Immunogenicity and Safety of Different Dosing Schedules of Trivalent Influenza Vaccine in HIV-infected Pregnant Women

Determining the contribution of influenza to early childhood morbidity and mortality in sub-Saharan Africa and the potential to prevent influenza disease through vaccination may contribute to reducing childhood deaths; since influenza illness is a vaccine preventable disease for which vaccines are developed, licensed and available at reasonable cost. Unfortunately, infants at highest risk for serious disease are those under 6 months of age, for whom trivalent inactivated influenza vaccine (TIV) is poorly immunogenic and not licensed. As pregnant women also have an increased risk of serious illness (3.3-5.5 fold for hospitalization for influenza-associated acute cardio-respiratory illness) from influenza infection, one strategy to prevent the complications of influenza in pregnant women and young infants is through maternal TIV immunization, which is recommended by the WHO. This could result in direct protection of the women and protection of the young infant consequent to transplacental transfer of TIV induced antibody.

Barriers to administration of vaccines during pregnancy including lack of information on effectiveness and concerns about safety probably explain the virtual non-existent use of TIV in pregnant women from low-middle income countries, including South Africa. Although there was a national campaign for influenza vaccination of pregnant women in South Africa during 2010 due to the concern of continued circulation of H1N1-2009pdm influenza virus, the uptake of vaccine remains poor. TIV immunization of pregnant women is still not provided as standard of care to pregnant women attending antenatal-clinics in South Africa, in part because of absence of any data from any African setting with regard to its risk-benefit ratio.

The immunogenicity and efficacy of TIV in HIV-infected adults was only recently documented in an African setting. This placebo-controlled, community-based randomized, placebo controlled trial, conducted at Themba Lethu HIV clinic, Helen Joseph hospital, reported that TIV was associated with a 75% reduction in influenza-confirmed illness. The results of the study also confirmed the safety of TIV among African HIV-infected adults. The study, however, only included 7 women who were pregnant. In addition to no differences in solicited adverse event rates, there was also no difference in either cluster of differentiation 4 (CD4+) cell count changes or HIV viral control in those on antiretroviral treatment between TIV vaccinees compared to placebo recipients. This allayed previous concerns regarding the potential negative effect of TIV which centered around the observed transient increase in HIV-1 viral load, even in HIV infected individuals on ART and who were virologically suppressed (viral load <400 copies/ml). Decreases in CD4+ lymphocyte counts have also been observed in HIV-infected individuals post TIV vaccination. These changes, however, even in past studies were infrequent (4-18%) and resolved at later time-points and were considered to be clinically non-significant.

Only recently has data become available from another low-income country (Bangladesh) in which the benefit of maternal TIV vaccination was demonstrated by a 63% (95% confidence interval (CI) 5 to 85) reduction in laboratory-confirmed influenza illness in infants under 24 weeks of age in children born to mothers vaccinated with TIV and a 36% reduction in clinical illness in vaccinated mothers. There has, however, not been any study on the effectiveness of maternal immunization with TIV on influenza- associated morbidity and mortality either in the mothers or infants in African settings.

Much of influenza virus associated morbidity and mortality may be due to the synergistic lethality of influenza with bacterial pathogens leading to pneumonia as well as other viral co-infections. Superimposed bacterial infections, especially Streptococcus pneumoniae and in patients treated with antibiotics, Staphylococcus aureus, contribute to a large proportion (28-65%) of pneumonia deaths associated with influenza illness during pandemics. Unpublished data from the Bangladeshi study show that infants of mothers who received TIV (with pneumococcal conjugate vaccine, 7 valent (PCV7) given to infants) were better protected from acquisition of pneumococcal carriage during influenza season than infants of TIV-unimmunized mothers. No data in the African setting are available to support or refute this observation from Bangladesh. As introduction of pneumococcal conjugate vaccines (PCV) in low-income countries is a priority of the Global Alliance for Vaccines and Immunization (GAVI), with 7 African countries already approved, and many others to follow, to introduce PCV within the next five years.

Despite the encouraging results on maternal immunization from Bangladesh, and the preliminary data supporting that TIV is efficacious mainly in HIV-infected non-pregnant adults, further data are needed to advocate for routine use of TIV during pregnancy in settings with a high prevalence of HIV. Reasons for this include that the impact of maternal HIV on the kinetics of TIV induced transplacental antibody transfer cannot be derived from available data. This is important as the primary focus of this proposal, and major potential public health benefit of maternal TIV vaccination, is targeted at protection of young infants. HIV infection is known to decrease placental integrity and lower antibody levels in the fetus and newborn. Furthermore, maternal hypergamma-globulinemia that is characteristic of HIV-infection may be associated with decreased neonatal antibody levels. This paradox is explained by the limited number of placental antibody receptors, resulting in immunoglobulin G (IgG) antibodies competing for available receptors and thereby decreasing vaccine-specific antibody transport. Preterm birth increases with HIV, chronic maternal disease or malnutrition. Transfer of maternal antibody which is gestational age dependant, may be more affected by maternal immunization in sub-Saharan Africa where these conditions are common.

In 2011, two Maternal influenza vaccination trials are being conducted in pregnant women from the Soweto community, one in HIV-infected women, and the other in HIV-uninfected women. In both these trials (HREC reference number 101106 and 101107) the immunogenicity of a single dose of TIV administered between 20 and 36 weeks of gestation was investigated.

HIV-uninfected women showed a good response one month post vaccination with 85.3%, 92.7% and 98.5% of women having seroprotective levels of antibodies to the vaccine strains, H3N2, H1N1 and B respectively, however, HIV-infected women showed an inferior response to vaccination. The CD4+ cell count at baseline (prior to vaccination) affected the immunological response mounted by HIV-infected women: 12.5%, 12.5% and 62.5% of women with CD4+ cell count <200, and 50%, 66.7% and 77.8% of women with CD4+ count ≥500 had seroprotective levels to vaccine strains, H3N2, H1N1 and B respectively, one month post receipt of single-dose TIV.

The overall aim of this project is to evaluate the safety and immunogenicity of 3 different dosing options of TIV vaccination of HIV-infected pregnant women: single dose, double dose (at same time point) and two-doses (1 month apart)