The impact of HBV or HCV infection in a cohort of HIV-infected pregnant women receiving a nevirapine-based antiretroviral regimen in Malawi

Patient population included women enrolled in the SMAC (Safe Milk for African Children) study, an observational study conducted in Malawi and aimed to assess the safety and pharmacokinetics of ART (Anti Retroviral Therapy) administration to breastfeeding women [9]. The study was conducted within the DREAM (Drug Resources Enhancement against AIDS and Malnutrition) Program, designed and managed by the Community of S. Egidio, an italian faith-based non-governmental organization. The study was approved by the National Health Sciences Research Committee of Malawi (approval number #486). A separate informed consent was signed by all participants. HIV-positive pregnant women, naive to antiretrovirals, older than 16, with no baseline grade 3 or 4 laboratory toxicity and no concomitant treatment for tuberculosis were eligible for the study. ART was administered as soon as possible after the first trimester if the women met the criteria for treatment (CD4+ < 350/mm³), and treatment was continued indefinitely after pregnancy. The women not needing treatment for themselves received antiretroviral prophylaxis from week 25 of gestational age until 6 months after delivery, which represented the recommended duration of breastfeeding at the time of the study (enrolment occurred between February 2008 and February 2009). All women received a combination regimen based on lamivudine (3TC, 150 mg twice daily) and nevirapine (NVP, 200 mg twice daily) plus either stavudine (d4T, 30 mg twice daily) or zidovudine (ZDV, 300 mg twice daily). Women experiencing nevirapine-associated toxicity were allowed to substitute this drug with lopinavir-ritonavir (LPV/r, 400/100 twice daily). Women were followed for 2 years after delivery. Complete blood count (Sysmex KX-21N hematology analyzer, Sysmex Corporation, Kobe, Japan) and liver enzymes (Humastar 180 Chemistry Analyzer, Human Diagnostics, Germany) were determined every 2 weeks for the first 2 months and every month thereafter during pregnancy, at 1, 3 and 6 months postpartum and then every three months until 24 months. CD4+ cell count (Epics XL-MCL, Beckman Coulter, Brea, CA, USA) and HIV-RNA determination (branched-DNA version 3.0 assay, Siemens Diagnostics, Deerfield, IL, USA) were performed every 6 months after delivery. Laboratory abnormalities were classified according to the Division of Acquired Immunodeficiency Syndrome/National Institute of Allergy and Infectious Diseases (DAIDS/NIAID) grading system, 2004 revision [10]. An ALT level of 1.25-2.5 times the upper limit of normal (ULN) defined grade 1 hepatoxicity, 2.6-5 times the ULN defined grade 2, 5.1-10 times the ULN defined grade 3, and > 10 times the ULN defined grade 4 hepatoxicity.

For HBV, women were tested using the Murex HBsAg Version 3 with Confirmatory Assay (Murex Biotech, Dartford, UK). The presence of HBeAg was assessed using the ARCHITECT HBeAg Reagent Kit (Abbott Diagnostics Division, Wiesbaden, Germany). Serum HBV-DNA was quantified using the COBAS TaqMan HBV Test (Roche Molecular System, Branchburg, NJ, USA) with a limit of quantification of 6 IU/ml. For HCV all sera were first tested with the Innotest HCV Ab IV assay (Innogenetics, Ghent, Belgium). Low and high-positive results were re-assayed using the Ortho HCV 3.0 ELISA test (Ortho-Clinical Diagnostics, Raritan, NJ, USA). Samples reactive with both assays were considered anti-HCV positive [11]. In anti-HCV positive women the presence of HCV-RNA was tested in at least 3 different samples using the Versant HCV-RNA 1.0 assay (Siemens Diagnostics) with a detection limit of 15 IU/ml. Only patients with detectable HCV-RNA were considered HCV-positive.

Characteristics in the hepatitis coinfected and the HIV monoinfected groups were compared using the Mann–Whitney test for continuous variables and the chi square or Fisher’ s exact test for categorical variables. The Kaplan-Meier method was used to define the time to the development of liver toxicity and the Cox proportional hazards model to identify factors associated with its emergence. Variables were included in the multivariate model if they were significant at a P value of less than 0.25 in univariate analysis. For all statistical tests, two-sided P values less than 0.05 were considered to be significant. Data analysis was done using the SPSS software system 20.0 (IBM, Somers, NY, USA).

Over the total follow-up time 125 women (40.5%) developed a new grade ≥ 1 ALT elevation (15/28 or 53.6% coinfected women and 110/281 or 39.1% HIV + only), 28 (9.1%) (2 coinfected and 26 HIV + only) a grade ≥ 2, and 6 (1.9%) (all HIV + only) a grade 3 toxicity. Twelve % of the cases occurred during pregnancy and 57.6% within 6 months of delivery (83.3% of the cases in coinfected women and 53.2% of the cases occurring in monoinfected women). In a Kaplan-Meier univariate analysis, the risk of developing an ALT elevation of any grade was higher for HBV or HCV infected women (P = 0.033 by log-rank test), while no significant association was found for grade ≥ 2 liver toxicity (P = 0.706) (Figure 1). In a multivariate Cox model that adjusted for age, CD4+ count > or < 250/mm³ and hemoglobin level, the presence of hepatitis B or C remained significantly associated with the development of liver toxicity of any grade (hazard ratio [HR]: 1.797, 95% CI 1.042-3.098, P = 0.035) (Table 2). In a similar multivariate model, based on the same variables reported above, but that considered only grade ≥ 2 ALT elevations, the only predictor of the emergence of toxicity was a baseline CD4+ cell count > 250/mm³ (HR: 3.296, 95% CI 1.125-9.655, P = 0.030).

Fifteen cases of ALT increase of any grade and 8 cases of grade ≥ 2 increase developed during pregnancy. Thirteen out of the 28 (46.4%) total cases of grade ≥ 2 toxicity developed within the first 18 weeks of treatment. Of these cases, 12 occurred in the 211 women with baseline CD4+ > 250/mm³ and 1 among the 98 women with baseline (CD4+ ≤ 250/mm³, P = 0.069). There was no difference in the occurrence of toxicity after 18 weeks according to the baseline CD4+ stratum.

In HBV-infected women, after controlling for age, CD4+ count above or below 250/mm³ and hemoglobin level, a baseline HBV-DNA level > 10,000 IU/ml was significantly associated with the development of liver toxicity of any grade (HR:3.278, 95% CI 1.056-10.180, P = 0.040). HBV-DNA measurements during treatment were available for 20 women at Month 6 and for 14 women at Month 12 postpartum. At 6 months 9/20 had a detectable HBV-DNA level (but only 3 had levels > 10,000 IU/ml) while at 12 months 8/14 had detectable HBV-DNA levels (5 had levels above 10,000 IU/ml), but it has to be considered that 6 of these women, not meeting the criteria for treatment, had interrupted treatment 6 months postpartum. Women with detectable HBV-DNA at 6 months had a higher median ALT level at 12 and 24 months compared to those with undetectable HBV-DNA (59.2 vs 18.5 IU/ml and 35.9 vs 18.2 IU/ml, respectively, P = 0.001 and P = 0.043).

There was no difference in the probability of developing liver toxicity of any grade between women receiving zidovudine or stavudine.

Among the 3 women who discontinued nevirapine because of liver toxicity none was coinfected.

Among the 6 women who died during follow-up, one was coinfected with HBV (no death was liver-related). There was no difference in the probability of survival between the 2 groups (P = 0.511 by log-rank test) (Figure 2).

Among the 237 women with HIV-RNA determination 6 months postpartum, 21/23 coinfected women vs 160/214 HIV + only had < 50 HIV-RNA copies/ml (P = 0.118). No difference was detected also considering the proportion of women with HIV-RNA < 1,000 copies/ml (21/23 vs 197/214, P = 1.0). Among the women continuing therapy, the proportion of women with < 50 copies/ml or < 1,000 copies/ml was not significantly different between the coinfected and the monoinfected populations at Month 12 or 24 (Figure 3).

Median CD4+ increase at Month 6 was 185 cells/mm³, with no difference between the two groups (185 cells in the monoinfected vs 179 cells in the coinfected group, P = 0.819) (Figure 4). Among the women continuing therapy, the median CD4+ increases at Month 12 and Month 24 were 218 and 294 cells, respectively with no difference between those coinfected and those HIV + only (279 vs 217 cells, and 439 vs 294 cells, respectively, P = 0.123 and P = 0.172).