Novavax has shared prelim efficacy data by press release:
ir.novavax.com/news-releases/…
90% overall, 50% in S Africa (60%) in HIV-.
But there's more...
ir.novavax.com/news-releases/…
90% overall, 50% in S Africa (60%) in HIV-.
But there's more...
@carlzimmer has written a nice story about it here:
Article makes very provocative statement - that previous exposure doesn't protect against new variant B.1.351 found initially in S africa. But where are data to support this?
https://twitter.com/carlzimmer/status/1354898936482455553
Article makes very provocative statement - that previous exposure doesn't protect against new variant B.1.351 found initially in S africa. But where are data to support this?
Press release is vague on this. It just says this.
This is v important because previous data have suggested that antibody neutralization of B.1.351 is lower for people w/ previous exposure (& absent for ~1/2), & vaccines provide lower but still substantial protection that most thought would protect against disease.
If this trial actually has data showing that previous exposure does NOT protect against symptomatic disease (the endpoint in this study), that would be qualitatively new information, & actual data should be shared asap.
If trial doesn't have data to strongly support this statement it shouldn't say this & @carlzimmer should revise article ASAP to remove this very strong claim.
Novavax released a presentation document but info there is also very vague (no data!):
ir.novavax.com/static-files/e…
h/t @BrianOFahey1
This is extremely important - @Novavax please provide actual data on this.
ir.novavax.com/static-files/e…
h/t @BrianOFahey1
This is extremely important - @Novavax please provide actual data on this.
A couple other points:
Lower efficacy in S Africa could be due to variant B.1.351 in S Africa, or due to other differences b/w UK & S Africa trials (age of pop, etc.).
Lower efficacy in S Africa could be due to variant B.1.351 in S Africa, or due to other differences b/w UK & S Africa trials (age of pop, etc.).
If lower efficacy is due to B.1.351, worrisome & efficacy against B.1.351 wouldn't match FDA guidelines b/c 95% confidence interval must exclude 30% efficacy & it doesn't: (95% CI: 19.9 – 80.1)
Here's a slide from a webinar by
@ShabirMadh w/ v crude data on reinfection. Simple math suggests comparison is:
2%=62 cases in N=3080 sero-
2%=26 cases in N=1300 sero+
But 2% could be 1.501% or 2.499% which would change result enormously.
@ShabirMadh w/ v crude data on reinfection. Simple math suggests comparison is:
2%=62 cases in N=3080 sero-
2%=26 cases in N=1300 sero+
https://twitter.com/kakape/status/1354922533997473798
But 2% could be 1.501% or 2.499% which would change result enormously.
If this holds up to proper analysis this would be extremely worrisome.
Caveat - it's not severe disease cases (too few to analyze), so previous infection might still protect against that. Need data to assess.
Caveat - it's not severe disease cases (too few to analyze), so previous infection might still protect against that. Need data to assess.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
