Investigation of Integrase Inhibitor Resistance Mutations in gp41 in Clinical Samples

Mutations conferring resistance to HIV Integrase Strand Transfer Inhibitors (INSTI) can occur outside integrase, including in env gp41, in vitro, but it remains unclear whether these arise under INSTI selection in vivo.

Using a large database of clinically-derived HIV sequences linked to antiretroviral treatment histories, we sought to identify mutations in gp41 associated with INSTI exposure by comparing integrase and gp41 amino acid frequencies in INSTI-naïve versus INSTI-treated individuals. Gp41 was investigated because this region is routinely sequenced to assess fusion inhibitor resistance.

We identified 72 individuals with subtype B HIV for whom a genotypic INSTI resistance test performed after ≥3 months of INSTI exposure revealed susceptibility to all INSTIs (HIVdb v8.8; score<15), and for whom plasma INSTI concentrations were detectable by mass spectrometry. Gp41 sequencing was successful for 52 (72%) of these. The median INSTI exposure duration in this group was 20 (Q1-Q3:10-39) months, with raltegravir (>54%), dolutegravir (52%) and elvitegravir (23%) being the most frequently prescribed. Comparison of gp41 amino acid frequencies between this group and a comparison group of 1221 gp41 sequences from INSTI- naïve individuals using Fisher's exact test with Benjamini-Hochberg correction for multiple comparisons identified the gp41 substitution V182I (OR=3.75, p=2.2x10-4, q=0.01) as over-represented among INSTI-treated persons. When comparing gp41 sequences pre- and post-INSTI therapy in this group however, no evidence of INSTI-driven selection was observed at this position.

While off-target INSTI substitutions may arise in vivo, there is currently insufficient evidence to recommend expanding INSTI resistance testing to include Env.