Novavax reports 43% efficacy against B.1.351 South African variant but negative impact in HIV positive participants

Simon Collins, HIV i-Base

On 5 May 2021, results from a phase 2/a/b study of the Novavaz NVX-CoV2373 vaccine reported 43% efficacy against the B.1.351 South African variant. Importantly, it also reported the results by HIV status, showing negative results in participants who were HIV positive.

The analysis included 2684 participants who were seronegative for SARS-CoV-2 at baseline and randomised (1:1) to vaccine or placebo injections. Of these, 94% were HIV negative and 6% were HIV positive, with results reported separately by HIV status. Although more than 4,300 participants were originally enrolled, one-third were later found to be seropositive for SARS-CoV-2 at baseline.

This was a generally young population at low risk of COVID-19 (median age 32, with only 4% > 65 years) and the primary endpoint was mild/moderate symptoms (rather than hospitalisation or mortality).

Overall efficacy, seven days after the second dose, was 49.4% (95% CI: 6.1 to 72.8), based on 15 vs 29 cases in the active vs placebo group, respectively. At the time of the study, national incidence of the B.1.351 variant was approximately 93%.

Among the HIV negative group, symptomatic COVID-19 was observed in 11 vs 27 participants in the active vs placebo groups respectively: efficacy 60.1% (95% CI: 19.9 to 80.1)

The corresponding efficacy in HIV positive participants, based on 4 vs 2 cases in active vs placebo groups respectively, was 52.2% (95% CI: −24.8 to +81.7). This showed a negative impact of the vaccine, although the numbers in this subgroup were small.


The study discussion notes that these results are preliminary and that the B.1.351 sequencing analysis was post hoc.

Also that the vaccine effects in the HIV positive group represented a relatively small fraction of the trial population and the study was not powered for efficacy results by HIV status.


Shinde V at al. Efficacy of NVX-CoV2373 covid-19 vaccine against the B.1.351 variant. NEJM. DOI: 10.1056/NEJMoa2103055. (5 May 2021).

This report was first posted on 9 May 2021.

Links to other websites are current at date of posting but not maintained.