CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

life. Factors showing an association univariately (p<0.1) were considered in multivariable linear regression models. Results: mtDNA content at birth (0-3d) was obtained for 114 HEU and 86 HUU children. In a multivariable model of mtDNA content at birth (n=200) that included HEU/HUU status, gestational age (GA) at birth, and maternal smoking during pregnancy, lower GA was the only factor independently associated with higher mtDNA content (p<0.001). Among infants born at term (GA>37w, n=168), although HEUs had significantly lower GA at birth (median: 38.9 vs. 39.7, p<0.01), the multivariable analysis revealed that only HEU status (p=0.005) was independently associated with higher birth mtDNA content. In a separate model that investigated maternal cART parameters among HEUs (n=114), neither duration nor type of in utero cART exposure were associated with mtDNA content at birth. Additionally, infant AZT prophylaxis did not affect mtDNA content at 6w vs. birth. Lastly, among age and sex-matched children (n=214:214), HEU children continued to have higher mtDNA content than HUUs (p<0.01) throughout the first three years of life. Conclusion: HEUs and infants born preterm have higher mtDNA content at birth, an effect that persisted up to age three. This may represent a long-term effect, possibly resulting from adaptive mitochondrial biogenesis in response to in utero stresses. 880 INCREASED INFLAMMATION AND MONOCYTE ACTIVATION IN HIV- EXPOSED UNINFECTED INFANTS Marisa M. Mussi-Pinhata 1 , Adriana Weinberg 2 , Qilu Yu 3 , Rachel A. Cohen 3 , Sahera Dirajlal-Fargo 4 , Nicholas Funderburg 5 , Emily Bowman 5 , Nahida Chakhtoura 6 , Grace A. McComsey 4 1 University de São Paulo Ribeirão Preto, Ribeirão Preto, Brazil, 2 University of Colorado Denver, Denver, CO, USA, 3 Westat, Inc, Rockville, MD, USA, 4 University Hospitals of Cleveland Medical Center, Cleveland, OH, USA, 5 The Ohio State University, Columbus, Ohio, USA, 6 National Institute of Child Health and Human Development, Bethesda, MD, USA Background: HIV infection is accompanied by high levels of inflammation that predict increased mortality in adults. HIV-exposed uninfected (HEU) infants also have increased infectious morbidity and mortality, but little is known about their levels of inflammation and immune activation. In this study, we assessed how inflammatory and monocyte activation markers correlated between mothers and HEUs at delivery and compared markers between HEU and HIV- unexposed (HU) infants at birth and at 6 months of life. Methods: The study enrolled term singletons ≥2500g at birth and their mothers. Samples obtained at birth and 6 months from 86 HEU mother-infant pairs enrolled in the NICHD cohorts in Brazil were analyzed and compared to 88 HU mother-infant pairs. All HIV-infected mothers received ARV during pregnancy. HEUs received neonatal zidovudine prophylaxis and formula. HUs were born to healthy mothers, and most received formula feeding by 4 weeks of age. Infants had clinical and laboratory evaluations at birth and 6 months. IL-6, TNFRI, TNFRII, sCD14, sCD163, IP-10, VCAM, OxLDL, D Dimer, and hsCRP were assayed by ELISA. Data were analyzed using two-sample t-tests, correlation coefficients and linear regression models. p<0.005 was used to adjust for multiple comparisons. Results: Among HIV-infected mothers, 81.4% had HIV-RNA <1,000 copies/ mL prior to delivery. In addition, they were older (27 vs. 24 years), and more frequently non-white (64.4% vs. 25.0%) when compared to uninfected mothers. Compared to HU, HEU infants were born more frequently by C-section (59% vs. 32%), with a lower median gestational age (38.6 vs. 39.3wk) and weight (3.2 vs. 3.3kg); and reached lower weight (5.9 vs. 8.5kg) and height (53.5 vs. 68.8 cm) at 6 months of age. Majority of inflammatory markers were significantly higher (p≤0.005) in HEU compared with HU at birth, but at 6 months only TNFRI and IL-6 remained higher (Table). Among HU mother-infant pairs, infant IL-6, TNFRI, TNFRII, sCD14, sCD163 levels at birth were associated with maternal levels at delivery (r≥0.31; p≤0.0004). For HEU pairs, the only association was for IP-10 (r=0.34; p<0.0001) at birth. Conclusion: HEU infants had higher inflammation and monocyte activation than HU at birth, which for some markers persisted to 6 months of life, and was related to maternal inflammatory status. Inflammation may contribute to the increased HEU infectious morbidity and poor growth. This hypothesis warrants further investigation.

881 INNATE IMMUNE ACTIVATION AMONG HIV-1 EXPOSED UNINFECTED INFANTS FROM BOTSWANA Pilar Garcia Broncano 1 , Samuel W. Kgole 2 , Gosego Masasa 2 , Terence Mohammed 2 , Sikhulile Moyo 2 , Joseph Makhema 2 , Xu G. Yu 1 , Jennifer Jao 3 , Roger L. Shapiro 4 , Mathias Lichterfeld 5 , Kathleen M. Powis 6 1 Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA, 2 Botswana Harvard AIDS Institute Partnership, Gabarone, Botswana, 3 Icahn School of Medicine at Mt Sinai, New York, NY, USA, 4 Harvard University, Boston, MA, USA, 5 Brigham and Women’s Hospital, Boston, MA, USA, 6 Massachusetts General Hospital, Boston, MA, USA Background: HIV-exposed uninfected (HEU) infants continue to experience increased morbidity/mortality from infectious causes compared with infants born to HIV-uninfected women. To explore possible immune etiologies, we conducted an immune profiles analysis, comparing profiles from a Botswana- based cohort of HEU to HIV-unexposed (HU) infants. Methods: Multiparametric flow cytometry was used to quantify proportions and phenotypic characteristics of innate immune cells (monocytes, dendritic cells and NK cells) and adaptive T and B cell-mediated immune responses using peripheral blood mononuclear cells (PBMCs) collected at 3-months of life from HEU and HU infants enrolled in a longitudinal gut microbiome study in Botswana. All HIV-infected women received ≥6 weeks of ART prior to delivery. All infants were born full-term and all HEU infants tested HIV negative at 3 months of life. Results: Thirty-three infants (17 HEU and 16 HU) were studied cross- sectionally. HEU infants had lower proportions CD14+monocytes compared with HU infants (p=0.013). The proportion of CD14+ CD16- “classical” monocytes was significantly reduced in HEU infants, while proportions of CD14+ CD16+ non-classical/intermediate monocytes, associated with increased activation and inflammatory responses, were markedly increased (Figure). Proportions of NK cells, the primary innate immune system effector cell, were lower in HEU infants (p=0.026). Upregulation of NKp30, a surface marker denoting terminal activation, on CD56hiCD16low NK cell subset was noted among PBMCs of HEU infants. Frequencies of CD4+ and CD8+ T cell subsets, proportions of regulatory T cells and expression of immune activation markers on these T cell populations did not differ between groups. Proportion of late memory B cells (Bm5, IgD- CD38-) was lower in HEU infants (p=0.026). While fewer HEU infants exclusively breastfed through 3 months (71% vs 88%), the difference was not significant (p=0.40). Despite these findings, there was no significant difference in hospitalizations in the first year of life between the groups. Conclusion: In this cohort, in utero exposure to HIV-1 and ART was associated with a distinct immunological profile characterized by increased immune activation in the innate immune system. Longitudinal evaluations are needed to determine whether abnormal innate immune activation among HEU has longer- term clinical consequences.

Poster Abstracts

CROI 2018 334