AstraZeneca’s Covid-19 vaccine less effective when given in bigger doses: Report
The news reached Sarah Gilbert Saturday evening that the Covid-19 vaccine she’s developed with AstraZeneca Plc appeared to work. But the University of Oxford professor had expected a key number: Was it more than 90% effective, as others have been -- or less?
Instead, when her colleague Andrew Pollard called with the results, he wanted to show her slides instead of simple figures. “I didn’t really understand why we would have to go through slides,” she recalled. “But then it became clear -- because it’s rather more complicated in our trial.”
That complexity has led uncertainty to swirl around one of the front-runners along with Pfizer Inc. and Moderna Inc. in the race for a shot to end the pandemic. Questions about the most effective dose of the vaccine, its safety record and the partners’ approach to testing it have cast doubt on whether the US Food and Drug Administration will clear it.
Early Monday, AstraZeneca and Oxford reported results after 131 trial participants in the U.K. and Brazil contracted Covid 19. The average efficacy of 70% prevention sounded good, within the expectations of analysts and above a 50% standard the FDA had set for Covid vaccines.
The study also found 16 severe cases, all of them among people who didn’t get AstraZeneca’s shot, according to Moncef Slaoui, the former GlaxoSmithKline Plc researcher who heads up the US’s Operation Warp Speed program. It was good news for Astra’s vaccine program that had been halted for six weeks in the US while regulators reviewed a serious adverse event that the company never gave details on. Oxford will report details of those adverse events in its efficacy study that it will submit for review in the coming weeks.
The puzzling part about Astra’s results was that bigger doses were less efficacious. The vaccine was only 62% effective in a group that got two full doses spaced about a month apart. But among about 2,700 people who got a half-dose followed by a full, the number rose to 90%.
Positive data from another vaccine of substantial efficacy with potential advantages in storage, transportation and affordability will be welcome, said Jesse Goodman, a former head of the FDA’s vaccines office who was later the agency’s chief scientist.
“What’s not clear yet is, is this effect a chance observation or does it reflect something that the higher dose may adversely affect immune response,” he said in an interview. “We need more details to begin to understand it.”
Geoffrey Porges, an analyst with SVB Leerink, was among the most critical voices, calling the data “premature and insufficient” in a note to clients and predicting the vaccine would “never be licensed in the US” The shares fell 3.8% in London.
AstraZeneca and Oxford are running trials of the vaccine in different parts of the world, and are currently only studying two full doses of the vaccine in their US study which is expected to enroll 30,000 volunteers. But the company is still recruiting participants and could add another arm to conduct further studies of the regimen using the half-dose, said Ruud Dobber, AstraZeneca’s executive vice president and president of biopharmaceuticals.
“Let’s be a bit more patient and see how the FDA will react before making such harsh statements,” he said in an interview with Bloomberg TV.
The U.K.’s Medicines and Healthcare Products Regulatory Agency has already started its analysis based on data it’s received in a rolling review, said Chief Executive June Raine, in a statement. Even the two full-dose regime’s lower efficacy number hit the World Health Organization’s benchmark, Chief Scientist Soumya Swaminathan said. Lower- and middle-income nations are looking to the shot, priced far lower than those of Pfizer and Moderna, as a way out of the pandemic.
But a lower dose producing a better result has many scratching their heads, including Gilbert herself. The researchers tested a regime with an initial half-dose to find the smallest amount that could still generate a strong immune response, Gilbert said, not to boost effectiveness.
“I was surprised,” she said in an interview. “I really hadn’t expected that.”
Its superiority over two full doses could be good news, allowing the company to make more inoculations from the same overall volume of vaccine. That could be important with the need to vaccinate billions of people around the world.
The vaccine uses a harmless chimpanzee adenovirus – the cause of some common colds – as a vector that’s then inserted with the spike protein of the coronavirus to generate an immune response.
It’s theoretically possible that a full initial dose generated antibodies to the adenovirus vector itself, which might have limited the immune response to the coronavirus spike protein, Gilbert said. But her team measured the antibodies to the adenovirus in earlier studies and found only a small effect.
“I’m not really sure if that’s the full answer. We’ll look into it some more,” she said. “But it may be something subtler in terms of inducing a high-quality immune response by giving just the right amount of vaccine antigen at the first dose and then expanding it with the second dose.”
One possible explanation for the varying results is that the lower initial dose does a better job of getting through the body’s defenses, allowing the vaccine to infect cells and create the immune response, said Michael Kinch, a drug development expert and associate vice chancellor at Washington University in St. Louis.
Another theoretical reason is more complex: The immune system may become inured by the initial high dose of vaccine and then ignore the subsequent dose, a phenomenon known as tolerance or immune desensitization. Either one would have the potential to blunt the effects of the shot.
“The headline number of 90% might ultimately prove to be right,” Kinch said, “but the dose-response inversion could be a sign for concern.”
The Oxford team is still analyzing its data to better understand how the vaccine works. Gilbert said she’s confident the higher efficacy number will roughly hold up when more data comes in.
“It could go up a bit or down a bit,” she said. “I don’t think it’s going to massively change, but it may diverge from that initial 90% in either direction when we get to a slightly larger analysis.”