Secondary integrase resistance mutations found in HIV-1 minority quasispecies in integrase therapy-naive patients have little or no effect on susceptibility to integrase inhibitors

Abstract

The goal of this study was to explore the presence of integrase strand transfer inhibitor (InSTI) resistance mutations in HIV-1 quasispecies present in InSTI-naïve patients and to evaluate their in vitro effects on phenotypic susceptibility to InSTIs and their replication capacities. The RT-RNase H-IN region was PCR amplified from plasma viral RNA obtained from 49 HIV-1 subtype B-infected patients (21 drug naïve and 28 failing highly active antiretroviral therapy [HAART] not containing InSTIs) and recombined with an HXB2-based backbone with RT and IN deleted. Recombinant viruses were tested against raltegravir and elvitegravir and for replication capacity. Three-hundred forty-four recombinant viruses from 49 patients were successfully analyzed both phenotypically and genotypically. The majority of clones were not phenotypically resistant to InSTIs: 0/344 clones showed raltegravir resistance, and only 3 (0.87%) showed low-level elvitegravir resistance. No primary resistance mutations for raltegravir and elvitegravir were found as major or minor species. The majority of secondary mutations were also absent or rarely present. Secondary mutations, such as T97A and G140S, found rarely and only as minority quasispecies, were present in the elvitegravir-resistant clones. A novel mutation, E92G, although rarely found in minority quasispecies, showed elvitegravir resistance. Preexisting genotypic and phenotypic raltegravir resistance was extremely rare in InSTI-naïve patients and confined to only a restricted minority of secondary variants. Overall, these results, together with others based on population and ultradeep sequencing, suggest that at this point IN genotyping in all patients before raltegravir treatment may not be cost-effective and should not be recommended until evidence of transmitted drug resistance to InSTIs or the clinical relevance of IN minor variants/polymorphisms is determined.

FIG. 1.

Raltegravir and elvitegravir susceptibilities of clonal RT-IN recombinant viruses. Shown is a scatter plot of mean FC values for raltegravir (RAL) and elvitegravir (EVG) obtained from 345 recombinant clones and analyzed in 1 or 2 experiments, each performed with four replicate determinations in duplicate plates. SD bars are shown only for clones with mean FCs ± SD of more than the BCO. The squares indicate clones obtained from HAART-experienced patients, and the circles indicate clones obtained from HAART-naive patients; all patients were naïve to integrase inhibitors.

FIG. 2.

Integrase amino acid substitutions found in recombinant viral clones from patients naïve to integrase inhibitors with resistance to elvitegravir. Patient 15 was drug naive; patients 18 and 50 were HAART treated. Consensus B was used as a reference strain for the definition of mutations. The gray-shaded cells represent mutations detected both in the plasma population and in specific recombinant viral clones; the asterisks indicate quasispecies in population sequences. The black-shaded cells represent mutations detected only in specific recombinant viral clones; FC values ± SD that were more than the biological cutoffs are in boldface. ND, FC was not determined. The mutations found in plasma were the same as those we observed by sequencing the pool of clones (>200/patient) obtained for all 3 subjects.

FIG. 3.

Replication capacities of recombinant clonal viruses. (A) Bar graph of percentages (average and standard deviation) of virus produced in cell-free supernatant after 5 days of infection with 10,000 pg/ml of p24 gag Ag for each recombinant and control virus in C8166 T cells. The infectivities of recombinant viruses carrying INI resistance mutations (15_7 and 15_19, T97A; 18_13, G140S; 50_2, E92G), recombinant viruses without specific INI resistance mutations (47_19 and 18_14), two control mutant viruses (T66I and E92Q), and the molecular clone HXB2 were analyzed. Experiments were done in triplicate. (B) Bar graph of percentages (average and standard deviation) of virus produced in cell-free supernatant after 5 days of infection with 10,000 pg/ml of p24 gag Ag for each site-directed mutant and control virus in C8166 T cells. One experiment, representative of three done in triplicate, is shown. *, P < 0.05; t test.