Detection of genotypically drug-resistant HIV-1 variants and non-B subtypes in recently infected antiretroviral-naive adults in Italy

Abstract

The treatment of a growing number of infected patients with numerous anti-HIV-1 compounds has raised the concern that drug-resistant viruses are increasingly being transmitted to newly infected individuals [1]. Cohort studies aimed at defining the prevalence of drug-resistant strains in untreated individuals have been reported in the past few months [2–6]. Those studies have indicated frequencies of transmission of drug-resistance ranging from 0 to 20% for reverse transcriptase inhibitors (RTI) and from 0 to 4% for protease inhibitors (PI) in the United States, Canada and Switzerland. In order to define the prevalence of the transmission of HIV-1 drug-resistant variants in Italy we analysed the HIV-1 reverse transcriptase and protease sequences from all subjects shown to seroconvert from HIV-1-negative to HIV-1-positive status between June 1996 and May 2000 at three different infectious disease clinics in Tuscany, central Italy (n = 116). Risk factors for HIV-1 infection were male homosexuality for 61 (52.6%), heterosexual contacts for 32 (27.6%), intravenous drug abuse for eight (6.9%), and unknown for 15 (12.9%) subjects. Genotypic resistance testing was performed by infrared sequencing with a LI-COR 4000L instrument [7] at a median of 2 months (range 0–24 months) since the documented seroconversion in the absence of any previous antiretroviral treatment. Reverse transcriptase and protease sequences were deposited at GenBank under accession numbers AF209213–AF209388. Mutations conferring resistance to RTI and PI were detected in the virus populations obtained from 15 (12.9%) and one (0.9%) of the subjects, respectively (Table 1). Twelve individuals (10.3%) were infected with virus containing zidovudine (ZDV) resistance mutations occurring most frequently at reverse transcriptase codons 41 (M41L), 67 (D67N) and 215 (T215Y/F). A mutation at codon 215 compatible with incomplete reversion from resistant to wild type (T215D/S) [8] was the only sign of pre-existing ZDV resistance in HIV-1 sequences from another two patients (nos. 35 and 113). In addition, a T215S substitution was detected in the HIV-1 pol gene from patient 11 in the presence of the M41L ZDV resistance mutation. Five (4.3%) patients harboured HIV-1 sequences with substitutions at codon 184 (three M184I and two M184V) responsible for resistance to lamivudine and, to a lesser extent, didanosine and zalcitabine. HIV-1 sequences from one of these patients (no. 49) also had a substitution at codon 74 (L74I) possibly involved in resistance to didanosine and zalcitabine. A K103N mutation conferring resistance to the whole class of licensed non-nucleoside RTI was found in the virus population from one (0.9%) subject only (no. 13). Likewise, HIV-1 sequences with key mutations responsible for resistance to PI (M46I and L90M) were identified only in subject 83. These mutations were accompanied by secondary substitutions expected to remodel the PI-resistant protease. Similar to other reports [2,4], secondary mutations at polymorphic sites (L10I/F, K20R, L24I, L33F, M36I, D60E, L63P, A71T, V77I) were present in 91 (78.4%) of the HIV-1 sequences independent of the presence of resistance to RTI.Table 1: Drug resistance mutations detected in 15 drug-naive newly infected subjects. The alignment of reverse transcriptase and protease sequences to the consensus sequences of all HIV-1 subtypes (http://www.ncbi.nlm.nih.gov/retroviruses/subtype/subtype.html) identified eight patients (6.9%) infected with non-B subtypes (one A, one C, one E, two F, and three G). Separate alignments of reverse transcriptase and protease gave consistent results in each case. Of these, only the one infected with subtype C (no. 81) had drug resistance mutations. The prevalence of drug-resistant HIV-1 variants in the newly infected Italian patients analysed was comparable to that recently reported in the USA, Switzerland and Canada [4–6]. Notably, the populations considered in those studies and in our work were quite similar in terms of the number of patients and the time elapsed between seroconversion and analysis. The much lower rate of transmission of HIV-1 variants resistant to RTI reported in another US study [3] may have derived from testing subjects at one year or more after the acquisition of HIV-1 infection, possibly allowing drug-resistant virus to revert to wild type in the absence of drug pressure. However, no evidence of transmission of drug-resistant HIV-1 was detected in a recent survey of intravenous drug abusers in Canada [2] analysed at a median of 3 months after seroconversion. Overall, these data suggest that different rates of transmission of drug-resistant HIV-1 variants may be detected when testing populations with different risk factors and durations of infection. Finally, we consider that subtype analysis should be performed while assaying for genotypic drug resistance in patients with both new and established infection in order better to define the circulation of different HIV-1 subtypes and its possible implications [9]. Laura Romanoa Giulietta Venturia Rebecca Ferruzzia Maria Letizia Riccioa Paola Corsib Francesco Leoncinib Assunta Vinattieric Laura Incandelad Pier E. Valensina,e Maurizio Zazzia,e