Switch From Enfuvirtide to Raltegravir in Patients With Undetectable Viral Load: Efficacy and Safety at 24 Weeks in a Montreal Cohort

To the Editor: INTRODUCTION Enfuvirtide, approved in 2003, is the lone member of the antiretroviral class known as HIV fusion inhibitors. Two phase 3 studies, TORO 1 and 2, demonstrated enfuvirtide efficacy when combined with optimized background therapy.1,2 Enfuvirtide is administered as a twice-daily subcutaneous injection, but its use is limited by painful injection site reactions (ISRs). In the TORO studies, 98% of patients suffered from ISR, leading to treatment discontinuation in 4% of patients.3 The integrase inhibitor, raltegravir, was approved in October 2007 for use in salvage therapy for patients with virologic failure. The efficacy of this new class drug was evaluated in 2 phase 3 studies, BENCHMRK 1 and 2, which demonstrated the efficacy of raltegravir in treatment-experienced patients with few adverse effects associated with its use.4-6 Although only approved for use in patients with virologic failure, when raltegravir became available in Quebec, many patients virologically suppressed on an enfuvirtide-containing regimen switched their enfuvirtide to raltegravir. This retrospective study evaluates the experience of these patients. METHODS Patients were identified through systematic chart review. Inclusion criteria included a HIV-1 viral load (VL) <50 copies per milliliter at the time of a single substitution from enfuvirtide to raltegravir (no other changes were allowed in the regimen). Patients were included who underwent this switch from September 2007 through March 2008. Standardized data extraction forms were provided to 6 Quebec clinics. Demographic information, antiretroviral and genotype history, adverse effects, and laboratory data before and up to 24 weeks after the switch were extracted by coauthors via chart review. Changes in CD4, VL, and safety parameters at 24 weeks were analyzed by Wilcoxon signed rank test. RESULTS Twenty-eight patients were included from 6 difference sites in Montreal. Ninety-three percent (26 of 28) were male with a median age of 44 years (Table 1). Patient had received a median of 12 years of antiretrovirals [interquartile range (IQR): 10-14], and all patients had drug resistance mutations to at least 1 drug among the 3 classes of medications [eg, nucleoside reverse transcriptase inhibitor, nonnucleoside reverse transcriptase inhibitor (NNRTI), and protease inhibitor (PI)]. Patients had received a median of 25 months of enfuvirtide (IQR: 12-39). Few patients were on biojector, but most of them had to be administered enfuvirtide by injection.TABLE 1: Patient Characteristics and Measure of Biological Parameters At Each Time PointIn the background regimens, 27 of 28 patients were receiving a ritonavir-boosted PI (boosted darunavir in 11 of 28 patients, boosted tipranavir in 8 of 28, other boosted PI in 8 of 28), with 6 receiving a NNRTI (4 on etravirine, 1 on efavirenz, and 1 on delavirdine). The median baseline CD4 was 344 cells per microliter (IQR: 270-456). Twenty-four week data were available for 27 of 28 patients. Twenty-six patients remained with a VL <50 copies per milliliter. One patient had a VL of 56 copies per milliliter at 24 weeks; his concurrent regimen consists of emtricitabine-tenofovir with ritonavir-boosted darunavir. No significant changes, statistically or clinically, were observed in the CD4 counts (CD4 at baseline = 344, CD4 at week 24 = 345, P = 0.327). The CD4 cell count decreased in 9 patients (1-260 cells/mm3), increased in 16 patients (1-230 cells/mm3), and was unchanged in 2 patients. However, the CD4 percent was stable and did not correlate with the fluctuations in the absolute CD4 cell count. Raltegravir was well tolerated with no new clinical or laboratory events reported. ISR resolved in all patients. CONCLUSIONS Switching raltegravir for enfuvirtide in patients whose VL is well suppressed seems to be safe. The results of our cohort showed only 1 patient with a VL above 50 by 24 weeks after switch and that VL was only 56 copies per milliliter. We found that raltegravir was well tolerated in this patient group. Our results are quite similar to another recent retrospective cohort study that found that only 1 of 35 patients had a VL slightly above 50 copies (60 copies) at week 20 after a switch to raltegravir.7 The strategy of switching one class for another new class agents in which resistance is not expected seems to be safe. Other new agents for the switch, like maraviroc, are not an ideal option because with undetectable VL, CCR5-tropism cannot be assessed adequately. Moreover, using a new generation agent-like etravirine-may also be more risky because many patients with prior NNRTI exposure may have resistance patterns that may make etravirine less than fully active and cannot be assessed in patients who are fully suppressed on their current regimen. A switch from enfuvirtide to raltegravir in virologically suppressed patients did not lead to early virologic breakthrough, except for 1 patient with low-level rebound count at week 24. And CD4 cell counts remained stable overall. Although some patients in our cohort had decreases in CD4 cell counts, all patients had stable CD4 percent. Other safety parameters were stable, and all patients benefited from resolving injection sites reaction. Longer follow-up is required to ensure the durability of the observed benefits and to more fully evaluate the impact on CD4 cell profile. ACKNOWLEDGMENTS We thank Phil Grant and Andrew Zolopa for reviewing the article. Annie Talbot, MD*†‡ Nima Machouf, PhD§ Réjean Thomas, MD§ Suzanne Marcotte, MSc* Rachel Therrien* Bernard Lessard, MD*† Serge Dufresne, MD*† Richard Lalonde, MD‖ *UHRESS du CHUM: unité hospitalière de recherche et d'enseignement sur le SIDA du centre hospitalier de l'université de Montréal (Hôtel-Dieu, Notre-Dame, St-Luc) †Clinique du Quartier Latin ‡Stanford University §Clinique l'actuel ‖UHRESS du CUSM: unité hospitalière de recherche et d'enseignement sur le SIDA du centre hospitalier de l'université McGill (Chest)