Appropriate use of nevirapine for long-term therapy.

Correspondence Figure 1. Frequency of early symptomatic hepatic events in controlled and uncontrolled trials with nevirapine. Data are from Boehringer Ingelheim's Expanded Hepatic Analysis and are included in EU/US VIRAMUNE prescribing information; " early symptomatic events " refers to events occurring during the first 6 weeks of treatment. To the Editor—The recent article by Sanne et al. [1] described possible risk factors associated with hepatic adverse events that can occur while nevirapine is being taken in a clinical trial. We are concerned about several key issues either overlooked or minimized in the article. It is important to understand that the preliminary results of the FTC-302 study provided the impetus for Boehringer Ingel-heim (BI) to conduct a comprehensive analysis of hepatotoxicity in all large BI controlled and uncontrolled studies. This analysis resulted in warnings that female sex and higher CD4 cell count at the initiation of therapy increases the risk of hepatotoxicity, particularly during the first 6 weeks of treatment. This information has been widely communicated through changes to the prescribing information and through BI communications to physicians. Most recently, BI has recommended against the initiation of nevirapine treatment in women with CD4 cell counts 1250 cells/mm 3 or in men with CD4 cell counts 1400 cells/ mm 3 unless the benefit outweighs the risk (see http://www.fda.gov/cder/drug/ advisory/Nevirapine.htm). As shown in figure 1, there is a clear demarcation in the risk of developing nevirapine-associated symptomatic hepatitis at the CD4 cell count cutoffs mentioned above. In our comprehensive analyses, the frequency of patients with symptomatic he-patic events in the lower CD4 cell count groups (women with counts !250 cells/ mm 3 and men with counts !400 cells/ mm 3) was significantly reduced and was consistent with frequencies found in association with other antiretrovirals, including efavirenz (1%–2%). We are concerned that, because study FTC-302 excluded patients with CD4 cell counts !200 cells/mm 3 , the major predictive factor of hepatic events was removed from consideration, therein biasing the au-thors' conclusion that a low body mass index (BMI) is predictive of nevirapine hepatotoxicity. Additionally, there was only a small number of patients with BMIs !18.5 (4.4% of total), and these accounted for only a small proportion of patients with hepatotoxicity (11.3%). This result suggests that there were other explanations for the high rate of hepatic events observed in this study, including the mean CD4 cell count of 406 cells/mm 3 in patients with early …