COVID vaccines in HIV-Positive recipients: Need for more data from the Novavax South Africa study and with other vaccine platforms

The Jan. 28 Novavax press release that included preliminary results from its COVID-19 vaccine Phase 2b trial in South Africa appears to contain a significant difference in efficacy between volunteers who were HIV-positive and HIV-negative. If verified, then the implications warrant sustained attention for this protein-based adjuvanted COVID-19 vaccine, as well as other vaccine platforms e.g., mRNA, adenovirus-vectored, whole inactivated, and DNA, in terms of efficacy in HIV-positive persons.

The Novavax protocol for this trial in South Africa stated that up to 240 volunteers who were HIV-positive would be included. These volunteers would be “receiving highly active antiretroviral therapy (HAART) and have been using the same regimen for at least 8 weeks before screening . . . Medically stable at screening, as determined by the investigator, and free of opportunistic infections in the 1 year prior to first study vaccination…and have a HIV-1 viral load < 1000 copies/mL within 45 days of randomization in the study”.

The Novavax press release Jan. 28 stated (my boldtype added):

In the South Africa Phase 2b clinical trial, 60% efficacy (95% CI: 19.9 – 80.1) for the prevention of mild, moderate and severe COVID-19 disease was observed in the 94% of the study population that was HIV-negative. Twenty-nine cases were observed in the placebo group and 15 in the vaccine group. One severe case occurred in the placebo group and all other cases were mild or moderate. The clinical trial also achieved its primary efficacy endpoint in the overall trial population, including HIV-positive and HIV-negative subjects (efficacy of 49.4%; 95% CI: 6.1 – 72.8).

Thus, only 6% of the study population was HIV-positive. Yet, when this 6% of the population was included with the 94% HIV-negative volunteers, then the vaccine efficacy was decreased to 49.4% from the 60% vaccine efficacy reported in the HIV-negative volunteers.

For such a small (6%) part of the study population to bring the overall vaccine efficacy down so much, from 60% to 49.4%, means that the vaccine efficacy in this 6% of volunteers who were HIV-positive must be much lower than the overall 49.4% efficacy.  

While some might argue the numbers are too small to warrant attention, I would argue the opposite and that attention must be focused on all COVID-19 vaccine platforms and efficacy in HIV-positive populations, and other immunocompromised populations. In addition, vaccine efficacy is likely to be lower in more immunocompromised HIV-positive persons e.g, excluded from this trial because they were not on HAART, or viral load was not < 1,000 copies/mL.