icon star paper   Articles  
Back grey_arrow_rt.gif
Induction-Maintenance Study: efavirenz plus abacavir and Combivir switch to Trizivir
  This article reports the results from two studies. The first study (CNAF3008) examined induction quadruple therapy of abacavir, Combivir, plus efavirenz in treatment-naive patients. Following this induction therapy, the study investigators conducted a continuation study (ALF3002; Suburbs Study) to evaluate a simplified regimen with Trizivir (abacavir plus Combivir) for those study patients with an HIV viral load maintained <50 copies/ml at 48 weeks.
BRIEF SUMMARY: In the first phase using this 4 drug regimen viral response was good with 77% of the 31 patients enrolled achieving <50 copies/ml. The viral response in the maintenance phase was also good as 85% of the 20 patients in the maintenance phase had <50 copies/ml after 48 weeks. These are small studies. Here is a link to a report on this 4-drug approach including the pros and cons which was written after ICAAC (Sept 2002);
QUAD Therapy: pros and cons; written by Kimberly Smith, MD http://www.natap.org/2002/ICAAC/day27.htm
The induction phase study results were published in JAIDS (Oct 2002) and reported by Pierre de Truchis and the CNAF Group. Here are pertinent excerpts from the publication. The authors said in an effort to find PI-sparing, yet compact and potent regimens, we have designed a study combining three nucleosides (Combivir [3TC + ZDV + abacavir [ABC]) with an NNRTI (efavirenz [EFV]). The primary objective of this study, CNAF3008, was to evaluate long-term safety, tolerance, and efficacy. Because of a limited prior experience with abacavir combined to efavirenz, the study was designed as a pilot with small patient numbers.
This was aphase III, 48-week pilot, open-label, multicenter study (12 sites in France). Patients received Combivir (300 mg zidovudine and 150 mg lamivudine, both in one tablet) one tablet twice daily, plus abacavir 300 mg, one tablet twice daily, plus efavirenz 200 mg, three capsules once a day.
Thirty-one subjects treatment-naive patients enrolled in the study (5 women, 26 men, median age 35 years). Median VL at entry was 4.69 (50,000 copies/ml). Thirteen subjects (42%) had baseline VL >100,000 copies/mL. Median baseline CD4 cell count was 322 cells/mm3 (range, 98-858). Two subjects discontinued from the study before week 48 (pregnancy, personal choice), and 5 discontinued at least one study drug.
By week 48, the median CD4 count increased from 322 cells at baseline to 404 using the ITT analysis (median increase of +128 cells.
At week 8, the proportion of patients with VL <50 copies/ml was 52% (33%-70%) in the ITT (S = F). At week 48, the proportion of patients with VL<50 copies/ml 77% (59%-90%) in the ITT (S = F). The week 48 median decrease from baseline in VL was -2.99 log for the ITT population.
Similar results were observed for patients with VL >100,000 copies/ml compared with patients having VL <100,000 copies/mL at baseline, with the exception of a higher median reduction in VL in subjects with baseline VL >100,000 copies/ml.
Using the modified assay with an LLQ of 5 copies/ml the proportion of patients below the limit of assay detection was 42% (25%-61%) for the ITT (S = F) population. The median decrease from baseline at week 48 was -3.97 log (range: -5.48 ; -0.06) in ITT. This maximal median decrease was reached at week 24 and sustained to week 48. A shorter time to reach <5 copies/mL for subjects with baseline VL <100,000 copies/ml was observed. At week 16 and 48, the probability to reach <5 copies/ml was 60% and 100% for subjects with baseline VL <100,000; whereas the probability of reaching this goal was 8% and 83% in the >100,000 copies/ml group.
One subject experienced cutaneous Kaposi sarcoma (3 lesions) at week 12 with spontaneous remission at week 24.
Twenty-seven patients (87%) experienced at least one study drug-related adverse event (AE). The most common AEs were nausea and vomiting (13/31, 42%), dizziness/vertigo (12/31, 39%), malaise and fatigue (8/26%), skin rashes (7/31, 23%), and sleep disorders (6/31, 19%). Six patients (19%) permanently discontinued at least one study drug due to an AE: 1 patient discontinued all study drugs (vomiting, depression); 5 stopped efavirenz as a result of vertigo (2), insomnia (1), gastralgia (1), and isolated rash (1); 1 stopped abacavir for isolated rash and one Combivir for myositis. Six serious AE were reported: depression (5), abdominal pain (1). No definite hypersensitivity reaction to abacavir was observed.
Median cholesterol increased significantly from 177 mg/dl at day 14 to 203 at week 48, corresponding to a median increase of 15. Fasted glucose increased significantly about 10%. No significant differences were observed for fasting triglyceride measurements. No clinical signs or symptoms of lipodystrophy were reported in this study.
According to the patient self-report questionnaire, adherence was considered to be very good; 20/24 (83%) and 19/21 (90%) of subjects at week 24 and week 48 visits, respectively, stated that they did not missed a dose or missed doses less than once a week in the 4 weeks preceding the reported visit.
Author summary: This pilot study is one of the first studies to demonstrate that a PI-sparing quadruple regimen based on three nucleosides-abacavir and Combivir in combination with efavirenz-can lead to a profound and sustained decline in levels of plasma HIV-1 RNA. The treatment regimen was generally well tolerated without any new or unexpected side effects. Adverse event were generally transient and mild to moderate. Efavirenz was more commonly discontinued as a result of an adverse event. Although the incidence of rash alone was 23%, there were no cases of hypersensitivity to abacavir in this small study.
This pilot study demonstrates that this initial combination of Combivir, abacavir, and efavirenz led to a profound and rapid decline in VL, with a median reduction of 2.7 log as early as week 4. HIV RNA viremia quantification results with the modified assay at 5 copies/ml have been already published by Yerly et al. This test could be used to identify subjects reaching very low nadirs and could predict the intensity and durability of the regimen's antiviral effect. Our results further demonstrated the major reduction in VL (-4.0 log) obtained at treatment initiation showed a comparable viremia decay slope in subjects <100,000 or >100,000 copies/ml at baseline. Magnitude and durability of virologic response was higher than the one observed in the classic triple therapies analyzed by Bartlett et al. This profound suppression may maximize antiviral effect and durability of the response, helping to minimize emergence of resistant viral strains. All patients who remained on the quadruple regimen of ABC/COM/EFV reached VL <50 copies/ml throughout 48 weeks, irrespective of baseline VL.
The results of our pilot study allow the start of larger comparative studies of triple versus quadruple therapy. After study completion, it could be conceivable to propose a simplified regimen with Trizivir to subjects with a controlled plasma HIV-1 RNA. The follow-on study will evaluate long-term tolerance and efficacy of this compact maintenance regimen.
Here are the results reported by study investigators: De Truchis Centres hospitaliers Ile de France; GlaxoSmithKline, France. Subjects with VL < 50 copies/ml, CD4+ cell count > 100 cells and who had received abacavir+Combivir+ efavirenz throughout the initial 48 weeks of CNAF3008 study could be included in this trial and were switched to Trizivir one tablet BID. Genotype and phenotype were performed in case of virological failure.
The intent-to-treat analysis population (ITT) included all AZLF3002 study participants who received at least one dose of Trizivir following efavirenz discontinuation and with at least one efficacy criteria in AZLF3002 study. Percentage of subjects with plasma HIV-1 RNA < 50 copies/mL was calculated considering switch/missing data as failure and was associated with 95% exact confidence intervals.
Of 31 patients completing CNAF3008 study, 20 patients met the inclusion criteria and were enrolled in Suburbs between January 2000 and March 2000. 7/11 non-enrolled subjects stopped at least one study drug before day 0 Suburbs and 4 refused to participate.
The median patient age of the 20 participants was 37; 90% male; average VL <1.70 log (50 copies); 9 (45%) of patients had >100,000 copies/ml before starting the initial QUAD regimen; median CD4 was 349 before starting QUAD regimen and 443 when starting maintenance Trizivir regimen. 19/20 study participants completed week 48 on study medication. One subject discontinued TZV at week 40 due to drug related arteritis of lower limbs.
At week 48, the proportion of patients with plasma HIV-1 RNA < 50 copies/ml was 85% (17/20). Using the modified assay, the proportion of patients with plasma HIV-RNA < 5 copies/ml was 50% (10/20).
Of the 3 patients with plasma HIV-1 RNA > 50 copies/ml at week 48, one patient discontinued Trizivir at week 40; a second patient (844 copies/mL) had poor adherence, and the third patient had plasma HIV-1 RNA at 128 copies/ml. Median plasma HIV-1 RNA decreased from 4.83 log (day 0 CNAF3008) to < 50 copies/ml at week 24 and remained stable until the end of Suburbs. The corresponding median decrease of HIV-1 RNA between Day 0 (CNAF3008) to week 48 (Suburbs) was -3.71.
At week 48 similar efficacy results were observed for patients with baseline plasma HIV-1 RNA > 100,000 copies/mL and plasma HIV-1 RNA < 100,000 copies/ml in CNAF3008. 6/11 subjects with baseline HIV-1 RNA < 100,000 copies/mL and 4/9 with baseline >100,000 copies/mL had HIV-1 RNA < 5 copies/mL.
Five (25%) patients experienced a total of seven adverse events (AEs) considered to be drug related by the investigator. Most of the drug related AEs were mild in intensity (5/7). Two events (arteritis and diabetes) were grade 3. Only one stopped study regimen due to arteritis of lower limbs. Two patients experienced one serious adverse event: a worsening of crural hernia, and a drug related insulin-dependent diabetes.
No hypersensitivity reactions to abacavir were reported. Baseline median cholesterol increased significantly during CNAF3008 with a median increase of +0.67 mmol/L, (p = 0.004) to reach a median value of 5.46 mmol/L (3.44; 8.60). After EFV discontinuation, during Suburbs median cholesterol dropped significantly to the initial baseline value of CNAF3008 As reported above from about 200 to 177). There was no change in fasting triglycerides between CNAF3008 baseline and week 48 suburbs.
  icon paper stack View Older Articles   Back to Top   www.natap.org