South Africa's Astra Vaccine Snub Raises Serious Questions

Sam Fazeli, a Bloomberg Opinion contributor who covers the pharmaceutical industry for Bloomberg Intelligence, answered questions about the implications of South Africa’s decision this week to temporarily halt the rollout of the AstraZeneca Plc-Oxford University vaccine after a trial showed the Astra shot had limited efficacy against a new variant of the virus that emerged there. In the meantime, the country will speed up its supply of shots from Johnson & Johnson and Pfizer Inc.-BioNTech SE. The conversation has been edited and condensed.

What’s the rationale behind this move? Isn’t the Astra shot effective against the virus?

While AstraZeneca’s vaccine was found to offer good efficacy against previous forms of the virus and even a more transmissible variant that emerged in the U.K., it showed only 10% efficacy in a relatively small trial of 2,000 subjects against mild and moderate disease caused by the B.1.351 variant first found in South Africa. The data wasn’t able to show whether the Astra vaccine protects against severe illness from this variant, though chances are it might. Johnson & Johnson’s vaccine did show 85% efficacy against moderate to severe disease in a much bigger trial, in which most of the Covid cases arose from the variant. So presumably that’s why J&J’s vaccine is preferred over Astra’s in South Africa.

So the data on the J&J shot and the B.1.351 variant from South Africa is still good news, right?

Well, yes and no. An 85% efficacy against severe or critical disease is great, but it still means that 15% of patients did develop more serious Covid-19. This compares with close to 100% efficacy for vaccine trials conducted in other geographies where B.1.351 was not prevalent. Also, the efficacy of the J&J vaccine was 57% against moderate to severe disease. When you overlap those two, it suggests that a much lower efficacy against moderate disease dragged it from 85% to 57%. While we don’t know whether its efficacy would be as low as AstraZeneca’s vaccine when only looking at mild-to-moderate disease, the data tell us it’s not going to be great. Moreover, this leaves us with potentially two vaccines that leave at least some people susceptible to severe disease while they are not preventing mild disease, and by inference, a higher risk of viral transmission.

What risks does this pose?

Continued transmission with even mild to moderate disease brings two possible risks with it. One is that this allows the virus to continue to mutate and evolve more vigorously than if the vaccines had prevented mild disease, as they do for the non-variant versions. There is also the risk that there will be a higher rate of “long Covid” from infections with the B.1.351 variant than non-variant viral infections. I say this because there is probably a correlation between disease severity and long Covid, though this still needs to be proved.      

Can’t vaccines be adjusted to target a new virus variant?

This is true. I have often written about how fast the vaccines using messenger RNA technology from Pfizer Inc.-BioNTech SE and Moderna Inc. — as well as others from J&J, Astra and Novavax Inc., which use different vaccine methods — can be adapted to any new variant. That is not the problem. And some good news has just been published by the Fred Hutchinson Cancer Research Center showing that people who were previously infected with either an older version of the virus or the B.1.351 variant and then vaccinated with the Pfizer-BioNTech or Moderna vaccines responded equally strongly in the form of generating high antibody levels. The key finding was that the vaccination induced antibodies that could neutralize a virus that was engineered to look similar to the B.1.351 variant, telling us that the immune system is not “blind” to this variant. But this does not tell us that future new vaccines will be as efficacious as those developed against non-variant virus. It is possible that a new-variant vaccine may run afoul of “original antigenic sin” – that is, even as the immune system gets stimulated by the new vaccine booster shot, it may actually produce a lot of antibodies against the original version of the virus. This is all theoretical, but we can’t just assume everything will go our way.

Presumably there will be more variants. How can we stay ahead of them?

To keep one step ahead of this virus, we have to vaccinate as fast as possible to get that 85% efficacy that the J&J vaccine showed against severe disease in our communities; while not bulletproof, it’s still strong protection against the worst Covid-19 outcomes. We also need to keep testing and routinely conduct full genomic analysis at a much bigger scale on samples from around the globe, not just the U.K. and South Africa, where the most troublesome variants were first spotted. This way, we can identify outbreaks of B.1.351 or other new variants before they have taken root in the community and it’s too late. Last, we should develop the next round of vaccines as fast as possible and determine their potential for offering protection as early as possible so we can adjust our expectations. This may sound a little like doom and gloom, but it’s not. We do have vaccines that will hopefully significantly reduce the burden of the disease on our societies and health-care systems. What I am saying, though, is that we cannot just assume that science will forever beat biology and evolution and that we should price that into our expectations. 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Bloomberg Opinion provides commentary on business, economics, politics, technology and markets.

©2021 Bloomberg L.P.

BQ Install

Bloomberg Quint

Add BloombergQuint App to Home screen.