Efavirenz versus nevirapine as a non‐nucleoside reverse transcriptase inhibitor in initial combination antiretroviral therapy for HIV infection
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To determine which NNRTI—EFV or NVP—is more effective in terms of viral load suppression when given in combination with two NRTIs as part of initial ARV combination therapy for HIV infection in adults and children.
Background
A total of 33.2 million people were living with HIV in 2007, and 22.5 million of this number were living in sub‐Saharan Africa (UNAIDS 2007). In countries with the financial resources and healthcare infrastructure to support HIV treatment, highly active antiretroviral therapy (HAART) has reduced hospitalisation, morbidity, and mortality among the HIV‐infected population (Mocroft 1998; Palella 1998; Gilks 2006;Hogg 1997).
For countries with limited resources, the World Health Organisation (WHO) recommends a public health approach to antiretroviral treatment to improve access, simplify clinical decision making, standardize regimens, and standardize monitoring, management of toxicity, and drug interactions (Gilks 2006).
The WHO model list of essential medicines describes the following three classes of antiretroviral (ARV) drugs used for treatment and prevention of HIV infection: nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs), non‐nucleoside reverse transcriptase inhibitors (NNRTIs), and protease inhibitors (PIs). Ritonavir, a synergistic enhancer, is recommended for use as a pharmacological booster to enhance the effects of PIs, and not as an antiretroviral drug in its own right (WHO 2007).
The International AIDS Society (IAS) ‐ USA recommends that the initial treatment regimen should include two NRTIs and efavirenz (EFV) or a boosted PI (Hammer 2008). The WHO recommends that for low‐resource settings, initial ARV therapy should include two NRTIs and one NNRTI, which can be either nevirapine (NVP) or EFV (Gilks 2006). Delavirdine (DLV), another NNRTI, is not included in the core list of essential medicines published in March 2007 by the WHO.
For HIV‐infected infants with no exposure to maternal or infant NNRTIs, or whose exposure to maternal or infant ARV treatment is unknown, standard NVP‐containing triple‐drug therapy should be started. For HIV‐infected infants with a history of exposure to single‐dose NVP or to NNRTIs containing maternal ARV therapy or preventive ARV regimens, a PI‐based triple ARV therapy regimen should be started. Where PIs are not available, affordable, or feasible, NVP‐based therapy should be used (WHO 2008). Efavirenz has not been studied adequately in children under the age of 3 years or among children weighing less than 13 kg.
For any initial regimen, the potency, durability of efficacy, ease of administration and storage, tolerability, and toxicity must be balanced with cost and availability (Gilks 2006).
Apart from the high cost of PIs, the pill burden, and the dietary constraints, they are linked to serious, long‐term metabolic disorders, most notably an increased risk of lipodystrophy and hyperlipidaemia (Moyle 2000; BHIVA 2001). A meta‐analysis of 12 trials revealed that NNRTI‐based regimens were better at virologic suppression than PI‐based regimens (Ioannidis 2006).
The NNRTIs have a more favourable adverse effect profile than PI’s, are less expensive, and are easier to administer. They also are more cost effective (Beck 2008).The main disadvantage of this group is that a single mutation may confer resistance to the entire class of NNRTIs because cross‐resistance among agents of this class is nearly universal (Deeks 2001; CDC 2002).
Nevirapine may be responsible for severe or fatal hepatotoxicity and a rash which may present in severe form as Stevens Johnson syndrome, but it is the NNRTI of choice for pregnant women because EFV may be teratogenic. It also may cause a rash or symptoms within the central nervous system (CNS), such as dizziness, somnolence, insomnia, drowsiness, nightmares, hallucinations, and poor concentration (DHHS 2001).
The authors of a previous systematic review which compared EFV to NVP in three ‐drug regimens concluded that their short‐term efficacy (42 to 52 weeks) was identical (Neuwelt 2003). The authors did not address other issues, such as adherence and clinical evolution.
Providing evidence on the more appropriate choice of NNRTI in terms of efficacy, durability, and tolerability is important to patients, caregivers, and policy makers all over the world. With such evidence, the cost and availability of each drug then could be evaluated on a secondary basis.
Objectives
To determine which NNRTI—EFV or NVP—is more effective in terms of viral load suppression when given in combination with two NRTIs as part of initial ARV combination therapy for HIV infection in adults and children.
Methods
Criteria for considering studies for this review
Types of studies
-
Randomised controlled trials
-
Observational cohort studies
Types of participants
Adults and children infected with HIV and without prior exposure to ARV therapy. Women who have received short courses of NNRTIs for the prevention of mother‐to‐child transmission (PMTCT) of HIV will be included. Owing to the paucity of knowledge on the effects of EFV in children under 3 years, this age group will not be included.
Types of interventions
We will consider triple‐drug ARV combination regimens for initial therapy that contain two NRTIs plus either EFV or NVP at any dose (EFV + 2NRTIs versus NVP + 2NRTIs). The NRTIs in each combination need not be specified, but if specified, they must be the same in both arms. Trials with additional arms will be included, but only the EFV‐ and NVP‐containing arms will be evaluated in the review.
We will compare the therapeutic efficacy of triple‐drug regimens containing EFV and NVP using plasma HIV RNA concentration as a surrogate marker for clinical progression. Plasma HIV RNA has been demonstrated to be a reliable predictor of HIV disease progression (Lau 2007, Mellors 2007).
The stavudine (D4T) lamivudine (3TC) and NVP Cochrane review (Siegfried 2006) analyzed studies comparing this regimen to any other available regimen used in the treatment of HIV/AIDS in newly or previously treated adults and adolescents. Some of the studies included in that review may be analyzed if they compare this regimen with another containing D4T, 3TC, and EFV in patients who have never received ARV therapy.
Types of outcome measures
Primary outcomes
-
Percentage of participants achieving undetectable plasma HIV RNA concentration (viral load) over time.
Secondary outcomes
-
Time to undetectable plasma HIV RNA
-
Change in mean CD4 cell count
-
Clinical progression to AIDS
-
Rate of discontinuation of treatment
-
Rate of treatment failure (virological rebound)
-
Time to virological failure.
-
Any adverse events (clinical and biological). These events will be reported as mild, moderate, severe, or severe‐life threatening and graded from one to four as per WHO guidelines (WHO 2006).
Search methods for identification of studies
See: HIV/AIDS Collaborative Review Group search strategy.
The identification of studies will be done with the assistance of the HIV/AIDS Review Group Trials Search Coordinator. We will formulate a comprehensive and exhaustive search strategy in an attempt to identify all relevant studies, regardless of language or publication status (published, unpublished, in press, or in progress).
We will search the following electronic databases.
1. MEDLINE (1996 ‐ 2008).
2. EMBASE (1996 ‐ 2008).
3. AIDSearch, which includes coverage of the following conferences.
a. International AIDS Conference (1996 ‐ 2008).
b. Conference on Retroviruses and Opportunistic Infections (1996 ‐ 2008).
c. The British HIV Association conference (1996 ‐ 2008).
d. International Congress on Drug Therapy in HIV Infection (1996 ‐ 2008).
4. The Cochrane Central Register of Controlled Trials (1996 ‐ 2008).
5. The WHO International Clinical Trials Registry Platform and Clinicaltrials.gov for ongoing trials.
The search strategy will include text terms such as efavirenz, EFV, EFZ, Sustiva, Stocrin, nevirapine, NVP, Viramune, Nevimune, non‐nucleoside reverse transcriptase inhibitor, NNRTI, PI sparing, non‐PI containing, etc. We will perform handsearches of the reference lists of all pertinent reviews and studies found. Finally, we will contact research organizations and experts in the field for unpublished and ongoing studies. Searches will be conducted for 1996 to 2008, the years during which NNRTIs have been approved for use and allowed on the market.
Data collection and analysis
Selection of studies
Critical appraisal of all identified citations will be done independently by the two authors to establish the possible relevance of the articles for inclusion in the review. Studies will be reviewed for relevance based on study design, types of participants, interventions, and outcome measures. Any disagreement will be resolved by discussion or by contacting an independent author. We will give reasons for excluding potentially relevant trials in an excluded studies table.
Data extraction and management
We will design and test a data extraction form. Data will be extracted independently by the two authors using the agreed upon form. Both authors will verify the extracted data. Extracted information will include methods, participant characteristics, interventions, and outcomes.
In the event that the authors disagree on the abstraction of study details, an independent author will be contacted to resolve the conflict. We will attempt to reach the principal investigators of the studies in case of any missing data or need for clarification about the studies.
Assessment of risk of bias in included studies
The risk of bias will be assessed using the following.
-
Sequence generation: how the allocation sequence was generated and whether it was adequate.
-
Allocation concealment: how the allocation sequence was concealed and whether it was adequate.
-
Blinding of participants, personnel, and outcome assessors.
-
The description of the completeness of outcome data for each main outcome.
-
Selective outcome reporting.
-
Baseline characteristics for observational studies.
-
Other potential sources of bias (e.g. funding).
Studies will be graded as being at high, low, or moderate risk of bias. "Risk of bias" tables will be filled independently by the two authors.
Measures of treatment effect
We will analyse the data using Review Manager 5.0.15 and present the results with 95% confidence intervals. We will calculate the relative risk (RR) and the odds ratio (OR) for binary data, the weighted mean difference (WMD) for continuous data measured on the same scale, and the standard mean difference (SMD) for data measured on different scales.
Subgroup analysis and investigation of heterogeneity
Studies first will be assessed for clinical heterogeneity. If studies are similar enough to combine, a meta‐analysis will be performed and statistical heterogeneity assessed. If there is significant unexplained statistical heterogeneity a random effects meta‐analysis will be performed.
Possible sources of heterogeneity in this review might include NRTI used, threshold for detection of viral RNA, drug dosage, and study design. If there is heterogeneity and the data are available we will explore this using the subgroup analyses of age (children/adolescents/adults), sex (male/ female), baseline CD4 count, type of NRTI, dosage, and study design.
For the purposes of this review, undetectable plasma HIV RNA (viral load) will serve as the primary endpoint. For the meta‐analyses, undetectable viral load will be defined as the less stringent 500 copies/mL cut off to include as many trials as possible. Subanalyses may compare drugs using the 50 copies/mL limit of detection.
Sensitivity analysis
Results from the included studies will be pooled to determine the OR and RR for the two regimens of achieving undetectable viral load. A sensitivity analysis will be performed to evaluate bias introduced by variability in study design, threshold of undetectable viral load, and specification of the two NRTIs. Finally, a test for homogeneity will be performed to ensure that the differences among the results of each trial could be expected by chance. A sensitivity analysis also will be conducted for studies with a high risk of bias.
