Morbidity, mortality, and gut virome ecology of Kenyan children exposed to versus not exposed to maternal HIV

Mother-to-child transmission of HIV has decreased rapidly with the expansion of optimized antiretroviral therapy (ART) regimens for all individuals living with HIV, resulting in a growing population of children who are HIV-exposed but uninfected (CHEU). Historically, CHEU have experienced greater morbidity and mortality than children who are HIV-unexposed uninfected (CHUU). The etiology of these differences remains unclear, but likely includes a combination of maternal and child biological, socio-behavioral, and environmental factors. With universal ART, many mothers living with HIV initiate optimized ART before pregnancy, which has been shown to improve maternal immunology and general health, and to lower overall morbidity and mortality among CHEU in the first years of life. However, few studies have directly compared the incidence and mechanisms of morbidity and mortality among CHEU and CHUU in the current era of optimized, universal ART to assess whether historic differences in these outcomes persist. Such data are needed to inform new directions of investigation and intervention that support the health of CHEU worldwide. To address this gap, this dissertation leveraged data from the Linda Kizazi Study—a Nairobi, Kenya-based cohort of healthy mother-infant pairs followed from the third trimester of pregnancy through two years postpartum—to describe and compare outcomes among CHUU and CHEU born to women living with HIV on ≥6 months of optimized ART. In Chapter 2, we used survival analysis methods to compare the risk of acute diarrhea, respiratory tract infections, malaria, hospitalization, and all-cause mortality between CHEU and CHUU. In Chapter 3, we used survival analysis methods to determine the probability, timing, and correlates of primary CMV infection among all infants and by HIV exposure status. In Chapter 4, we described the ecology of the infant gut “virome” (collection of all viruses) and assessed whether the richness (number of unique viruses), relative abundance, and alpha and beta diversity of viruses differed between CHEU and CHUU. In the Linda Kizazi cohort, most mothers living with HIV initiated ART before pregnancy and most CHEU received antiretroviral and cotrimoxazole prophylaxis. The incidence of respiratory tract infections, including pneumonia, in the first two years of life was lower among CHEU than CHUU; this association may have been mediated by exclusive breastfeeding since a significantly greater proportion of CHEU than CHUU were exclusively breastfed for 6 months. However, there were no significant differences between CHEU and CHUU in mortality or other measures of morbidity (acute diarrhea, malaria, hospitalization). There also was no association between HIV exposure and the probability or timing of primary CMV infection, though both CHEU and CHUU had a two times greater risk of CMV acquisition for every log10 increase of CMV DNA in their mother’s breast milk. Compared to CHUU, CHEU had lower gut virome richness and a lower relative abundance of several Anelloviruses (a family of commensal viruses), but the alpha and beta diversity and relative abundance of other viral families were similar between HIV exposure groups. Findings from this dissertation suggest that universal optimized ART can successfully reduce morbidity and mortality among otherwise healthy CHEU, resulting in outcomes similar to their CHUU peers. Our results differ from historic cohorts in the pre-ART era, which showed greater morbidity and mortality and earlier CMV acquisition among CHEU than CHUU. To our knowledge, this is the first study to describe the gut virome of CHEU; our data suggest that the overall gut virome composition in infancy is not substantially altered by HIV exposure despite some differences in the presence versus absence of specific taxa, but further study is needed to understand why there were some differences in overall richness and the relative abundance of Anelloviruses. Overall, these results emphasize the need to continue expanding access to optimized ART for all women living with HIV and provide additional evidence that supports current recommendations to exclusively breastfeed all infants for 6 months.