The epidemiology of hepatitis B virus infection in HIV-infected and HIV-uninfected pregnant women in the Western Cape, South Africa

Abstract

Objectives: Persistent hepatitis B virus (HBV) infection is a major cause of morbidity and mortality in sub-Saharan Africa. The HIV epidemic has the potential to affect its biology. Immunisation protocols established in the pre-HIV era are based upon data showing predominantly horizontal infant transmission. This study aimed to determine whether HIV co-infection will change the epidemiology of HBV both by increasing infectivity and by favouring the escape of viruses bearing phenotypically altered HBsAg.

Methods: This retrospective cross-sectional study used antenatal samples from the 2008 Antenatal Sentinel HIV and Syphilis Prevalence Survey in the Western Cape, South Africa. All HIV-infected women were age and race-matched to HIV-uninfected women. Samples were tested for serological markers of HBV and HDV infection. HBV viral load, consensus sequencing and genotyping were performed. Luminex technology was used to determine HBsAg phenotype. All samples from HIV-infected women were tested for traces of antiretroviral drugs by mass spectrometry.

Results: This study showed a trend toward loss of immune control of HBV in HIV-infected women with 3.4% of samples containing HBsAg, 18.9% contained HBeAg. In contrast, 2.9% of samples from HIV-uninfected women contained HBsAg and 17.1% of these HBeAg. The median HBV load in the HIV-infected group was 9.72×10(7)IU/ml and in the HIV-uninfected group 1.19×10(6)IU/ml. Genotyping showed 63/68 samples belonged to genotype A and the remainder genotype D. Mutations in the precore region were found in 35% and 33% of samples from HIV-infected and HIV-uninfected respectively. Although no major epitope ablation was found, marked variation in HBsAg profiles in HIV-infected group was demonstrated. No HDV infection was detected.

Conclusion: HIV-HBV co-infected women exhibit a degree of immune escape. One in six HBV-infected pregnant women, irrespective of HIV status is HBeAg seropositive. HBV immunization of newborns in sub-Saharan Africa should be implemented.

Keywords: Antenatal; HBV; HIV; Sub-Saharan Africa; Virus escape.

Fig. 1

Median plasma HBV viral loads (expressed in IU/ml) according to HIV status in HBeAg positive (left) and in HBeAg negative, anti-HBe positive (right) samples displayed as box (interquartile range (IQR)) and whisker (range) plots.

Fig. 2

‘Hot spot’ analysis of amino acid changes across the HBsAg. Solid bars are sequences from 25 HIV-uninfected women and the hatched bars from HIV-infected women. The circled area indicates amino acid changes in the major antigenic region between codons 120 and 150.

Fig. 3

Epitope variance of plasma HBsAg expressed as box (middle quartiles) and whiskers (range) in 20 HIV-uninfected and 33 HIV-infected women. Reactivity against each monoclonal antibody solid phase is shown independently (MAb P2D3 is against a linear epitope and MAbs HB07 and HB05 are against different conformational determinants in the second loop of HBsAg) and is expressed as the percentage reactivity of each mAb as part of the total reactivity for that sample. ● denotes those samples whose reactivity fell 2 standard deviations outside the expected range for that specific MAb.