HIV-1 infection in Tanzania : HIV antibody testing strategies and lymphocyte subset determinations

Abstract

The objectives of the studies that are reported in this thesis were to evaluate HIV antibody testing strategies and prognostic markers and to study inummological deterioration of HIV-1 infected individuals. Alternative testing strategies for confirmation of HIV seropositivity were evaluated using combinations of screening assays. Wellcozyme recombinant, Organon Vironostika, Abbott recombinant and Enzygnost anti-HIV-1/2 enzyme-linked immunosorbent assays (ELISAs) and two simple/rapid assays, HIV-CHECK and Serodia were evaluated using 1,861 sera. Several combinations of these assays were found to have a diagnostic accuracy comparable to a screening assay followed by Western blot analysis (WB). The cost of confirming a positive sample was 2-3 times less compared to ELISA followed by WB analysis. A modified WHO alternative confirmatory testing strategy was also evaluated using two consecutive ELISAs, Enzygnost HIV-1/2 and Wellcozyme HIV-1 recombinant assays.

It was demonstrated that a second ELISA based on different antigens and a different test principle could be used as an alternative to the WB assay for confirmation of HIV antibodies. Three simple/rapid 141V assays, Capillus, Serocard and Determine were evaluated using 1412 fresh sera to formulate an alternative confirmatory strategy for the diagnosis of HIV infection. All assays had a sensitivity of 1OO%. In an alternative confirmatory strategy the use of Capillus followed by Serocard or Determine gave a specificity of 99.9% and 99.8% respectively. Serocard followed by Determine gave a specificity of 99.3%. A testing strategy with 100% specificity (95% Cl; 99.6- 100%) could be achieved by the sequential use of all three simple/ rapid assays or by repeat testing by Capillus followed by Serocard.

Immunological and virological predictive markers associated with an increased risk of mother-to-child transmission (MTCT) of HIV-1 were determined among Tanzanian women. T lymphocyte subsets, p24 antigenemia and beta-2 microglobulin (beta-2M) levels were measured in HIV-1 infected mothers. Among 138 HIV-1 seropositive mothers 20.7% transmitted HIV infection to their offspring. Mothers with CD4+ T lymphocytes <20% and beta-2M > 2.0 mg/I had a significantly higher rate of MTCT of HIV-1 infection (54%). HIV-1 p24 antigenemia was significantly more common in transmitting mothers than in non- transmitting mothers [14% vs. 2.3% (p--0.05)]. Therefore, low CD4+ T lymphocytes% and a high level of beta-2M, especially when present in combination, appeared to be useful predictive markers for MTCT of HIV-1 in Tanzanian women.

Reference values of lymphocyte subsets were determined in healthy HIV antibody negative individuals. In addition two methods for enumeration of lymphocyte subsets, SimulSET and MultiSET, were compared using whole blood from both HIV seronegative and seropositive individuals. In seronegative Tanzanian subjects, the percentages of CD3+ and CD4+ T lymphocytes and the CD4+: CD8+ T lymphocyte ratios were significantly lower while the percentage of natural killer cells was significantly higher compared to the levels of the corresponding parameters reported among Europeans. Seronegative Tanzanian females had significantly higher levels of CD3+ and CD4+ T lymphocytes and CD4+: CD8+ T lymphocyte ratios compared to seronegative males. There was acceptable agreement between the SimulSET and MultiSET flow cytometric methods.

The rate of decline of CD4+ T lymphocytes was determined in a cohort of hotel workers. The mean duration of followup in workers who had three or more CD4+ T cell determinations and were used for calculations of CD4+ cell slopes was 71.4, 52.9 and 86.0 months for HIV-1 seroprevalent (n=94), HIV-1 seroincident (n=77) and seronegative subjects (n=325), respectively. The median yearly decline of the absolute number and percentage of CD4+ T lymphocytes (CD4+%) was -21.0 cells/µl and -1.3% respectively for the HIV-1 seroprevalent subjects and -22.0 cells/µl and -1.5%, respectively for the seroincident subjects. In seroincidents the mean duration to a CD4+ T lymphocyte level corresponding to definition of AIDS was 13.3 years or 11.8 years for CD4+ counts and percentage, respectively. HIV-1 seroprevalent individuals who died had a yearly median loss of -52.0% CD4+ T lymphocytes /PI and -2.9% CD4+ T lymphocytes compared to a median loss of -18.5 CD4+ T lymphocytes /µl and -1.2 % CD4+ T lymphocytes for those who survived (p = 0.105 and p<0.01 respectively). Among all the HIV-1 seropositive individuals followed up for a minimum of five years, 14. 1 % did not show a net decline of CD4+ T lymphocytes. In conclusion, the rate of CD4+ T lymphocyte decline in HIV-1 infected individuals in our population is similar to that reported in Europe and North America and a substantial proportion of HIV-1 infected individuals in this population are long-term-non-progressors.