• Slide 452: CCR5 D32 Deletion and Response to HAART in HIV-1-Infected Patients (Authored by S. Guerin, L. Meyer, I. Theodorou, F. Boufassa, M. Magierowska, C. Goujard, C. Rouzioux, P. Debre, J.F. Delfraissy, and Seroco/Hemoco Study Group. INSERM U292, Le Kremlin-Bicêtre; Hosp. Pitié Salpêtrière, Paris; Hosp. Bicêtre, Le Kremlin-Bicêtre; and Hosp. Necker, Paris, France)
  Click here to view the original abstract

• Slide 456: Randomized Trial of Continued Indinavir (IDV)/ZDV/3TC vs. Switch to IDV/ddI/d4T or IDV/ddI/d4T + Hydroxyurea in Patients with Viral Suppression (Authored by D. Havlir, P. Gilbert, K. Bennett, A. Collier, M. Hirsch, P. Tebas, E. Adams, D. Goodwin, S. Schnittman, M.K. Holohan, D. Richman, and the ACTG 5025 Study Team. NIAID ACTG, Bethesda, MD; Glaxo Wellcome, Res. Triangle Park, NC; and Bristol-Myers Squibb, Wallingford, CT)
  Click here to view the original abstract

• Slide 457: Protease Inhibitor Class-Sparing Maintenance Therapy with Abacavir (ABC) + Lamivudine (3TC) + Zidovudine (ZDV) in Patients with Long-Term Suppression of HIV-1 RNA (Authored by M. Opravil, S. Yerly, A. Lazzarin, H.J. Furrer, J.P. Chave, P. Vernazza, E. Bernasconi, M. Battegay, C. Python, L. Perrin, B. Hirschel, and the Swiss HIV Cohort Study. Univ. Hosp. Zurich, Geneva, Bern, and Basel; Cantonal Hosp. St. Gallen and Lugano; Lausanne; Glaxo-Wellcome; Switzerland; and S. Raffaele Hosp., Milano, Italy)
  Click here to view the original abstract

Slide 452

People who have the delta 32 deletion in one of their two genes encoding for the chemokine receptor CCR5 have been shown to progress more slowly to AIDS and death than those who have wild-type CCR5. A group of French researchers analyzed the effects of this deletion on response to protease inhibitor-containing antiretroviral therapy.

166 patients who were antiretroviral therapy naive and starting a PI-based regimen were included. Patients were considered to have an adequate viral load response to therapy if they achieved a viral load <500 copies/mL or if viral load declined by two logs; an immunologic response was measured by an increase in CD4 cell counts by at least 50 cells. At months 6 and 12 of treatment, 82% of the 22 patients with the delta 32 deletion had a virologic and immunologic response, vs. 49% of the patients with the wild type CCR5 (p = <0.05). By multivariate analysis, the presence of the delta 32 mutation was an independent predictor of successful therapy, as were fewer prior regimens before starting the PI and choice of a protease inhibitor other than saquinavir.

This study adds to a body of data supporting an overall more favorable course for patients who have the delta 32 mutation. It is likely that testing for the presence of this mutation will become incorporated into clinical trials in the future.

Slide 456

Given the high rates of virologic failure seen in clinical practice, it is important to explore the strategy of preemptive intensification of antiretroviral therapy even for patients with viral loads below the level of detection. Diane Havlir presented the results of such a strategy from ACTG trial 5025.

Patients were eligible if they had been on AZT/3TC/indinavir for more than six months, had a viral load <200, and a CD4 cell count >200. Furthermore, they needed to be naive to d4T or ddI. They were then randomized to switch the nucleosides to d4T/ddI/hydroxyurea (HU), or d4T/ddI alone, or to continue AZT/3TC. Indinavir was continued in all patients. A dose of 600 mg twice a day of HU was used. The planned study endpoints were time to either viral load >200 or toxicity.

At entry, the mean CD4 cell count was 617, and 83% of the patients had viral loads <50 copies. The study began in November 1998, and was terminated prematurely in September 1999 due to unacceptable toxicity in the HU group. Specifically, the HU group reached 14 toxicity endpoints (mostly pancreatitis, neuropathy, and elevated LFTs), compared with six for ddI without HU and two for continued AZT/3TC. Of particular importance, there were two deaths related to pancreatitis in the HU group; a third death occurred secondary to a stroke, but this patient also had pancreatitis. Antiviral activity between the three groups was similar, but CD4 cell counts dropped significantly in those receiving HU.

The results of this study suggest that the strategy of combining HU 600 mg twice daily with d4T/ddI increases the risk of toxicity without improving efficacy. Patients who are receiving ddI/HU should be monitored for closely for the development of neuropathy by questioning and physical exam, as well as have periodic blood amylase and lipase measurements.

Slide 457

Can a protease inhibitor-containing regimen be simplified by the substitution of an alternative agent for the protease inhibitor? In this carefully-conducted prospective study, the strategy of substituting abacavir (ABC) for the protease inhibitor is compared with continuing the protease inhibitor in patients receiving AZT/3TC as dual nucleosides.

To be eligible, patients needed to have a viral load result less than assay for at least six months on a protease inhibitor plus AZT/3TC regimen; furthermore the viral load had to be <50 at study entry. In an important screen for prior AZT resistance, plasma samples on stored specimens were checked for the 215 mutation, and patients were excluded if this was found.

The mean CD4 cell count at study entry was 543 for the PI group, and 585 for the ABC group; they had received the protease inhibitor containing regimen for 36 and 28 months, respectively. At 48-weeks follow-up, virologic failure had occurred in 5 of 79 patients who remained on the protease inhibitor, versus 9 of 84 who switched to ABC, a non-significant difference. CD4 cell counts did not differ between the groups. When viral isolates were sequenced from those who experienced viral breakthrough on ABC/AZT/3TC, evidence of resistance mutations to these drugs was found most frequently at positions 184, 215, 41, and 70. Cholesterol and triglyceride levels fell rapidly in those switching to ABC, and remained significantly lower than the protease inhibitor group throughout the study.

The results of this study suggest that a substitution of ABC for the PI can be done with relative safety -- at least for the short term -- in patients receiving PI/AZT/3TC, so long as there is no prior resistance to AZT. On a practical basis, absence of AZT resistance will most likely be the case in those who started their first antiretroviral therapy with a PI plus AZT/3TC and had never experienced a virologic rebound. Pending further long-term follow-up of virologic outcomes, such a strategy of ABC substitution should probably only be done in those with significant lipid abnormalities related to PI therapy.