| |
Durability of efavirenz compared to nevirapine with long-term follow-up of an antiretroviral-naive patient cohort; efavirenz CNS effects
|
| |
| |
9th Annual Conference of the British HIV Association. April 24-26; Manchester, UK. G Matthews, Y Gilleece, C Orkin, S Mandalia, MR Nelson, B Fisher,
M Bower and BG Gazzard. Chelsea and Westminster Hospital, London, UK
This report contains two studies reported at the Conference. The first reports a more durable response to efavirenz than nevirapine as measured by time to treatment failure and time to viral failure. At the Retrovirus Conference in Feb 2003 researchers reported the one year follow-up results from the 2NN Study, a comparison of efavirenz and nevirapine. The 2NN Study researchers reported that over the course of this one year study there was no statistically significant difference between nevirapine and efavirenz in terms of the percent achieving undetectable viral load after one year on therapy. The 2NN Study also found that patients receiving efavirenz were more likely to experience CNS (neuropsychiatric) side effects than patients receiving nevirapine, but patients receiving nevirapine were more likely to experience liver enzyme elevations and potential hepatotoxicity. The second study in this report finds that of 483 patients starting efavrienz 80 patients (16%) discontinued in I think the first two years due to adverse events, and 70% of the adverse events (11%) were due to neuropsychiatric reason (CNS). The second study was in treatment-naive patients but who had more advanced HIV with on average 176 CD4 cell count and 162,000 copies/mL viral load.
Durability
Background: Few long-term data directly compare outcomes for efavirenz (EFV) versus nevirapine (NVP) regimens in antiretroviral-naive patients. We provide durability data on non-nucleoside reverse transcriptase inhibitors (NNRTIs) with up to 260 weeks' follow-up.
Methods: Antiretroviral-naive patients initiating EFV or NVP since 01/96 were identified from a prospective observational database. Virological failure [two viral load (VL) measurements >500 copies/ml] or switch failure (discontinuation/switch) was identified. Multivariate analysis determining significant factors associated with failure and time to failure (TTVF) Kaplan-Meier (KM) curves was performed.
Results: 694 patients initiated NNRTI-based highly active antiretroviral therapy (HAART) (357 EFV, 337 NVP) since 01/96 with a total follow-up of 292 patient-years. No significant differences between EFV and NVP were found for sex, baseline VL or CD4, prior AIDS. EFV tended to be commenced in later years and with zidovudine/lamivudine (ZDV/3TC), NVP with stavudine/didanosine (d4T/ddI). In multivariate analysis, significant independent predictors of failure (virological and/or switch) were: prior AIDS illness [RH 1.48, 95% confidence interval (CI) 0.99- 2.22], backbone d4T/ddI (RH 1.96, 95% CI 1.21-3.17) and NVP HAART (RH 1.60, 95% CI 1.07-2.40). Patients receiving d4T/ddI as their nuke backbone were more likely to experience treatment failure than patients receiving AZT/3TC.
Editorial note: it is my understanding from a report by www.aidsmap.org that the study showed that patients with baseline CD4 count above 100 were 53% less likely to experience treatment failure than patients with below 100 CD4 counts at baseline.
Stratifying by year of therapy had no effect on outcome. 73 (10.5%) patients failed to achieve HIV VL <500 copies/ ml within 6 months of therapy.
Of 621 patients achieving <500, 31 (5.0%) later developed virological failure and 45 (7.2%) had failure as defined by switch of therapy. Kaplan-Meier curves showed an EFV benefit both in TTVF, time to viral failure, (P=0.0476) and time to treatment failure (VF±switch) (P=0.0324).
Conclusions: This cohort provides the strongest evidence yet that durability of EFV over NVP continues with long-term follow-up.
SIDE EFFECTS
Efavirenz: what happens in the long-term?
Oral abstract 9. L Swaden, CA Sabin, FC Lampe, MS Youle, MA Johnson and M Lipman. Royal Free and University College Medical School, London, UK
Background: Non-nucleoside reverse transcriptase inhibitors (NNRTIs)
have a known efficacy and side-effect profile. However, much of this is
derived from either drug trial data or a relatively short-term follow-up.
Methods: We analysed an observational cohort of 483 subjects (78%
male, 56% gay risk for HIV, 66% Caucasian origin) starting efavirenz
(EFV) at our HIV centre from 1996 onwards and followed up for a
median of 23.8 months.
Results: At the start of treatment, median CD4 count and HIV RNA viral
load (VL) were 176 cells/µl and 5.21 log 10 copies/ml, respectively, with
38% being antiretroviral-naive. Of the 427 subjects with a baseline VL
>500 copies/ml, 75% (Kaplan-Meier) achieved a VL <500 within the
first 6 months; this occurred after a median of 92 days. Virological
rebound (two consecutive VL >500 copies/ml) occurred in 16.7% by 1
year and was associated in multivariable analysis with younger age,
starting EFV before 2000 and heavy prior use of NNRTIs or protease
inhibitors (PIs).
201 (42%) of subjects stopped EFV; the proportions stopping by 3, 6, 12 and 24 months were 13%, 20%, 31% and 43% respectively (13%, 19%, 27% and 37% in those who were antiretroviral-naive).
The most common causes of this were adverse events (AEs) (n=80; 16% of the 483 patients stopped EFV due to adverse events), patient choice (n=47) and viral rebound (n=23); 70% of AEs were neuropsychiatric in origin. The incidence of severe AEs did not decrease with prolonged follow-up. Risk factors for stopping due to AEs included concurrent use of PIs and Caucasian origin.
Discussion: EFV is a potent drug, though there is a high incidence of discontinuation due to AEs. These occur at a much later time (years
rather than weeks) than reported previously.
|
|
| |
| |
|
|
|
