Toxicity of nevirapine-based ART differs in pregnant HIV-infected Mozambican and Thai women with higher CD4 counts

There may be differences in the rates of severe hepatotoxicity associated with nevirapine-containing highly active antiretroviral therapy (NVP-ART) in pregnant women with higher CD4 counts (above 250 cells/mm³) in Mozambican and Thai women, according to the findings of a prospective study published in the April 1^st edition of the Journal of Acquired Immune Deficiency Syndromes and recent Thai findings reported by Thai and US researchers at the 2007 Conference on Retroviruses and Opportunistic Infections in Los Angeles, USA.

ART has been successfully used for PMTCT in developed countries where HIV transmission to babies have been reduced to less than 2 %. Due to concerted national and international efforts, antiretroviral therapy is being rolled out in resource-poor countries for PMTCT. The World Health Organization (WHO) recommends that pregnant women with CD4 counts below 350 cells/ mm³ who have stage I or II clinical disease should receive ART during pregnancy.

Single dose nevirapine or NVP-ART has become the mainstay of PMTCT. However, its use in continuous antiretroviral therapy is fraught with an increased risk of hepatotoxicity or skin toxicity in women with higher CD4 counts (above 250 cells/ mm³). Furthermore, the risk of hepatotoxicity is higher in pregnant women than in non-pregnant women.

In Mozambique, the national treatment policy recommends NVP-ART as a first line therapy for HIV-infected pregnant women with CD4 counts equal to or less than 350 cells/ mm³). A team of Mozambican and US researchers investigated the association between drug toxicity and maternal CD4 counts in Mozambican women receiving NVP-HAART for PMTCT.

The study took place in Maputo in Mozambique. The study subjects were HIV-1-infected pregnant women and their infants. Women were recruited at the antenatal clinic, offered voluntary counseling and testing for HIV-1, and seropositive women were referred to the study hospital for CD4 cell count measurement. Women who did not meet the inclusion criteria received treatment or prophylaxis even though they were not study participants.

At enrollment, an oral questionnaire for demographic, socio-economic, and health information was administered. Physical examination included an assessment for pregnancy, overall health, and WHO clinical AIDS disease staging.

Baseline laboratory investigations included liver and kidney function tests, complete blood cell counts, urine analysis, and syphilis testing. The study participants received iron, folate, multivitamins, mebendazole, and, when needed, treatment against sexually transmitted infections.

Two weeks after enrollment, 146 out of the 163 HIV-1 infected pregnant women initiated HAART. The median gestational age at enrollment was 26 weeks (IQR 21-31 weeks) and the median CD4 count was 231 cells/ mm³ (IQR 160-286 cells/ mm³). Most of the study population (57 %) had WHO stage I or II disease, 40 % had stage III and 3 % had stage IV disease. The mean duration of follow up from the initiation of HAART was 39 weeks (range 8-58 weeks).

The participants were seen one to two weeks after the initiation of HAART in order to assess adherence and side-effects. Thereafter, antenatal follow up visits took place monthly until 34 weeks of gestation, and then every fortnight until delivery. During antenatal visits, information was collected regarding disease status, side effects, and severe adverse events. Liver function tests, hemoglobin (Hb) analysis for anaemia, and CD4 counts were carried out at defined intervals after starting HAART.

Severe hepatotoxicity (grade III or IV) was defined as aspartate aminotransferase (AST) or alanine aminotransferase (ALT) levels  5 times the upper limit of normal; moderate (grade II) and mild (grade I) hepatotoxicity were defined as AST or ALT concentrations > 2.5 times and > 1.25 times the upper limit of normal, respectively. The overall rates of NVP-ART-induced hepatotoxicity were high with 11/146 pregnant women developing > grade II toxicity. Severe hepatotoxicity was experienced by 4 women (3%) who all had baseline CD4 counts > 250 and less than or equal to 350 cells/ mm³ (p = 0.02).

Skin toxicity was seen in 10% of the women; 3% experienced the severe Stevens-Johnson syndrome. A greater proportion of Stevens-Johnson syndrome was associated with CD4 counts greater than 250 cells/mm³.

Severe neuropathy was defined as acute numbness or pain of the finger tips and toes. Severe and moderate anaemia were defined as hemoglobin 3.

Side effects and toxicities were managed according to national guidelines. The treatment regimen was changed in case of severe anemia and liver, skin, or neuro-toxicity.

The Thai study was carried out to assess the risk of NVP-HAART-induced hepatotoxicity in pregnant Thai women with a CD4 count of 250-350 cells/mm³. Moderate and mild (grade I-II) hepatotoxicity, asymptomatic hepatotoxicity, and grade III-IV rash were more frequent in pregnant women with CD4 counts > 350 cells/mm³ The poster did not report any severe hepatotoxicity. Overall, the study did not find any association between toxicity and CD4 counts of 250-350 cells/mm³.

The discordant findings of the Mozambican and Thai studies might be due to methodological differences such as the definitions of the various toxicities and follow-up schedules. Demographic, socio-economic, and genetic differences between the study populations as well as differences in co-infections might also have played a role. While it is not clear whether any of these possibilities account for the contradictory findings, it would be useful for similar studies at different sites to follow standardised clinical and laboratory protocols to facilitate the generation of results which might further inform PMTCT treatment policy.

The findings of both Thai and Mozambican studies have two important policy messages. NVP-ART can be effectively deployed in resource-poor settings for PMTCT. However, the severe hepatotoxicity observed in Mozambican patients with higher CD4 counts calls for a more closer laboratory monitoring as part of health care during PMTCT in Africans. In this regard, the need for low-tech assays for laboratory monitoring in resource-poor settings cannot be over-emphasised.