Levofloxacin and Hypoglycemia

To the Editor–The combination of tenofovir, emtricitabine, and nevirapine is recommended as first-line treatment for antiretroviral-naive patients in some guidelines [1, 2], although it has never been evaluated in randomized, controlled trials. By contrast, other guidelines recommend this combination only as an alternative [3]. We conducted an open-label, randomized, controlled trial to compare elevations in the plasma cholesterol and triglyceride levels after receipt of coformulated tenofovir-emtricitabine plus either nevirapine (200 mg twice daily after a 14-day course of 200 mg once daily) or atazanavir (300 mg once daily) plus ritonavir (100 mg once daily; ATV/r). Antiretroviral-naive HIV-1–infected adults with a CD4+ T cell count <250 cells/mm3 (for women) or <400 cells/mm3 (for men) were enrolled in the study. In light of unexpected virological failure and evidence from the DAUFIN study [4], an unplanned ad interim analysis was conducted in April 2007, when 7 patients had been enrolled in each arm. Virological failure was defined as a reduction in the HIV RNA level of <1 log10 copies/mL at week 4 or viral rebound at week 12 after an initial decrease. Three women and 2 former injection drug users were randomized to the nevirapine arm, whereas only men and no injection drug users were randomized to the ATV/r arm. The median ages of patients were 36 years (range, 28–44 years) and 43 years (range, 33–75 years) for the nevirapine and ATV/r arms, respectively. Three patients in the nevirapine arm and 2 patients in the ATV/r arm had HIV RNA levels >5 log10 copies/mL and CD4+ T cell counts <200 cells/mm3. Two premature treatment discontinuations occurred in the nevirapine arm (because of severe hypertransaminasemia for one patient and generalized rash for the other). All patients experienced a >1-log10 reduction in the HIV RNA level after 4 weeks. However, 3 patients (all of whom were in the nevirapine arm) experienced virological rebound at week 12, despite an initial decrease in the HIV RNA level. Viral genotypes noted at the time of virological rebound were K65R, Y181C, and G190A; T69N, K101E, Y181C, and M184V; and K103N, Y181C, and M184V. None of these mutations were present at baseline (figure 1). The remaining 2 patients in the nevirapine arm and all 7 patients in the ATV/r arm achieved an HIV RNA level <1000 copies/mL at week 12. Self-reported adherence to treatment was 100% for all patients but 1, who reported an adherence rate of 93% at visual analogue scale at week 12 and subsequently developed virological failure, even though they had trough plasma concentrations of nevirapine that were greater than the threshold of 3.0 mg/dL [5]. In addition, the remaining patients who experienced virological failure had nevirapine concentrations that always exceeded the efficacy threshold. All 3 patients who experienced virological failure had high HIV RNA levels (>5 log10 copies/mL) and low CD4+ T cell counts (<200 cells/mm3) at baseline. Risk of early virological failure was significantly higher in the nevirapine arm than in the ATV/r arm for both the intent-to-treat (P=.021, by Fisher's exact test) and on-treatment (P=.045, by Fisher's exact test) analyses. The trial was prematurely stopped. HIV RNA levels and resistance profiles for the 3 patients who experienced early virological failure by week 12 of treatment with tenofovir-emtricitabine plus nevirapine. Pro, protease gene; RT, reverse-transcriptase gene. In our small population of antiretroviral-naive patients, use of tenofovir-emtricitabine plus nevirapine was associated with unexplained early virological failure. This observation supports a similar finding by Rey et al. [4]. Although some guidelines are more rigorous in suggesting preferred treatment regimens that have been validated by randomized, controlled trials [3], other guidelines are more permissive [1, 2]. Our results appear to support the former approach, suggesting that tenofovir-emtricitabine plus nevirapine should be avoided if antiretroviral combinations that have been “validated” by randomized, controlled trials are available. We await more information from ongoing, randomized, controlled trials of tenofovir-emtricitabine plus nevirapine, and intensive ad interim monitoring should be performed to assess the risk of premature virological failure. Potential conflicts of interest. G.C. has served as a consultant on advisory boards for Abbott, Bristol-Myers Squibb, Jansen-Cilag, Roche; has served as a speaker for Abbott, Boehringer-Ingelheim, GlaxoSmithKline, and Roche; and has received research and educational grants from Pfizer. C.T. has served as a consultant on advisory boards for GlaxoSmithKline and Tibotec; has served as a speaker for Abbott, Boehringer-Ingelheim, Bristol-Myers Squibb, Gilead, GlaxoSmithKline, and Roche; and has received research and educational grants from Bristol-Myers Squibb. All other authors: no conflicts.