 |
|
|
| |
Results on First-Line ABC/3TC With RNA Above 100,000: ARIES Study of abacavir/3TC plus atazanavir/r
|
| |
| |
48th ICAAC
October 25-28, 2008
Washington, DC
Mark Mascolini
A trial of abacavir/lamivudine (ABC/3TC) plus atazanavir/ritonavir produced some results that contradicted an earlier finding of reduced efficacy with this nucleoside backbone in people starting therapy with more than 100,000 HIV RNA copies/mL [1]. But there are a variety of virologic measures to review. The full Aries poster results will also be reported by NATAP.
Controversy over first-line ABC/3TC arose when AIDS Clinical Trials Group (ACTG) investigators reported results of ACTG study A5202 at the International AIDS Conference in August 2008 [2]. This trial randomized previously untreated people to ABC/3TC or tenofovir/emtricitabine (TDF/FTC) plus either efavirenz or atazanavir/ritonavir. Among people starting therapy with a viral load above 100,000 copies, time to virologic failure--and to grade 3 or 4 side effects--proved significantly shorter in the ABC/3TC group than in the TDF/FTC group. At the same meeting session, however, GlaxoSmithKline investigators unfurled an analysis of ABC/3TC in six earlier trials showing no response difference in people beginning therapy with viral loads above or below 100,000 [3].
ARIES, the new study, involved 515 previously untreated people who started one of the ACTG A5202 regimens, ABC/3TC plus atazanavir/ritonavir, with a viral load above 1000 copies and any CD4 count [1]. At week 36 the protocol randomized people to continue taking that regimen or to drop the ritonavir boost, push the atazanavir dose to 400 mg once daily, and continue the other drugs. Median pretreatment viral load stood at 5.08 log (about 120,000 copies) and median CD4 count at 199. Slightly more than half of study participants began treatment with more than 100,000 copies. Most ARIES enrollees were men (83%) and white (63%). Everyone got screened for HLA-B*5701, the genetic wrinkle that predicts hypersensitivity reaction to abacavir.
The ARIES team defined virologic failure as failure to push the viral load below 400 copies by week 30 or a confirmed rebound above 400 copies. They also analyzed response according to the primary efficacy endpoints of ACTG A5202: a viral load at or above 1000 copies at or after week 16 and before week 24, or a confirmed rebound above 200 copies at or after week 24.
Given the gene screen for HLA-B*5701, a moderately high number of people--73, or 14%--dropped out of the study by week 36. Most dropouts resulted from adverse events (22%) or loss to follow-up (22%), though 11 of the 73 dropouts (15%) had an "insufficient viral load response" and 5 (7%) had protocol-defined virologic failure.
At week 36 of ARIES, 410 of 515 people (80%) had a viral load below 50 copies and 422 (82%) had a load under 200 copies in a time-to-loss-of-virologic response (TLOVR) analysis. The study group included 227 people with a pretreatment load under 100,000 copies and 288 with a starting load over 100,000. In the under-100,000 group, 190 (84%) had a viral load below 50 copies at week 36, compared with 220 (76%) in the over-100,000 group. Among people who began treatment with fewer than 100,000 copies, 193 (85%) had a 36-week load under 200 copies, compared with 229 (80%) who started with more than 100,000 copies.
At the 36-week point, ARIES investigators counted 15 virologic failures, 5 in the 227 people (2.2%) with fewer than 100,000 copies before treatment and 10 in the 288 people (3.5%) with more than 100,000 copies before treatment. (That breakdown did not appear in the poster presentation of these results, but it did in the pdf abstract file of study results distributed to the media before ICAAC.) (from Jules: its in the poster I have, and in my ARIES report)
These virologic failure results--and the week-36 sub-50 rate--seem to bolster the ACTG A5202 conclusion that ABC/3TC-containing regimens lack some punch in people starting therapy with a high viral load (from Jules: but in most studies including 96 Weeks ARTEMIS resented here at this ICAAC there is a falloff between viral response in the patient group with <100,000 copies/ml at baseline compared to the >100,000 c/ml patient group at baseline). But when ARIES researchers filtered their response results through an A5202 endpoint sieve, they saw little difference between people starting ABC/3TC plus abacavir/ritonavir with more versus fewer than 100,000 RNA copies. The overall virologic response rate according to A5202 endpoints was 96.5%, with rates of 98.1% in people with fewer than 100,000 copies before therapy and 95.1% in people with more.
The ARIES investigators proposed that the worse sub-50-copy rate with a pretreatment load above 100,000--but an equivalent sub-200-copy rate in the 100,000-plus and sub-100,000 groups at week 36--suggest that some patients who began treatment with a high load were responding well but had yet to reach the 50-copy threshold.
In a 12-study meta-analysis at this meeting, reported separately by NATAP, independent investigators found evidence that TDF/FTC outmuscles ABC/3TC when combined with a first-line ritonavir-boosted PI, regardless of pretreatment load [4].
References
1. Squires K, Young B, DeJesus E, et al. Atazanavir/ritonavir + abacavir/lamivudine in antiretroviral-naive HIV-1 infected HLA-B*5701 negative subjects demonstrates efficacy and safety: the ARIES trial. 48th Annual International Conference on Antimicrobial Agents and Chemotherapy (ICAAC). October 25-28, 2008. Washington, DC. Abstract H-1250a.
2. Sax P, Tierney C, Collier A, et al. ACTG 5202: shorter time to virologic failure (VF) with ABC/3TC than TDF/FTC in treatment-naïve subjects with HIV RNA >100,000. XVII International AIDS Conference. August 3-8, 2008. Mexico City. Abstract THAB0303.
3. Pappa K, Hernandez J, Ha B, et al. ABC/3TC shows robust virologic responses in ART-naïve patients for baseline (BL) viral loads of >100,000c/mL and <100,000c/mL by endpoint used in ACTG5202. XVII International AIDS Conference. August 3-8, 2008. Mexico City. Abstract THAB0304.
4. Hill AM, Sawyer WS. Effects of NRTI backbone on efficacy of first-line boosted PI-based HAART--meta-analysis of 12 clinical trials in 4896 patients. 48th Annual International Conference on Antimicrobial Agents and Chemotherapy (ICAAC). October 25-28, 2008. Washington, DC. Abstract H-1254.
|
| |
|
|
|
|
|
