December 17, 2005
In 2003, the seminal 2NN Study taught us that dual non-nucleoside reverse transcriptase inhibitors (NNRTIs) -- specifically efavirenz (EFV, Sustiva, Stocrin) and nevirapine (NVP, Viramune) -- are not better than single NNRTIs. 2NN also lent support to the use of nevirapine for first-line HIV treatment and took some steam out of the efavirenz train.
However, one of the adverse effects associated with nevirapine in 2NN -- hepatotoxicity -- has since become an Achilles' heel of sorts for the drug. A review of all the nevirapine studies to date revealed that the risk for hepatotoxicity was highest with therapy initiation in women with a CD4+ cell count of 250 or more and in men with a CD4 + cell count of 400 or more. This led to a change in the Dept. of Health and Human Services' guidelines on HIV treatment that now recommends that nevirapine be avoided in treatment-naive patients meeting these CD4+ cell criteria.1
This study by Stephen Storfer, of Boehringer Ingelheim, and colleagues sought to apply our current knowledge of nevirapine-associated hepatotoxicity to the original 2NN data, in hopes of better understanding the hepatic abnormalities discovered in 2NN. What they discovered was that something strange appears to have occurred in one of the study's 65 sites -- and that there's no easy explanation for the occurrence.
A Look Back: The 2NN Study
Therapy consisting of a single NNRTI with two nucleoside reverse transcriptase inhibitors (NRTIs) had become standard of care by the late 1990s. However, no large trial had compared the two most commonly used NNRTIs -- nevirapine and efavirenz -- either head-to-head or combined with each other as dual-NNRTI therapy. 2NN was designed to do just that, as well as to examine the role of nevirapine dosed once daily instead of the usual twice daily.
The 2NN trial enrolled 1,216 treatment-naive patients from 65 trial sites spanning 17 countries in six continents. Study patients were randomized into one of four arms: nevirapine once daily, nevirapine twice daily, efavirenz once daily or a combination of nevirapine and efavirenz twice daily. All patients received an NRTI backbone of stavudine (d4T, Zerit) and lamivudine (3TC, Epivir).2
The original analysis revealed no difference in efficacy, as measured by virologic failure, between the nevirapine and efavirenz arms. However, the nevirapine once-daily arm was found to have a higher incidence of hepatic laboratory abnormalities designated as asymptomatic hepatic adverse events (AsymHAE) compared to the nevirapine twice-daily or efavirenz arms. Subsequent analysis also found symptomatic hepatic adverse events (SymHAE) to be higher in the nevirapine once-daily arm than the nevirapine twice-daily arm.3
Other trials, such as ATLANTIC, found lower rates of asymptomatic and symptomatic hepatic adverse events with nevirapine once daily than the 2NN trial indicated, which prompted the developer of nevirapine, Boehringer Ingelheim, to perform further study subanalysis.4 They uncovered an association between adverse event-related treatment failure and geographic region; in fact, they discovered that nevirapine-associated adverse events came primarily out of one trial center in Bangkok, Thailand.5
This current subanalysis, presented by Dr. Storfer, was designed to tease out the relationship between adverse events and geography. Dr. Storfer et al sought to reexamine the 2NN data while controlling for geographic region and CD4+ cell count at initiation of therapy, given that we now know that a higher incidence of hepatic adverse events is correlated with a CD4+ cell count of 400 or higher in men and 250 or higher in women.4
Hepatotoxicity in Bangkok: Does It Hold Up to the New Criteria?
For the sake of clarity, comparisons to the dual-NNRTI arm of the original 2NN study were omitted here. In this safety analysis, hepatic adverse events from 2NN were examined based on the same rules used in the Viramune Expanded Hepatic Analysis.3 These rules, which were used to identify possible cases for clinical review, were designed to be as inclusive as possible. They focused on two broad categories of adverse events:
In total, 121 patients (10.4%) had events that met these criteria. Their cases were then reviewed by two blinded reviewers, with a third reviewer used only if a consensus could not be reached.
Asymptomatic hepatic adverse events were defined as those with ALT/AST elevations of five times or greater that were not associated with clinical symptoms of hepatitis.
Rates were then analyzed by geographic site; specifically, the Bangkok site was compared to the 64 other trial sites as a whole.
Rates were also analyzed based on whether or not patients would be started on nevirapine by current CD4+ cell count criteria (i.e., avoid nevirapine at a CD4+ cell count of 400 or higher in men and 250 or higher in women), criteria which did not exist when the initial 2NN Study took place.
2NN was originally a three-arm study. The fourth arm, nevirapine twice daily, was added five months after study initiation, with the primary outcome adjusted to test for equivalence between the new arm and the efavirenz once-daily arm. Treatment randomization was recalculated. Boehringer Ingelheim feels that the differences between the three-arm phase and the four-arm phase are too great to consider them one study, so analyses contrasting nevirapine twice daily against the other study arms are based only on the four-arm phase. (Analyses comparing efavirenz against once-daily nevirapine are based on the whole trial.)
Efficacy differences were derived by subtracting the response to the first arm from the response to the second arm. Treatment response was defined as a confirmed HIV RNA below 50 copies/mL at 48 weeks without any discontinuation of NNRTI, incident AIDS-defining illness or confirmed rebound followed by return to HIV RNA below 50 copies/mL.
The Bangkok site contributed 162 of the 967 patients who received a single NNRTI in the 2NN Study. A comparison of the rates of hepatic adverse events appears in the chart below.
|SymHAE % (AsymHAE%)|
|Population||CD4+ Cell Count Criteria||NVP QD||NVP BID||EFV QD|
|Total 2NN||None||8.6 (7.0)||5.6 (4.9)||2.1 (3.8)|
|Bangkok||None||16.3 (10.2)||2.3 (9.3)||1.4 (5.8)|
|Women < 250 cells/mm3|
Men < 400 cells/mm3
|18.4 (7.8)||0.0 (9.3)||0.0 (5.8)|
|64 Other Sites||None||6.2 (5.6)||6.3 (4.2)||2.6 (3.2)|
|Women < 250 cells/mm3|
Men < 400 cells/mm3
|4.2 (2.8)||3.2 (3.9)||3.6 (3.9)|
When comparing hepatic adverse event rates within each of the populations above, the following were the only pairwise comparisons that were statistically significant (P<.05):
No statistical significance was found after the new CD4+ cell count initiation criteria was applied to the Bangkok site and the 64 other sites.
No statistically significant differences in treatment response or virologic response (defined as having HIV RNA below 50 copies/mL) were seen when the researchers compared nevirapine once daily to efavirenz, nevirapine twice daily to efavirenz or nevirapine once daily to nevirapine twice daily. This was true for the "Total 2NN" population as well as the "64 Other Sites" population.
Compared to all other sites in the trial, an excess of hepatic events was seen among patients in the nevirapine once-daily arm at the Bangkok site (even after applying the CD4+ cell criteria [see table]). By contrast, in the other 64 trial centers, rates of symptomatic hepatic events in the nevirapine and efavirenz arms were quite similar.
Additionally, when the new CD4+ cell count initiation criteria is applied, this excess in hepatic events clears, leaving behind comparably low rates of symptomatic and asymptomatic hepatic events in these 64 centers. Only in the Bangkok site did the excess remain. Efficacy analyses revealed no differences in treatment response or virologic response between treatment arms and treatment sites for single NNRTIs.
At the non-Bangkok sites, use of the CD4+ cell count criteria resulted in a reduced rate of hepatic adverse events on nevirapine; these rates dropped to a level similar to that seen with efavirenz once daily. Again, however, the Bangkok site was different; no significant decrease was discovered in these events after the criteria was applied; in fact, there was actually an increase in symptomatic hepatic adverse events in the nevirapine once-daily arm.
The Bangkok Mystery Deepens
This new analysis indicates that something unusual occurred among the 2NN study patients at the Bangkok study site. Even after applying the new CD4+ cell count exclusion criteria, rates of hepatic events both symptomatic and asymptomatic were much higher in Bangkok than the other 64 study sites combined. While this study confirmed the unusually high rates at a single study site, further research will need to be done to uncover the reason for the difference.
Were there racial/ethnographic differences that resulted in impaired/altered hepatic metabolism? Was low body mass index an issue? Were there other, less obvious causes, such as dietary differences or concomitant use of hepatotoxic herbal remedies? Was hepatitis B or C coinfection, while low overall in the study, an issue at this particular site? All of these issues need to be examined to determine why toxicity rates should be so different at one specific site.
Beyond the Bangkok conundrum, this study highlights the importance of clinical use of CD4+ cell count criteria to reduce nevirapine hepatotoxicity risk. After all, in all sites except Bangkok, application of the criteria virtually eliminated the difference in hepatic events between the nevirapine arms and the efavirenz arm.
This subanalysis shows us that 2NN can teach us much more than the original, most obvious conclusion, which was: Don't use two NNRTIs in the same regimen. We can look forward to further subanalysis and future research into NNRTI hepatotoxicity based on this trial.
|Please note: Knowledge about HIV changes rapidly. Note the date of this summary's publication, and before treating patients or employing any therapies described in these materials, verify all information independently. If you are a patient, please consult a doctor or other medical professional before acting on any of the information presented in this summary. For a complete listing of our most recent conference coverage, click here.|