Transmission of HIV-1 variants resistant to the three classes of antiretroviral agents: implications for HIV therapy in primary infection

Over the past 3 years, the transmission of multidrug-resistant HIV has been detected in few cases [1–4]. Recently, a trend towards an increase in the incidence of multidrug-resistant HIV from 0.4 to 5.8% in US cities was highlighted [5]. The slightly increasing frequency of this transmission can be related to the sequential use of several antiretroviral drug regimens and the incomplete suppression of viral replication. Consequently, the transmission of drug-multiresistant variants could impair the success of the management of antiretroviral therapy in newly infected patients. We report a new case of transmission of an HIV-1 variant with multiple mutations associated with resistance to protease and both nucleoside and non-nucleoside reverse transcriptase (RT) inhibitors. The index patient was a 54-year-old heterosexual woman, who had been tested negative for HIV-1 antibodies 9 months before visiting. She reported having had, 3 months before visiting, a preservative accident during sexual intercourse with her HIV-1-infected partner. One month later, she noted an episode of fever. At visiting time, in November 1999, two enzyme immunosorbent assays for HIV-1 antibodies (Axsym HIV1/2gO, Abbott GmbH Diagnostika, Wiesbaden-Delkenheim, Germany; Access HIV-1/2 New, Bio-Rad, Marnes La Coquette, France) were strongly reactive; Western blot was complete (New Lav Blot I, Bio-Rad); p24 antigen (VIDAS HIV P24 II, BioMérieux sa, Marcy l'Etoile, France) was undetectable; the plasma HIV-1-RNA level was 6.0 log10 copies/ml (COBAS Roche Amplicor HIV-1 Monitor test version 1.5, Roche Diagnostics, Meylan, France); the CD4 cell count was 597 cells/μl; and physical examination was normal. A diagnosis of primary HIV-1 infection was made. The patient insisted on the immediate initiation of antiretroviral therapy, and received zidovudine plus lamivudine (300/150 mg × 2/24 h by month), and nevirapine (200 mg × 2/24 h by month), which was continued without interruption for 15 months. Genotypic analysis in the RT and protease genes from plasma HIV-1 RNA by pol gene sequencing [6] was performed just before the initiation of therapy and showed multiple mutations associated with antiretroviral resistance to the three classes of drugs (Table 1).Table 1: Genotypic analysis of the drug-resistance mutations of HIV-1 from the index and source patients at baseline and during follow-up. After 15 months of triple therapy, a poor virological response was noted: the viral load did not fall below 2.98 log10 copies/ml (detection limit of the assay 200 copies/ml) despite good adherence to treatment. CD4 cell counts were constantly higher than 350 cells/μl, and no opportunistic infection occurred. Genotypic testing performed from the source patient's plasma HIV RNA 3 months before transmitting virus to the index patient showed the same genotypic resistance profile in the RT and protease genes that were later found in the index patient (Table 1). Genotypic analysis performed during a follow-up of 12 months showed only slight differences in both individuals. Phylogenetic analysis of the RT and protease genes indicated that the viral sequences from the source patient and index patient clustered together, with boostrap values of 100%. A high similarity (97.7%) within the RT sequences was found in the source and index patient's sequences (data not shown). This case report confirms previous cases of transmission at the time of primary HIV infection of multiple-drug genotypically resistant HIV-1, which showed an impaired effect of antiretroviral therapy on viral replication. For example, in one index case, the HIV-1-RNA level remained at 3.3 log10 copies/ml 7 months after the initiation of triple therapy [3]. In another case, there was only a slower decrease in the number of copies of plasma HIV-1 RNA, contrasting with the undetectable levels of HIV RNA within 12 weeks after starting treatment in 36 other patients treated with similar drug combinations [1]. Conversely, another study [7] reported the case of a patient who chose to defer therapy: at month 5 after diagnosis, the multidrug-resistant strain was replaced by a more susceptible population of viruses; the HIV-1-RNA level that had a spontaneous decrease showed a 10-fold increase; an increase in the CD4 cell count of 67 cells/μl (17%) was noted between baseline and month 5, suggesting that the absence of treatment had no marked deleterious effect. It thus remains unclear whether altering treatment to replace a poorly fit resistant virus population by a replication-competent wild-type virus is a relevant therapeutic strategy in the setting of primary HIV infection. The implications of these data for the management of antiretroviral therapy in patients who acquire multidrug-resistant HIV-1 are unknown, as well as the clinical, virological and immunological evolutions in such patients. Therefore, sequential follow-up in the same situations either in the presence or in the absence of highly active antiretroviral therapy pressure is needed. Philippe Colsona Isabelle Ravauxb Nouara Yahia Christian Tourrèsa Hervé Gallaisb Catherine Tamaleta