Objectives: To assess the virologic noninferiority of an antiretroviral treatment simplification with coformulated zidovudine/lamivudine/abacavir (group 1) vs. coformulated zidovudine/lamivudine plus nevirapine (group 2) in HIV-1-infected patients receiving successful first-line highly active antiretroviral therapy.
Methods: This is a prospective, multicenter, open-label, comparative, randomized, noninferiority study. A delta of 15% for differences in virologic suppression <200 copies/mL between groups was prespecified with a 1-sided 0.025 significance level.
Results: A total of 134 patients were included into this study: 68 were allocated to group 1 and 66 to group 2. By intention-to-treat analysis (switch equals failure), the percentage of virologic suppression <200 copies/mL (<50 copies/mL) at week 48 was 71.0% (65.1%) and 73.0% (63.3%) in groups 1 and 2, respectively (estimate for differences [<200 copies/mL]: −2.1, 95% CI: −17.4-13.1, P = 0.783). Thirteen and 14 patients in groups 1 and 2, respectively, discontinued therapy due to adverse events. Dyslipidemia improved in both groups, with a higher improvement in low-density lipoprotein cholesterol (P = 0.049) in group 1.
Conclusions: Group 1 is not inferior to group 2 regarding virologic suppression <200 copies/mL. Both strategies improve lipid profile.
From the *AIDS Care Unit, Lluita contra le SIDA Foundation, Germans Trias i Pujol Hospital, Universitat Autònoma de Barcelona; †Hospital Virgen de la Macarena, Sevilla; ‡Hospital Xeral de Vigo; §Hospital Costa del Sol, Marbella; ∥Hospital Clínico Universitario, Santiago; and ¶Hospital Juan XXIII, Tarragona, Spain.
Received for publication November 9, 2004; accepted March 17, 2005.
The study received an unrestricted grant from GlaxoSmithKline and Boehringer Ingelheim. The authors had complete control over the data and the final manuscript.
Reprints: Anna Bonjoch, Fundació Lluita contra la SIDA, Hospital Universitari Germans Trias i Pujol, Universitat Autònoma de Barcelona, Ctra. de Canyet, s/n, 08916 Badalona, Spain (e-mail: email@example.com).
Antiretroviral treatment simplification from protease inhibitor (PI)-based approaches to nonnucleoside analogue (NNRTI)-including combinations preserves the antiviral efficacy of the former while exerting noteworthy benefits on lipid metabolism, treatment adherence, and quality of life.1,2 However, attempts at simplification with coformulated triple-nucleoside analogues (NRTIs) have provided divergent results.3-7 These strategies have a simpler and more comfortable administration, avoiding food constraints; have a beneficial impact on lipid metabolism, leading to fewer requirements of lipid-lowering drugs8; and are not subject to significant drug-drug interactions. However, when triple-NRTI regimens are given to patients with prior suboptimal NRTI drug exposure as a simplification strategy, these regimens clearly perform worse than nevirapine- or efavirenz-based combinations.7 Nevertheless, in patients with neither NRTI-related mutations nor prior suboptimal NRTI exposure, triple-NRTI regimens appear to perform as well as NNRTI-based combinations.
This study aims to prospectively evaluate the efficacy of treatment simplification with coformulated abacavir/lamivudine/zidovudine (fixed-dose combination) vs. coformulated lamivudine/zidovudine plus nevirapine in HIV-1-infected patients whose only antiretroviral experience is an ongoing first-line highly active antiretroviral therapy (HAART) approach that achieved and maintains profound virologic suppression.
MATERIAL AND METHODS
This is a multicenter, open-label, blinded, centrally randomized, parallel, comparative, controlled, noninferiority clinical trial. Patients with suppressed viral load at least in the 6 months prior to inclusion in the study and receiving a first-line treatment approach containing PIs or NNRTIs plus 2 NRTIs were randomly assigned to 1 of the 2 following arms: abacavir/lamivudine/zidovudine fixed-dose combination (Trizivir; GlaxoSmithKline, Tres Cantos, Madrid, Spain, group 1) or lamivudine/zidovudine fixed-dose combination (Combivir; GlaxoSmithKline, Tres Cantos, Madrid, Spain) plus nevirapine (Boehringer Ingelheim, Sant Cugat, Barcelona, Spain, group 2). This trial was approved by the Ethics Committee of each center and was authorized by the Spanish Health Authorities.
The primary endpoint was the proportion of patients with plasma HIV-1 RNA <200 copies/mL at week 48 of follow-up. Secondary endpoints included the proportion of patients with plasma HIV-1 RNA <50 copies/mL at week 48, the evolution of CD4 counts, change of lipid metabolism and body shape, the safety profile of regimens, as well as patient quality of life, adherence, and satisfaction with the ongoing antiretroviral treatment.
We designed a noninferiority study to compare the triple-nucleoside arm with respect to the nevirapine-containing arm. The sample size was calculated assuming an equivalence range of 15% between groups in the percentage of patients with HIV1- RNA levels <200 copies/mL at week 48, with a statistical power of 80% and a 1-sided 0.025 significance level.
Protocol Procedures and Study Definitions
Baseline evaluation was used to collect sociodemographic characteristics and antiretroviral treatment history. At baseline and at subsequent visits (12, 24, 36, and 48 weeks of follow-up), plasma HIV-1 RNA, CD4 cell counts, lipid profile (under fasting conditions), anthropometrical measures, adherence to antiretroviral treatment, quality of life, and patient satisfaction with therapy were determined. Lipid abnormalities were defined according to each laboratory's upper limit of normality. The severity of adverse clinical events and laboratory abnormalities were evaluated according to the AIDS Clinical Trial Group toxicity grading scale.9 HIV-1 RNA was quantified using the NASBA assay (Organon Teknika, Barcelona, Spain) or Amplicor Roche Molecular Systems. Ultrasensitive HIV-1 RNA analysis (with a limit of detection of 50 copies/mL) was performed in all but 2 participating hospitals according to center availability.
Adherence to treatment was self-reported by patients and was considered appropriate if subjects assured that they had taken at least 95% of the prescribed doses during the previous 2 weeks. Quality of life was evaluated by the MOS-HIV questionnaire (Medical Outcome Study for HIV-positive patients). Effort to take medication was assessed with 10-point Likert scales.
The statistical analysis was performed according to a per-protocol (PP) and an intention-to-treat (ITT) analysis. Patients who were lost to follow-up or who withdrew their consent before the second visit (and consequently only performed the screening or the baseline visit) were excluded from the final analysis. Subjects who were lost to follow-up after the second visit or who discontinued the study regimen due to other reasons apart from virologic failure were censored from the PP analysis but were considered as treatment failures for the ITT analysis. The corresponding 95% CIs were calculated. Results were considered significant at P < 0.05. Multiple comparisons were adjusted by a least significant difference test.
Overall, 134 patients were included in the study, 68 being allocated to group 1 and 66 to group 2. Most of the patients were male (82%). Median age at baseline was 39 years in both groups, and baseline CD4 cell counts were (median, interquartile range [IQR]) as follows: in group 1, 530 cells/mL (396-858), and in group 2, 554 cells/mL (385-780). The two study groups were comparable at baseline regarding demographic characteristics and immunologic parameters. Up to 89% of the patients in group 1 and 94% of the subjects in group 2 were receiving PIs at inclusion in the study. Overall, the most frequent PIs were nelfinavir (35%), ritonavir-boosted indinavir (21.6%), and unboosted indinavir (17.2%). Most frequent NRTI strategies were: lamivudine + stavudine (45%) and zidovudine + lamivudine (20%). Five patients from group 1 and 3 patients from group 2 were lost to follow-up or withdrew their consent before the second control visit and were excluded from the final analysis (P = 0.74).
Efficacy Measures and Immunologic Parameters
According to the ITT analysis, at the 48th week of follow-up, 71% of patients in group 1 and 73% of patients in group 2 had plasma HIV-1 RNA levels <200 copies/mL. Testing for differences between P1 − P2 = 0 yielded an estimate for the difference of −2.1% (95% CI: −17.4%-13.1%, P = 0.783) (PP, 98.6% in group 1 and 98.5% in group 2, estimate for difference: 0.04% [95% CI: −5.5%-5.6%, P = 0.988)]). In the ultrasensitive HIV-1 RNA ITT subanalysis (corresponding to 91.5% of the overall study sample), percentages of viral suppression <50 copies/mL at week 48 were 65.1% for group 1 and 63.2% for group 2 (estimate for difference: 0.010 [95% CI: −0.151-0.173, P = 0.897] [PP, 78.0% in group 1 and 73.4% in group 2, estimate for difference: 0.045, 95% CI: −0.123-0.214, P = 0.599]). Upon treatment analyses, median baseline values of CD4 cell counts were 492 cells/μL (IQR 396-792) in group 1 and 579 cells/μL in group 2 (IQR 426-846) (P > 0.05). A nonstatistically significant median CD4+ count increase from baseline of 96 cells/μL and 43 cells/μL was observed in groups 1 and 2, respectively.
Analysis of Virologic Failures According to Genotypic Resistance Mutations
One patient belonging to group 1 had early virologic failure at week 12 of follow-up. At his inclusion in the study, he was receiving didanosine/stavudine/ritonavir. Genotypic analysis at virologic failure showed mutations M184V and T215Y and the T69STG insertion. One patient from group 2 experienced confirmed virologic failure at week 48 of follow-up. His prior treatment included zidovudine/lamivudine/nelfinavir. Although his initial adherence was correct, it worsened during the study. Genotypic analysis at failure yielded mutations M41L, K101E, M184V, and G190A.
Safety, Metabolic Profile, and Body Fat Distribution
Thirteen and 14 patients, respectively, discontinued therapy due to toxicity. Eighty-seven percent of treatment discontinuations occurred within the first 12 weeks of therapy. The most frequent adverse event was rash in both groups (5 patients per group, respectively). Grade 3 or 4 toxicity was detected in only 1 patient from group 1 (anemia) and in 4 patients from group 2 (3 clinical hepatitis and 1 rash).
Median total cholesterol decreased from 5.5 mM (IQR: 4.1-6.2) at baseline to 4.75 mM (IQR: 4.0-5.3) at week 48 in group 1 (P = 0.018) and from 5.6 mM (IQR: 5.1-6.4) to 5.06 mM (IQR: 4.5-5.7) in group 2 (P = 0.03), respectively. Also, low-density (LDL) lipoprotein cholesterol ameliorated from 3.59 mM (IQR: 2.6-4.4) to 2.68 mM (IQR: 2.1-3.2) at week 48 (P < 0.001) in group 1; and from 3.6 mM (IQR: 2.9-4.2) to 3.11 mM (IQR: 2.5-3.6) at week 48 (P = 0.046) in group 2, respectively. The proportion of patients with hypercholesterolemia declined from 55.6% at baseline to 30.5% at week 48 in group 1 (P = 0.019) and from 68.9% to 42.2% in group 2 (P = 0.019). The proportion of patients with elevated LDL cholesterol decreased from 60% to 20.6% (P = 0.003) in group 1 and from 59.3% to 41.7% (P > 0.05) in group 2. All the variations in lipid profile were comparable between groups except for LDL cholesterol at week 48 (P = 0.049), with a more pronounced decrease in group 1. The subanalysis of patients who had prior exposure to PIs (92% of the patients included) showed similar results to the overall group. Overall, 47% of patients (31 patients from group 1 and 32 from group 2) presented lipodystrophy at baseline. No significant changes were demonstrated in body fat distribution throughout the study, and no new cases of lipodystrophy were detected.
Adherence, Quality of Life, and Patient Satisfaction With Treatment
Adherence remained high in both groups during the study follow-up (group 1-baseline: 94.3 ± 12.5, week 48: 96.7 ± 10.9, P = 0.87; and group 2-baseline: 97.3 ± 5.5, week 48: 97.6 ± 6.1, P = 0.79). Differences between baseline and week 48 were not significant in either group. Quality of life remained at adequate levels from baseline to week 48 according to the General Health Scale (group 1-baseline: 68.4, week 48: 67.5; and group 2-baseline: 67.1, week 48: 67.4). The effort to take medication decreased significantly regarding all the scales evaluated (P < 0.001). A more pronounced decrease in effort to take medication under was observed in group 1 at week 48 with regard to group 2 (P = 0.023).
This prospective, open-label, randomized, comparative, noninferiority trial shows that, in patients receiving successful PI- or NNRTI-containing first-line HAART combinations, treatment simplification with coformulated abacavir/lamivudine/zidovudine fixed-dose combination is not inferior to coformulated lamivudine/zidovudine plus nevirapine in terms of the maintenance of viral suppression <200 copies/mL for up to 1 year. Our study lacks the statistical potency necessary to yield definitive conclusions regarding the noninferiority of regimens according to ultrasensitive HIV-1 RNA measurements. However, results obtained by both conventional and ultrasensitive methods are consistent. In fact, only 1 patient per group experienced virologic rebound during the study. This study clarifies that an effective and safe simplification of standard HAART regimens can be achieved with coformulated triple-NRTIs if this approach specifically targets patients with a low likelihood of harboring nucleoside analogue-associated resistance mutations (NAMs). This includes previously antiretroviral-naive patients reaching a rapid, permanent, and durable suppression of plasma viral replication below detectable limits at the initiation of their first PI- or NNRTI-based HAART. Conversely, and according to prior studies,4,5 simplification with triple-NRTI combinations should not be attempted in patients with a history of suboptimal drug pressure or in those suspected of harboring NAMs for any reason. When most of our patients initiated antiretroviral therapy, pretreatment genotyping was not a standardized procedure in the clinical management of HIV-1-positive persons; therefore, these data are unavailable for this study. In patients with persistent viral suppression, assumptions regarding NRTI resistance can only be made based on clinical history because suppressed viral load precludes conventional resistance analyses. Our findings are concordant with previous studies suggesting that patients with a low probability of harboring NAMs or those without detectable NAMs are able to maintain long-term viral suppression with a triple-NRTI simplification regimen.3,6,7 Regarding the patient from group 2 who experienced treatment failure, it is likely that resistance mutations accumulated in stepwise fashion during the study due to suboptimal treatment compliance.
There were no differences in the proportion of adverse events between the study arms. The high frequency of treatment discontinuation due to adverse events in our study may be related to previous optimal tolerance to treatment of our patients. Notably, although the overall rate of treatment withdrawals for adverse events was elevated and presented a similar magnitude in both study arms, only 16% of adverse events seen in our study were in fact grade 3 or 4. In this context, any disturbance, even mild effects, as well the awareness of potentially dangerous effects, such as hypersensitivity reactions or hepatitis, probably induced physicians and patients to return to the previous well-tolerated strategy.
Lipid abnormalities improved early and in sustained fashion in both groups, in agreement with previous reports showing benefits of PI-sparing treatments on lipid metabolism.8,10 Improvements were comparable in the two arms with the exception of a more pronounced decline in LDL cholesterol in patients from group 1. This improvement is clinically important and implies a decrease in dietary restrictions, lipid-lowering agents, and probably in the overall cardiovascular risk of the patient. No remarkable changes in body fat maldistribution were found in this study. One partial explanation for the lack of improvement in lipodystrophy observed in our study is the perpetuation of mitochondrial toxicity through the continuation of NRTI exposure,10 although other factors, such as the lack of more sensitive measurements of lipodystrophy in our study (dual-energy x-ray absorptiometry or CT scan) cannot be discarded. Moreover, 48 weeks of follow-up may be insufficient to detect significant variation in body fat distribution.
In conclusion, patients receiving effective first-line PI- or NNRTI-based HAART can safely simplify therapy with coformulated zidovudine/lamivudine/abacavir or coformulated zidovudine/lamivudine plus nevirapine, as long as their likelihood of harboring NAMs is extremely low. Careful individual study of tolerance, treatment history, and baseline genotypic testing, when available, is crucial in determining the optimal simplification approach strategy. In a context of low likelihood of NAMs, the two strategies are similar in terms of virologic suppression, immunologic recovery, improvement of metabolic abnormalities, and the effort to take medication.
1. Negredo E, Cruz L, Paredes R, et al. Virological, immunological, and clinical impact of switching from protease inhibitors to nevirapine or to efavirenz in patients with human immunodeficiency virus infection and long-lasting viral suppression. Clin Infect Dis
2. van Leth F, Phanuphak P, Ruxrungtham K, et al. Comparison of first-line antiretroviral therapy with regimens including nevirapine, efavirenz, or both drugs, plus stavudine and lamivudine: a randomized open-label trial, the 2NN Study. Lancet
3. Katlama C, Fenske S, Gazzard B, et al. TRIZAL study: switching from successful HAART to Trizivir™ (abacavir-lamivudine-zidovudine combination tablet): 48 weeks efficacy, safety and adherence results. HIV Med
4. van Leeuwen R, Katlama C, Murphy RL, et al. A randomized trial to study first-line combination therapy with or without a protease inhibitor in HIV-1 infected patients. AIDS
5. Hoogewert M, Regez RM, Schouten WE, et al. Change to abacavir-lamivudine-tenofovir combination treatment in patients with HIV-1 who had complete virological suppression. Lancet
6. Opravil M, Hirschel B, Lazzarin A, et al. A randomized trial of simplified maintenance therapy with abacavir, lamivudine, and zidovudine in human immunodeficiency virus infection. J Infect Dis
7. Martinez E, Arnaiz J, Podzamczer D, et al. Substitution of nevirapine, efavirenz or abacavir for protease inhibitors in patients with human immunodeficiency virus infection. N Engl J Med
8. Martínez E, Conget I, Lozano L, et al. Reversion of metabolic abnormalities after switching from HIV-1 protease inhibitors to nevirapine. AIDS
9. AIDS Clinical Trial Group. Table of Grading Severity of Adult Adverse Experiences
. Rockville, MD: Division of AIDS, National Institute of Allergy and Infectious Disease; 1996.
10. Moyle GJ, Baldwin C, Langroudi B, et al. A 48-weeks randomized, open-label, comparison of three abacavir-based substitution approaches in the management of dyslipidemia and peripheral lipoatrophy. J Acquir Immune Defic Syndr
The members of the SimplifiHAART study group: A. Bonjoch, R. Paredes, J. Miranda, E. Martínez-López, J. A. Muñoz-Moreno, S. Gel, S. Videla, E. Masmitjà, N. Pérez-Alverez, B. Clotet, AIDS Care Unit, Lluite contra la SIDA Foundation, Germans Trias i Pujol Hospital, Universitat Autònoma de Barcelona, Barcelona; J. Gálvez, D. Morales, L. Orbea, J. Rodríguez, M. A. Muniain, A. Domínguez, M.J. Ríos, Servicio de Medicina Interna, Hospital Virgen de la Macarena, Sevilla; C. Miralles, J. De la Fuente, A. López, A. Ocampo, Y. Barbazán, C. González, Hospital Xeral de Vigo; J. de la Torre, J. L. Prada, A. Del Arco, Hospital Costa del Sol, Marbella; A. Prieto, E. Losada, Hospital Clínico Universitario, Santiago; C. Viladés, F. Vidal, J. Peraire, Hospital Juan XXIII, Universitat Rouira i Virgili, Tarragona; M. Márquez, J. Santos, J. Ruiz, M. González, Hospital Universitario, Málaga; A. Mariño, A. Fernández, Hospital Arquitecto Marcide, El Ferrol; D. Peña, M. Pérez, Hospital de la Línea de la Concepción; A. Masabeu, Hospital de Palamós; J. Pasquau, Hospital Universitario Virgen de la Nieves, Granada; M. Rubio, Hospital Universitari Arnau de Vilanova, Lleida; J. A. Terrón, Hospital General de Jerez; M. J. López, J. Corredoira, Hospital Xeral Calde, Lugo; and H. Knobel, A. González, A. Guelar, Hospital del Mar, Barcelona.