LONDON – South Africa has decided against using the 1 million doses of Astrazeneca plc’s COVID-19 vaccine already in the country to start the rollout of its national vaccination program, in favor of opening a phase IIIb study of Johnson & Johnson’s single shot product.

“Over the next few days, we expect to announce a plan to expedite vaccination using [J&J’s] investigational vaccine in health care professionals in South Africa,” said Glenda Gray, CEO of the South African Medical Research Council and chair of the country’s research committee on COVID-19.

With the vaccine rollout network now in place, the national vaccination program was due to start this month, initially immunizing around 1 million health care staff with the Astrazeneca vaccine.

That has now been put on hold after data publicized on Feb. 7 showed the Astrazeneca product provided “minimal protection” against mild to moderate COVID-19 infection caused by the B1.351 variant of SARS-CoV-2 first identified in South Africa.

The finding, based on a small study involving 1,749 volunteers, is markedly different from earlier analyses of the data carried out up to the end of October 2020, before B1.351 emerged. These found Astrazeneca’s vaccine was 75% effective in reducing mild and moderate disease 14 days after a single dose.

Since then, B1.351 has become the dominant variant at all clinical trial sites and 95% of all volunteers who became infected after receiving the second dose of the Astrazeneca vaccine had B1.351.

Efficacy went from 75% after one dose to there being “very little difference between the vaccine group and placebo,” said Shabir Madhi, professor of vaccinology and director of the Vaccines & Infectious Diseases Analytics unit at the University of Witwatersrand, principal investigator of the trial run in collaboration with Oxford University, the co-developer of the Astrazeneca vaccine.

“There was 2% lower risk, which is not statistically significant. We can’t say the vaccine protects against B1.351,” Madhi said.

He was speaking on Sunday evening at a webinar hosted by South Africa’s health minister, Zweli Mkhize, to outline new developments in the country’s COVID-19 vaccine acquisition and rollout program. Gray and Madhi presented in-depth data from the Astrazeneca and Johnson & Johnson COVID-19 vaccines trials in the country, while other experts commented on the implications for vaccines rollout.

All the experts agreed the most important factor in terms of public health benefit is that COVID-19 vaccines prevent serious disease, hospitalization and death.

The 1,749 volunteers in the Astrazeneca trial were all low risk, with an average age of 31 and few co-morbidities, meaning protection against moderate to severe disease, hospitalization or death could not be assessed.

The finding that it does not provide protection against mild to moderate disease caused by B1.351 does not preclude Astrazeneca’s vaccine being effective against severe disease. In addition to using a similar adenoviral vector to Johnson & Johnson’s vaccine, Astrazeneca’s product has been shown to induce T cells, which, said Mahdi, is “particularly important” in protecting against serious infections. More data are needed, he said.

Changes to vaccine rollout

In comparison to Astrazeneca’s timing, the South African arm of the phase III study of Johnson & Johnson’s single shot vaccine started in November 2020, after B1.351 had become dominant. Top-line data reported on Jan. 29 showed lower efficacy, at 57%, in South Africa, compared to the U.S., where it gave 72% protection against moderate to severe disease, and in Latin America, where efficacy was 66%.

“It was 57% effective, where nearly all cases were due to B1.351,” Gray said.

Across all geographies, Johnson & Johnson’s vaccine was 85% effective in preventing severe disease 28 days post vaccination. Efficacy against severe disease increased over time, with no cases post day 49.

Based on that, it is now planned to roll out the Johnson & Johnson vaccine as an expansion of the phase III study, leveraging the existing clinical trial infrastructure, but also administering it through the network of immunization centers that has been put in place for the national program.

Johnson & Johnson last week applied to the FDA for emergency use authorization; Gray said a rolling review of the product has started in South Africa.

As Mahdi noted, that leaves hanging the question of what to do with 1 million doses of Astrazeneca vaccine on hand in South Africa. “Might it still have some role to play? Might it be of some use in protecting against severe disease?” he said.

Salim Abdool Karim, co-chair of South Africa’s ministerial advisory committee on COVID-19, told attendees of the webinar the key question is to look at the data available for all vaccines in relation to B1.351 and decide how to move forward in that context.

Peer-reviewed clinical data are available for only four vaccines, of which Johnson & Johnson’s is the only one to have clinical data showing it is effective in preventing severe disease and hospitalization in people infected with the B1.251 variant. For the others, “roll out implementation studies” are needed, said Karim.

That signals a new approach to the national vaccination program, he said. Step one will involve testing efficacy of vaccines in 100,000 people to assess hospitalization rates. If acceptable, step two will be a wide scale rollout.

“We can still proceed with our rollout, but we need to do it wisely,” Karim said. Deployment of the stock of Astrazeneca vaccine is on “temporary hold” pending clinical efficacy information in step one.

In the face of the disappointing Astrazeneca trial news, Mkhize was forced to justify the decisions to procure the vaccine and then to decide not to use it. At the time of choosing, the Astrazeneca vaccine both looked to be effective and was going to be available first, he said.

“The vaccine program will proceed. Now the approach will be to do implementation studies with the vaccine rollout network we have already created,” said Mkhize.

The protocols are being drawn up, and over the next four weeks it is expected to start implementation studies of Astrazeneca, Johnson & Johnson and Pfizer Inc./Biontech SE’s COVID-19 vaccines.

South Africa also is in talks to procure Russia’s Sputnik V, China’s Sinopharm and Moderna Inc.’s vaccines, and will open negotiations with Novavax Inc., Mkhize said.