepared’ for the virus. These bromides of March augured more than seventy thousand deaths in the UK, and one and a half million worldwide.
The virus’s genetic code has been available since January. We knew precious little about the subtleties of coronaviruses, but we did already know that they rip their way into host cells using a protein complex known as Spike. Block Spike, with a vaccine that raises antibodies to it, and you block the virus. There are plenty of ways to do this. You might use a killed authentic Sars-CoV-2 virus; or a different, live but innocuous virus with Spike bolted on; or the Spike protein plus an adjuvant (something that promotes an aggressive immune response); or the messenger RNA that codes for a piece of Spike, so your own cells make the protein. Two vaccines of this last type have proven blessedly effective. The Pfizer/BioNTech and Moderna vaccines are about 95 per cent likely to prevent symptomatic infection. To put this in context, we are content if the annual and very well understood seasonal influenza vaccine is 60 per cent effective. The deputy chief medical officer, Jonathan Van-Tam, compared it to your team scoring two consecutive goals in a penalty shootout. Penalties are too prosaic; in footballing terms these are goals only Maradona could have scored. But even that falls short of conveying quite how remarkable it is to have created a vaccine, using hitherto unproven technology, that’s 95 per cent effective against a novel virus – in less than a year.
The good news doesn’t stop there. The AstraZeneca/Oxford vaccine – in which Spike is bolted onto an innocuous chimpanzee virus – is also highly effective. Had it not been for the garish success of the mRNA vaccines, we would have been celebrating the headline figure of ‘70 per cent effective’. Behind that composite figure are two slightly different trials (one in Brazil, the other in the UK). Where two full doses of the vaccine were given it was found to be 62 per cent effective. Part of the UK trial gave participants an initial half dose of the vaccine, followed by a full dose; the control group was given a meningitis vaccine. This regime appeared to be 90 per cent effective. Bizarrely, it turned out to have been discovered by accident. AstraZeneca explained that it had meant to administer a full first dose, but ended up halving it thanks to ‘serendipity’ (what scientists would usually describe as a mistake). At the time of writing, the full details of the trial are not available and the 90 per cent figure cannot be relied on, but it is at least biologically plausible. A potential disadvantage of bolting Spike onto a different virus is that you will also raise antibodies against that virus. Perhaps the smaller initial dose delivers enough Spike, but not too much of the viral vehicle. This is no more than informed speculation at present; further investigations are underway to see if it is the correct explanation. It’s the reason the Russian vaccine, Sputnik V, uses two slightly different types of virus in its two doses. We are told Sputnik is 92 per cent effective. Russian scientists could be forgiven for being reluctant to deliver bad news – but it does seem that any vaccine which elicits a good response to Spike is going to work well enough.
Why might AstraZeneca’s 62 per cent be ‘enough’? There’s an even more welcome statistic than the headline figures: nobody in the vaccine arms of any of the trials has been reported to have suffered a severe Covid infection. This suggests that even for the unlucky people in whom the vaccines do not prevent a symptomatic infection, they at least work well enough to prevent a severe infection. We don’t yet know whether reducing symptom severity is the only thing the vaccines do. In principle, it could be that they turn what would have been severe infections into mild infections, and mild infections into asymptomatic infections. It’s important to know if infections are prevented as well as reduced in severity, because otherwise an asymptomatic or mild infection in a vaccinated person could be transmitted to an unvaccinated vulnerable person. This was looked at directly in the UK AstraZeneca trial: participants returned swabs every week to check for the presence of the virus. The Pfizer vaccine trial is looking at it indirectly, by checking for the presence of antibodies against the viral nucleoprotein. This would indicate an infection, as vaccination will only induce antibodies to Spike, whereas infection with the virus also directs antibody production to this structural protein – vital for the virus, but a target of dubious significance for our immune systems. The betting is that vaccines with 90 per cent or better efficacy would also prevent a significant number of asymptomatic infections. If so, we can finally talk of the end of the pandemic.
Are the highly effective Pfizer and Moderna vaccines ‘better’ than the AstraZeneca vaccine? We can’t draw that conclusion yet. The results reported so far aren’t directly comparable: the different trials used subtly different case definitions, and took place in different countries at different times. We will get more information when the full data are published. What’s certain is that the mRNA vaccines have a disadvantage both in price and – because of their relative instability – ease of deployment. The Pfizer vaccine needs to be stored at -70°C, which creates annoying if not insoluble logistical difficulties.
The Pfizer vaccine had the advantage of recruiting principally in the United States. Thanks to the astounding incompetence of the outgoing Trump administration, there was no lack of transmission to unvaccinated participants and so the Pfizer trial reached the necessary numbers first. As I write, it has just been approved for use by the MHRA (the UK’s national regulator). This is a highly credible independent authority which has appropriately expedited the paperwork without compromising standards. I am unimpressed by criticisms insinuating undue haste: it’s not axiomatic that everything British has to be useless; occasionally we contrive to get things right. I expect we will be jabbing the Pfizer vaccine into upper arms in the next few weeks. If I am offered it, I shall certainly take it, as I would any other vaccine that passed the stringent safety and efficacy requirements of the MHRA.
How many vaccines will there be? Given that antibodies to Spike seem to do the job, and all the vaccines in later stage clinical trials induce plenty of these antibodies, it seems we shall have a good number to choose from. Again, excellent news. We will need to vaccinate most of the population as fast as we can, so the more vaccines that can be manufactured in different places and in different ways, the better. Richer countries can easily afford the more expensive mRNA vaccines (at $19.50 or $32-37 a dose, still a ridiculous bargain compared to the economic damage that would otherwise be inflicted). AstraZeneca, commendably, is not aiming to make a profit on its vaccine ‘during the pandemic’ and will sell it at about $3 per dose. I hope other firms follow suit.
None of these vaccines will work, however, until or unless they actually get administered. A significant number of health and social care workers will be vaccinated soon, and a start will probably be made on the most vulnerable patient populations, but the great majority of vaccine deployment will be next year. Cue the metaphors. Boris Johnson has spoken of the ‘distant bugle of the scientific cavalry’, and of the ‘morale-boosting bugle-blasting excitement of Wellington’s Prussian allies coming through the woods on the afternoon of Waterloo’. Abstruse metaphors to many, but perhaps not so absurd to their intended audience. The abandonment of social distancing has become an alternative clarion call for some on the right of Johnson’s party. Not for them the discipline of the Iron Duke; instead, a reckless effort to prove their bravery by re-enacting the Charge of the Light Brigade (but putting other people’s lives at stake).
Until the vaccines arrive – like the cavalry, if you want – it makes sense to be extremely cautious and to suppress transmission of the virus as much as reasonably possible. More stringent social distancing measures would perhaps be unendurable if required for many months or years, but the very best case vaccine scenario is unfolding. Spreading infection now will cost lives, and livelihoods; in a few months’ time, when the most vulnerable have been vaccinated, we will be back to near normal. The new restriction tiers are all less stringent than the national lockdown that was in place in England until early December. A relaxation of the rules over Christmas is planned, with up to three households allowed to form a temporary ‘bubble’. If you are planning to do this, I urge you to exercise the utmost caution and resist the impulse to hug anyone. A seasonal catch-up from two metres away in a well-ventilated space would be wiser. Johnson’s joke that ‘’tis the season to be jolly careful’ is better than he usually manages, but it may also be better than his actual policy if people interpret the rules loosely.
A loose political interpretation of ‘the science’ we have allegedly been following is especially to be resisted. In marked contrast to the spring, Sage minutes are now published (after a short delay). This has many beneficial effects, not least that it’s obvious when politicians decide to ignore scientific advice. The advice is also open to wider scientific scrutiny. If the minutes from February and March would have been heavily criticised, the more recent deliberations are seriously impressive in their scope and attention to detail. Flashy but mistaken suggestions have been studiously ignored; errors (about the effectiveness of masks, for example) have been rectified; and the increasingly excellent vaccine data have been taken into account. The press briefings from scientists, medics and politicians have been increasingly assured. Given all this, it was baffling that a second lockdown wasn’t implemented earlier, perhaps to coincide with half-term, as suggested by Sage and taken up by the Welsh government. In England, the November lockdown was delayed, and was as a consequence longer and more damaging. Early social distancing is key – until the vaccines arrive.
But is there a way to achieve something closer to normality before the vaccines are widely deployed? Probably. Other countries, such as South Korea, managed it. A key fact about this virus is that some people remain infectious but asymptomatic, and a lot of transmission clearly occurs before people develop symptoms. The only way to prevent transmission, therefore, without insisting on stringent social distancing measures for everyone, is by extensive testing and tracing. NHS Test and Trace was judged to have all but failed by September. It was overwhelmed with demand, then turnaround times lengthened to the point where it wasn’t doing anything useful. Sage documents suggested it was having a very marginal benefit despite its huge cost. There was also the Excel incident, when 15,841 positive cases dropped off a spreadsheet (it was a wildly out-of-date file format) and were not passed on for contact tracing. An enterprising pair of economists seized on this opportunity – a quasi-randomised intervention, courtesy of code from 1987 – to work out if the testing and tracing was doing any good. According to their analysis, it was surprisingly effective: more than a hundred thousand infections could have been prevented had the error not occurred.
In any case, there’s every reason to believe that Test and Trace is better now. Testing capacity has greatly increased and turnaround times have improved. New ‘lateral flow’ tests, which are less sensitive but give a result in less than an hour, have been piloted with some useful effect, most notably in Liverpool. Take-up wasn’t as high as hoped, and such enterprises always run the risk of targeting the worried well, but even so there’s been a dramatic reduction in cases. However, one-off testing of a self-selecting population will have at best a modest effect. Systematic approaches with regular testing are much more likely to work. A variety of such approaches have been employed. At the Francis Crick Institute we test everyone individually every week. It’s a big effort, but keeps our laboratories and our testing pipeline open. It’s possible to make this approach work on a larger scale. Cambridge University designed a pooled testing system for its students. Pools of five to ten students contribute swabs that are tested together every week. Students contributing to positive pools are then retested individually. Take-up has been high, and the effect has been striking. In the most recent week (23-29 November) just 0.03 per cent of students (3 out of 9329) tested positive – tenfold lower than at the beginning of term, and thirty-fold lower than in the general population. Other universities with similar programmes are also reporting excellent results. Universities across the UK are now working with the government to get similar approaches to work for their students and staff (I am directly involved in this). It’s important because although students are unlikely to be severely affected by Covid-19, they might spread the virus across the country – especially since they are more likely to have mild symptoms or none at all.
The end is in sight. Effectively deployed testing may be able to ameliorate social distancing until the vaccines arrive. We were woefully prepared for a coronavirus pandemic in March, but were another similar virus to emerge in 2022 we wouldn’t make the same mistakes. We should be wary of learning the wrong lessons, however. To have several highly effective vaccines for this horrible virus after less than a year is a quite astonishing achievement, among the greatest things that we – by which I mean both humanity in general and molecular biologists in particular – have ever accomplished. We’ve been skilful, but we have also been lucky. A Sars-CoV-2 vaccine turns out to be relatively easy to develop. The virus that causes the next pandemic may not be so forgiving.