The high rates of liver side-effects seen in a major study of nevirapine (Viramune) may have been driven by unusual results from one study centre, according to an analysis of the study’s findings by the drug's manufacturer, Boehringer Ingelheim. These findings were presented last month at the 45th Interscience Conference on Antimicrobial Agents and Chemotherapy in Washington.
The study also found that exclusion of patients starting nevirapine treatment with high CD4 cell counts eliminated the differences in side-effect rates between nevirapine and its major competitor, efavirenz (Sustiva). Efavirenz is manufactured by Bristol-Myers Squibb.
The 2NN study was a large, international trial comparing the two non-nucleoside reverse transcriptase inhibitors (NNRTIs) nevirapine and efavirenz in patients who had not taken HIV therapy before. The study contained four arms: patients were randomised to receive nevirapine once daily, nevirapine twice daily, efavirenz or a combination of efavirenz and nevirapine. All of the patients took a background combination of 3TC (lamivudine, Epivir) and d4T (stavudine, Zerit).
Although the study found no significant differences between the two drugs’ anti-HIV effects, it showed that nevirapine was more likely to cause liver problems than efavirenz.
The study also showed that patients taking nevirapine once a day had higher rates of liver toxicity than those taking it twice daily. Once-daily nevirapine is an experimental dose: currently, nevirapine is only licensed for twice-daily dosing.
As other studies did not reveal the same pattern of liver side-effects, Boehringer Ingelheim was keen to find out what was causing the 2NN study’s results. As a previous analysis of the study’s findings had revealed that one of the 65 study centres had higher rates of discontinuation due to side-effects, investigators from the company looked at the rates of liver-related side-effects at this centre in Bangkok.
They found that the increased rates of symptomatic liver side-effects seen in the study could have been driven by the results from the Bangkok site, which contributed 17% of the study’s patients. At this centre, 16% of the patients taking nevirapine once a day had symptomatic liver side-effects, while 2% of those taking nevirapine twice daily and 1% of the patients taking efavirenz did.
In the other 64 sites, only 6% of the patients taking nevirapine once daily had symptoms of liver side-effects. The same proportion of the patients taking nevirapine twice a day had liver side-effects, compared with 3% of those taking efavirenz.
Due to the risk of liver problems in patients starting nevirapine with high CD4 cell counts, current recommendations advise against the use of the drug in men with CD4 cell counts above 400 cells/mm3 and in women with CD4 cell counts above 250 cells/mm3. To investigate whether the application of these recommendations affected the results of the 2NN study, the investigators re-analysed the data after excluding the patients with higher CD4 cell counts.
For the Bangkok site, all of the symptomatic cases of liver side-effects were seen in patients taking nevirapine once a day, with 18% of the patients showing symptoms. No patients in the nevirapine twice-daily or efavirenz arms showed symptoms of liver side-effects.
In contrast, the rates of symptomatic liver side-effects were similar across all three arms in the 64 sites outside Bangkok: 4% of the 119 patients taking nevirapine once daily had liver side-effects, compared with 3% of those taking it twice a day and 4% of those taking efavirenz.
These findings suggest that implementation of the new guidelines for the initiation of nevirapine therapy can result in a similar risk of liver-related side-effects to nevirapine’s main competitor. “In non-Bangkok patients, applying the new nevirapine CD4 cell count criteria at initiation of nevirapine therapy is associated with a reduced rate of hepatic adverse events that is similar to efavirenz,” the investigators conclude.
The investigators also examined the rates of non-symptomatic liver side-effects, determined by the levels of liver enzymes in blood samples. Although the differences between groups were less dramatic than analysis of the symptomatic side-effects, the results followed a similar pattern.
Omission of the results from the Bangkok site also reduced the observed differences between the nevirapine once-daily and efavirenz arms in terms of the drugs’ anti-HIV activity.
Despite the strong contribution of the Bangkok site’s results to the overall conclusions of the 2NN study, the investigators do not offer an explanation for their observations. “The high rates of both symptomatic hepatic adverse events and non-symptomatic hepatic adverse events in the population studied at the Bangkok site warrants further study,” they write.
Any future analysis should hopefully explain whether the difference in reported liver toxicity between Bangkok and non-Bangkok patients has a biological basis, or whether it is a statistical artefact resulting from the way in which the 2NN study was administered.
In particular, it would be useful to know why the Bangkok centre contributed 22% of all participants in the once-daily nevirapine arm. In addition, questions surround why 30% of the 162 patients who received a single NNRTI at Bangkok received nevirapine once a day, compared to 22% of the participants in the overall study (as treated). The effect that these recruitment figures had on the overall result require further investigation.
It would also be useful to discover whether there was any temporal difference in the rate of diagnosed liver events during the first five months of recruitment at the Bangkok centre: during the initial period of the study, 2NN was a three-arm comparison of once-daily NNRTIs with dual NNRTI treatment. The fourth, twice-daily nevirapine arm was added after this phase of the study, when the primary outcome was adjusted to compare twice-daily nevirapine to once-daily efavirenz.
Storfer S et al. Analysis of hepatic events within the 2NN study: controlling for geographic region and CD4+ cell count at initiation of nevirapine therapy. 45th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, abstract LB-13, 2005.