The views and opinions expressed here are those of the authors and do not necessarily reflect the position of either Johns Hopkins University and Medicine or the University of Washington.
For perhaps a month, we had the naivete to think that with the high efficacy of both the Moderna and Pfizer mRNA vaccines, we were ahead of this virus. Sitting pretty, so to speak: 95% effectiveness in overall disease and nearly 100% efficacy in preventing severe disease. We just needed to pump out vaccines and all would be well. This illusion – I’ll now call it delusion – has abruptly ended. The bubble burst thanks to the scientific prowess of our South African colleagues, who had the foresight to set up a surveillance system in a very systematic, regionalized way to see how the virus was mutating. This is the way molecular epidemiology and pandemic control should be undertaken. They detected incredibly sudden changes in the virus’s mutational pattern – not single changes, but multiple changes in several areas of the spike protein.
These changes appeared in regions of the SARS-CoV-2 virus that we had not seen before. They are “dangerous” for our current vaccine strategies because the mutations occur in the same area of the virus that the Covid-19 vaccines train our immune system to look for: the spike protein that gives this novel coronavirus both its name and its distinct appearance. More specifically, these mutations are found in the Receptor Binding Domain (RBD), which defines the part of the spike protein where the virus attaches to our own cells and to which neutralizing antibodies are directed. Some of these mutations looked like they might not be accessible or resistant to the neutralizing antibodies that arise after natural infection or even vaccination. And there were changes in another area we call N-terminal domain (NTD) that we don’t really understand, but we know that some monoclonal antibodies ¬– shown to be among the most promised therapeutics developed to combat the virus – also likely work by binding to that region.
The South Africans reported these changes to the United Kingdom (UK) because the predominant strain in South Africa in the early parts of the epidemic was similar to the one in the UK, something one would expect with the large international travel between the two countries. This led UK molecular biologists to look harder at strains circulating in their country. They then found a new variant that was starting to sweep across the country and create a second wave that we now see with increased rates of transmission and increased rates of hospitalization. This variant had one of the mutations in the RBD part of the genome, the one that appears associated with increased replication in the nose and transmission to other people, but fortunately not the changes that alter the neutralizing antibodies.
Very quickly, scientists put two and two together and saw two things were happening: 1) the virus was evolving more rapidly than previously detected, with mutations that enhanced both binding to human cells and virus replication; and 2) these two characteristics were being detected in the same virus. The UK variant had a change in just one area of the RBD gene that increased the ability of the virus to attach to its receptor in the human body. Scientists were finding that the UK variant replicates faster in cells and has a higher rate of replication in the nose and subsequent rate of transmission. We had seen that behavior back in June when we had a variation in the U.S. where the original strain from Wuhan was overtaken by what we call the D to G614 strain; that’s what is mainly circulating in the U.S. This strain also increased attachment, but the F105Y mutation enhances these characteristics even more and the spread seems to be even greater. This UK strain is now called the B.1.1.7 strain, which has spread throughout Europe, and is being increasingly recognized in the U.S. The CDC predicts that this U.K. strain will be the predominant one in the U.S. over the next few months.
The South African variant, now called B.1.351 also has the increased transmissibility gene of F105Y, but in addition to the U.K. variant has a couple of nasty mutations. One, called E484K, reduces the protection conferred by antibodies that most people have after contracting COVID-19 and that are also induced by vaccination. This mutation has been seen in other unrelated strains, so it seems to be a frequent way the virus tries to escape from people who have acquired immunity. This means that people who have been infected with prior strains and developed immunity to those prior strains appear to be at risk for the virus strains that have the mutations from South Africa. We call this “escape from neutralization” or “adaptive immune escape” from the virus’s point of view. This is what RNA viruses do: influenza does this; HIV of course does this too.
In retrospect, we were a bit naïve about SARS-CoV-2, thinking that because this subgroup of coronaviruses edits its mutations more efficiently, we might escape from having lots of different strains. But natural selection is very powerful, as Mr. Darwin taught us. What we have now is more people having SARS-CoV-2 infection than ever before on earth; it’s ever increasing. And this is exacerbated by more interactions between people because we’re tired of social distancing; and by more people with past infections who have antibodies, thus providing selective pressure for the virus to further mutate and escape our immune system’s attack. So, we find ourselves in a situation where mathematically, this is not going to stop anytime soon.
What do we do about it? This week brought us data that says yes, we do need to manage this, but we don’t have to panic. We do really have evidence that our current tools are going to be okay, but we have to create strategies to get us back up to the optimal state. And, there’s a second important concept: vaccinating everyone globally is going to be necessary to stop the spread. What impacts one of us impacts all of us, including here in the U.S. Ongoing community spread anywhere with this virus means spread everywhere.
We now have hard clinical evidence that our current vaccines will work as well against the UK variant as they have for the variants circulating in the U.S. The best evidence of this is a 14,500-person clinical trial of the Novavax vaccine which showed a 90% efficacy rate in a UK study where the UK variant constituted 30% to 40% of the strains circulating at the time of the trial. The antibody concentration from this 2-dose vaccine is pretty similar to what we saw in the Moderna and Pfizer vaccines, and several groups have now shown that sera from people vaccinated with either of these vaccines also neutralize the UK B.1.1.7 isolate as well as the isolate from Wuhan. So, that’s good news.
More importantly, we got data from the Johnson & Johnson (J&J) vaccine trial out of South Africa just this past Friday. We were lucky. We didn’t know when we conducted the J&J trial that there would be marked viral strain differences between South Africa and the U.S. In fact, when I worked with the company to design the trial, the reason we went to South America and South Africa for an international trial was because J&J is a global company and they wanted a globally developed vaccine. My network was already working with them on their experimental HIV vaccines and we had great investigators in both South Africa and South America who were tackling the COVID-19 epidemic and wanted to help in developing a vaccine. We were conducting so many trials in the U.S., we felt we could distribute some of the work internationally and everyone would benefit. At the time, we didn’t realize how fortuitous this decision was. A month ago, after the South Africans noted that the B.1.351 variant was the main circulating strain in the country, we got nervous because South African scientists were reporting that the B.1.351 variant could not be neutralized by convalescent sera from persons infected with the earlier circulating strains of virus in South Africa. These data were confirmed by a number of U.S.-based laboratories.
Fortunately, this anxiety was alleviated with the recent announcement of the data from the trial. The J&J vaccine protected against hospitalization and death in 88% of the enrollees in South Africa – that’s 88%! Zero deaths in the vaccine group; six in the placebo group. Overall, the vaccine was 57% effective against moderate and severe disease. The efficacy number in the U.S. was 72%; in Brazil, 71%. So, yes, we do have an 18% to 20% difference in the effectiveness in the J&J vaccine. And I will add to that the Novavax vaccine, which showed 90% efficacy in the UK, showed 55% protection in South Africa. So, here too a reduction in efficacy to the South African variant using our current vaccines.
But to put this all in perspective, we must ask ourselves: What’s more important? Do we develop these vaccines to reduce the frequency and severity of us having sore throats and coughs? Or do we develop these vaccines to prevent us from getting hospitalized, or put on oxygen, put on a ventilator, or dying? I think all of us would take a vaccine that prevented us from dying, even though it might still mean a case of COVID-19 with a sore throat and a headache and some body aches for a couple days. I think I’d take that deal. I could elaborate, but I think that’s enough for one message. In my next blog, I’ll talk more about the J&J vaccine and where it fits in the vaccine response in the U.S. and globally. We need a bit more data to become public before we tackle this task. There are lots of opinions here and there’s nothing wrong with that. But it’s been a very good week. This past week has brought two new safe and effective vaccines – Novavax and J&J – into our world. That can be nothing but great news. As in all science, all good clinical trial outcomes lead to more good questions that we’ll need to solve to control this pandemic. To be continued