VRC01 Selects Rare HIV Escape Mutations After Acquisition in Antibody-Mediated Prevention Trials

Abstract

Broadly neutralizing antibodies (bnAbs) show promise in HIV prevention, yet viral escape remains a challenge. In the Antibody Mediated Prevention (AMP) trials, the CD4 binding site (CD4bs) bNAb VRC01 blocked acquisition by VRC01-sensitive strains. However, its influence on viral evolution post-acquisition is not fully understood. Here we analyzed >12,000 HIV env sequences from 47 participants from the AMP trials, identifying VRC01-mediated de novo escape mutations in 8 of 26 VRC01-treated participants but none in 21 placebo participants. These mutations were found at very low frequency (<1%) in global viruses. Escape mutations, primarily located in the Loop-D and β23/V5 regions of Env, conferred cross-resistance to several CD4bs bnAbs, while more potent CD4bs bnAbs like N6 and 1-18 largely retained their activity. Our findings demonstrate that prophylactic VRC01 can select for viral escape after infection, underscoring the need for next-generation bnAbs with improved breadth and potency to enhance durability and efficacy of antibody-based HIV prevention.

Keywords: HIV; HIV VRC01 escape mutations; VRC01 resistance; antibody mediated prevention; broadly neutralizing antibodies (bnAbs).

Figure 1.. Identification of putative post-acquisition VRC01 escape mutations in individuals with single-lineage infections by VRC01-sensitive viruses (IC₈₀ <3 μg/ml).

A-D. Approach used to identify escape mutations within the 34 amino acid VRC01 epitope, comparing a VRC01 high-dose recipient (V703_2372, A and C) and a placebo recipient (V703_1750, B and D). A and B. Logoplots showing amino acid (AA) changes between time point 1 (TP1) and time point 2 (TP2). The bottom panel shows the TP1 founder lineage consensus, the top panel shows TP2 variant frequencies, and the middle panel is scaled to show variant AAs. Sequence counts are shown in parentheses. C and D. Phylogenetic trees of TP1 (green) and TP2 (blue) sequences. Adjacent highlighter plots display AA changes within the VRC01 epitope relative to the sequence closest to the TP1 consensus. E. Number of participants with VRC01 variant sites (VS) in the Africa Trial (purple) and Americas Trial (green). Compared to the placebo group, the treatment group (pooled) had more VS in the Africa trial (*p = 0.036 two-sided Fischer’s exact test), but not the America trial (p=ns). F. Change in VS per day pooled across trials and grouped by treatment: placebo (grey), VRC01 low dose (light brown) and VRC01 high- dose (burnt orange). A significant difference was observed between VRC01 high-dose group the placebo group (two-sided Mann-Whitney *p=0.030).

Figure 2.. Post-acquisition VRC01 escape mutations.

A. IC₈₀ titers of the parental pseudovirus and the corresponding escape mutants in the VRC01 high-dose and low-dose groups. No escape mutations were detected in the placebo groups. The frequency of mutations at time point 1 (TP1) and TP2 is indicated. <LOD: below the limit of detection. B. Location of escape mutations on gp120. Some mutations conferred escape (more than 3-fold difference from TF in IC₈₀, blue bars), and mutations that others had no effect on VRC01 sensitivity (less than 3-fold difference from TF in IC₈₀, black bars). Mutations are colored by trial group. C. A crystal structure of the gp120 (PDB ID: 4LST) (grey), the VRC01 epitope footprint (red) and the escape mutations (blue). D. Close-up images of the interaction between VRC01 (wheat color) and the HIV gp120 protein (grey), with residues of interest shown in red. E. Close-up images showing interactions between VRC01 (wheat color) and gp120 (grey) with residues of interest are shown in red. Escape mutations were modelled using the mutagenesis tool in PyMOL Molecular Graphics System, version 2.4.2, Schrödinger, LLC.

Figure 3:. Frequency and kinetics of VRC01 escape following HIV diagnosis (Day 0).

A. Database (8,452 filtered sequence alignment, https://www.hiv.lanl.gov/) frequencies of: wildtype residues (AA in the transmitted founder lineage, light grey), escape residues (blue), other (neither wildtype nor escape, dark grey). B to I. Frequency (%) at TP1 and TP2 of sequences harboring escape mutations. Two participants (V703_2141 and V704_1481) received their infusion 23 or 27 days prior to sequencing, while the remaining six received their last VRC01 infusion on the same day as sequencing (day 0). Number of sequences per time point is shown in Supplementary Table S2.

Figure 4.. Relationship between escape mutations and bnAb sensitivity.

A. Heatmap of IC₈₀ neutralization titers of 54 pseudoviruses (rows) against 10 bnAbs (columns), 8 targeting the CD4bs epitope (labelled in green at the top), two controls [bnAb targeting the V2 apex epitope (lavender) and the MPER (pink)]. Rows and columns are ordered using hierarchical clustering: the top dendrogram illustrates the clustering of the bnAbs, whereas individual viruses are listed on the right hand side, with TFL plasmids in pink and mutants in black. The left side dendrogram illustrates the grouping of viruses based on their neutralization resistance levels, with the most sensitive viruses clustering at the bottom of the heatmap (dark red and orange hues) and the more resistant ones at the top (light yellow and blue). Colored bands on the left indicate trial study group (Treatment arm). B. Heatmap showing all pairwise correlation coefficients across all pairs of bnAb neutralization titers (columns in the heatmap shown in 4A). Correlation coefficients were calculated using random effects generalized linear models (GLMs) accounting for the Env backbone as a random effect. White cells indicate no statistical significance, whereas statistically significant correlation coefficients are by colored cells, color coded from red (perfect negative correlation) to dark blue (perfect positive correlation).

Figure 5.. IC80 titers of parental sensitive pseudoviruses (PSV) and corresponding resistant clones to bnAb panel.

Pie charts (A-H) showing resistance profiles IC₈₀ (μg/ml), with color key at the bottom of panel G. A-B. Scatter plot showing parental PSV IC80 values against different bnAbs (A) and the effect of VRC01 escape mutations on each bnAb sensitivity (IC₈₀) (B). Bars indicate the median IC₈₀ (μg/ml), and 2-way paired Wilcoxon p-values denote whether they were statistically different compared to VRC01. C - H. Effect of single mutations (C,D,E), in different parental PSV backbones (D,E), and double mutations combinations (F, G, H) on bnAb cross-recognition. The alluvial flow links mutant PSV sensitivity (colored by IC₈₀ value) (left) to bnAb (right). The bnAbs are colored by potency (right), with the darker shades of grey showing the increase of bnAb IC₈₀/ potency as determine in A. The potency is ordered differently on each plot to reflect differential impact of the different mutations. All amino acid positions correspond to HXB2 reference HIV virus.