Nevirapine toxicity in women predicted by liver function, not CD4 count, developing country study reports

Abnormal liver function tests at baseline, not CD4 cell counts over 250 cells/mm³, were predictors for severe liver damage and associated rash researchers among women in Zambia, Thailand and Kenya on a nevirapine-based antiretroviral regimen reported in a multi-country prospective cohort study published in the advance online edition of HIV Medicine.

Close to 70 percent of the estimated 3.5 million people in sub-Saharan Africa and Asia on antiretroviral treatment are on regimens that include nevirapine, a non-nucleoside reverse transcriptase inhibitor (NNRTI).

Nevirapine has proven effective, has no known adverse effects on the unborn child, or long term side-effects. However, it can cause damage to the liver, rash and life- threatening sensitivity, especially in the first few months.

Clinical studies have not given a consistent definition of nevirapine-associated liver damage. Serious liver damage is generally defined in one of three ways, note the authors:

• Severe liver damage: liver function tests defined by levels of serum alanine transferase (ALT) or aspartate transaminase (AST) that are greater than or equal to five times the upper limit of normal (ULN)
• Rash-associated liver damage: a rash associated with a 2.5 fold increase in ALT or AST above ULN
• Fatal liver damage

A retrospective analysis undertaken by Boehringer-Ingelheim, nevirapine’s original manufacturer, found that women with a baseline CD4 cell count over 250 cells/mm³   had a greater risk of rash-associated liver damage than those with a baseline under 250. This led the United States Food and Drug Administration (FDA) to issue a warning against women with CD4 cells counts over 250 to take nevirapine unless benefits outweighed the risks. Some studies have supported Boehringer-Ingelheim’s findings while others have not.

Subsequent to the World Health Organization’s recommendations of 2006 and 2009 increasing numbers of women in resource-poor settings have started and will start nevirapine-containing regimens at CD4 cell counts between 250 cells/mm³ and 350 cells/mm³.

The authors note that in spite of the increased numbers of women starting nevirapine-based antiretroviral regimens in resource-poor settings with CD4 cell counts at or above 250 cells/mm³, few studies have looked at the risk of liver damage among this population.

The Non-Nucleoside Reverse Transcriptase Inhibitor Response study, a prospective cohort study in Zambia, Thailand and Kenya was designed to identify risk factors for severe liver damage (grade 3 or 4) and rash-associated liver damage (rash with grade 2 or above liver damage) among HIV-infected women starting nevirapine-based antiretroviral treatment regimens. At enrolment and at weeks 2, 4, 8, 16 and 24 serum alanine transferase (ALT) and aspartate transaminase (AST) were measured and study participants were evaluated for signs and symptoms of hepatitis and rash.  

113 (14%) of the 820 women (497 Zambian, 192 Thai and 131 Kenyan) who started nevirapine-based antiretroviral treatment had abnormal liver function tests (ALT and AST levels) at baseline.

After starting nevirapine-based antiretroviral therapy a total of 41 (5%) women had grade 3 or 4 events (severe liver damage).  Severe liver damage was seen in 12% (13) of the 113 with abnormal liver function tests at baseline compared to 4% of women who had normal liver function tests at baseline (aOR 3.2 95% CI: 1.4-6.8).

Whereas severe liver damage was seen in 5% of those with CD4 cell counts below 250 cells/mm³ as well as in 5% of those with CD4 cell counts above 250 cells/mm³ at baseline (aOR 1.0, 95% CI: 0.-2.5). In a multivariate analysis other baseline variables that included age, body mass index, viral load, concurrent anti-tuberculosis therapy, WHO clinical disease stage and country were not associated with severe liver damage.

The analysis was repeated for each country and the results were the same as noted above.

Similarly rash-associated liver damage was seen in 7 (6%) of the 113 women with abnormal liver function tests at baseline compared to 20 (3%) of 699 women with normal baseline values (aOR 2.8; 95% CI: 1.1-7.1). However, among Thai women rash-associated liver damage appeared more frequently (7%) compared to Zambian (2%) or Kenyan women (2%) (aOR .4; 95% CI: 1.5-7.8). As with severe liver damage other baseline variables, noted above were not linked to rash-associated liver damage.

The analysis was repeated for each country with similar results. While women with a baseline CD4 cell count at or above 250 cells/mm³ were not at increased risk for rash-associated liver damage differing trends were seen in Zambia (OR 0.5; 95% CI: 0.01-3.8) and Thailand (OR 2.3; 95% CI: 0.4-10.3).

The results also showed a decreased risk of rash-associated liver damage among women with baseline CD4 cell counts of 50-199 cells/mm³ compared with those whose CD4 cell counts were below 50 or at and above 200 cells/mm³.

However, the authors caution this should not be seen as evidence to start nevirapine safely in women with CD4 cell counts of 50-199 cells/mm³.  Of the three participants (0.4%) who died with symptoms suggestive of severe liver damage one had a baseline count within this range (68 cells/mm³). All three were receiving anti-tuberculosis therapy. The other two had baseline CD4 cell counts under 50 cells/mm³.

The authors stress that within resource-poor settings health care workers need to pay attention to nevirapine-associated liver damage in women starting antiretroviral therapy regardless of CD4 count and in particular to concurrent use of anti-tuberculosis therapy . WHO recommends efavirenz is substituted for nevirapine because of less interaction with rifampicin.

Monitoring liver function early (at baseline and at 2 and 4 weeks) suggested that the majority of those at risk for liver damage can be identified.

The authors highlighted a possible alternative to reduce the risks of both liver damage and harm to the unborn child: women with baseline CD4 cell counts at or above 250 cells/mm³ can be started on an efavirenz-based regimen for at least six months then changed to nevirapine. The authors suggest that liver damage may not occur because nevirapine is started after the initial rapid immune recovery on antiretroviral treatment has begun.

Limitations according to the authors include:

• Rash may be more difficult to identify in dark-skinned participants; a standardised rash grading system was used to minimise bias.
• No data was available on other liver damaging toxins, for example alcohol and chronic exposure to aflatoxins (a poison produced by mold that may lead to cirrhosis and liver cancer; found in cereals, oilseeds as well as tree nuts).
• Few women in the study had baseline CD4 cell counts at or above 350 cells/mm³. So these findings are not necessarily applicable to women with higher CD4 cell counts.
• No evaluation was made of the possible effect chronic hepatitis B virus (HBV) co-infection may have had on increasing or confounding the association seen between baseline abnormal liver function tests and the risk of hepatitis after starting nevirpaine.

The authors conclude that among three to five percent of women in three resource-poor settings severe liver damage or rash-associated liver damage was seen in the first 24 weeks after starting a nevirapine-based antiretroviral therapy. Baseline abnormal liver function tests, not CD4 cell counts at or under 250 cells/mm³, predicted both severe liver damage and rash-associated liver damage.  

The authors recommend that in resource-poor settings where liver function (transaminase) testing “is available, testing should focus on early time-points and on women with abnormal baseline ALT or AST results.”

Peters PJ et al. Nevirapine-associated hepatotoxicity was not predicted by CD4 count ≥250 cells/µL among women in Zambia, Thailand and Kenya. HIV Medicine 2010. doi:10.1111/j.1468-1293.2010.00873.x