ABSTRACT
Many participants in HIV-1 vaccine trials, who have not previously been exposed to or vaccinated against HIV-1, display serum immunoglobulin antibodies that bind the gp41 region of HIV-1 envelope prior to vaccination. Previous studies have hypothesized that these pre-existing antibodies may be cross-reactive and may skew future vaccine responses. In 12 large studies conducted by the HIV Vaccine Trial Network (HVTN) (n=1470 individuals), we find wide variation among participants in the pre-vaccine levels of gp41-reactive antibodies as measured by the binding antibody multiplex assay (BAMA). In the absence of exposure to the gp41 immunogen, anti-gp41 IgG levels were temporally stable over 26-52 weeks in repeated measures of placebo recipients. The analysis revealed that the geometric mean of pre-vaccine anti-gp41 IgG response was greater among participants in South Africa compared with participants in the United States. With gene-level metagenomic sequencing of pre-vaccination fecal samples collected from participants in one trial (HVTN 106), we detected positive associations between pre-vaccine anti-gp41 IgG and abundance of genes from multiple taxa in the Eubacteriales order. The genes most strongly associated with higher baseline anti-gp41 IgG mapped to a clade containing Blautia wexlerae and closely related strains. In trials with vaccine products containing the full or partial portion of gp41 immunogen alongside a gp120 immunogen, we did not find evidence that individuals with higher baseline anti-gp41 IgG had different levels of anti-gp120 IgG after vaccination compared to individuals with lower pre-vaccine anti-gp41 levels (pooled estimate of standardized mean difference -0.01 with a 95% CI [-0.37; 0.34]).
