CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Conclusion: In utero exposure to HIV and ART without infection was associated with reduced basal ganglia and total grey matter volume in early infancy. To our knowledge this is the first cohort study to examine the neuroanatomy of HEU neonates. These findings are consistent with brain regions reported to be affected in HIV-infected children and suggest that HIV/ART exposure may impact brain structural development during pregnancy. 799 POSTNATAL LPV/R EXPOSURE, GROWTH, AND NEUROPSYCHOLOGICAL OUTCOMES AT SCHOOL AGE Nicolas Nagot 1 , Mandisa Singata 2 , Amandine Cournil 1 , Joyce Nalugya 3 , Souleymane Tassembedo 4 , James Tumwine 5 , Nicolas Meda 4 , Chipepo Kankasa 6 , 1 INSERM, Montpellier, France, 2 University of Fort Hare, East London, South Africa, 3 Makerere University College of Health Sciences, Kampala, Uganda, 4 Centre Muraz, Bobo-Dioulasso, Burkina Faso, 5 Makerere University, Kampala, Uganda, 6 University Teaching Hospital, Lusaka, Zambia, 7 University of Fort Hare, Alice, South Africa, 8 University of Bergen, Bergen, Norway Background: In the ANRS 12174 randomized controlled trial, HIV-exposed uninfected African neonates who received lopinavir-ritonavir (LPV/r) prophylaxis for one year exhibited slower growth from birth to week 50 compared with those receiving lamivudine (3TC). We assessed whether this difference in growth persisted over time, and was accompanied by differences in neuropsychological and clinical outcomes. Methods: Between February 2017 and February 2018, we conducted a cross- sectional clinical evaluation among former trial participants who completed the 50-week follow-up and who were not HIV-infected. In addition to HIV testing and clinical examination, neuropsychological outcomes were assessed using the Kaufman Assessment Battery for Children, 2nd edition (KABC-II), Tests of Variables of Attention (TOVA), the Movement Assessment Battery for Children, second edition (MABC-2), and the caregiver-reported Strengths and Difficulties questionnaire (SDQ). Results: Of 1101 eligible children, aged 5 to 7 years, 553 could be traced and analysed (274 in the LPV/r and 279 in the 3TC groups). Changes from baseline value in height-for-age, body mass index and weight-for-age Z-scores, were greater in the LPV/r group compared to the 3TC group (estimated differences ranging from 0.19 to 0.30), but Z-scores did not differ between groups at follow- up. No differences in the KABC-II and MABC-2 tests and SDQ questionnaire were found. A marginally better performance was observed for the 3TC group on the TOVA test. Clinical outcomes were similar between groups. Conclusion: The impact of LPV/r on growth did not persist over time after drug discontinuation. At school age, children exposed to LPV/r and 3TC at birth for one year had comparable growth and neuropsychological outcomes without evidence of long-term side-effects of LPV/r. It provides reassuring data on clinical outcomes for all HIV-infected children treated with this antiretroviral in early life. 800 MALNUTRITION IN HIV-EXPOSED UNINFECTED CHILDREN IN LONG-TERM OBSERVATIONAL FOLLOW-UP Lynda Stranix-Chibanda 1 , Jim Aizire 2 , Nonhlanhla Yende-Zuma 3 , Haseena Cassim 4 , Sherika Hanley 3 , Teacler Nematadzira 1 , Lucky Makonokaya 5 , Bonus Makanani 6 , Lillian Wambuzi Ogwang 7 , Taha E. Taha 2 , Mary Glenn Fowler 8 , for the PROMOTE study team 1 University of Zimbabwe, Harare, Zimbabwe, 2 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 3 CAPRISA, Durban, South Africa, 4 Perinatal HIV Research Unit, Soweto, South Africa, 5 University of North Carolina Project–Malawi, Lilongwe, Malawi, 6 University of Malawi, Blantyre, Malawi, 7 Makerere University– Johns Hopkins University Research Collaboration, Kampala, Uganda, 8 Johns Hopkins University School of Medicine, Baltimore, MD, USA Background: For unclear reasons, HIV exposed uninfected children (HEUs) are at risk of malnutrition, which increases childhood infections and mortality. Stunting, particularly in the first 2 years of life, potentially affects cognitive functioning and educational achievement, adult height and future risk of metabolic disease. Stunting in girls may be passed on to their future offspring. We set out to establish the rate of severe growth faltering and correlates of stunting in a cohort of HEUs aged 2-5 years in follow-up since birth in four African countries. Marianne Periès 1 , Joanne Batting 7 , Ingunn M.S. Engebretsen 8 , Thorkild Tylleskär 8 , Philippe Van De Perre 1 , Grace Ndeezi 5 , Jean-Pierre Molès 1

altered by immunosuppression and influence morbidity in HIV-exposed infants are unknown. We hypothesized that immunosuppression, as measured by low maternal CD4 count, alters BM virome and microbiome. Methods: We performed next-generation sequencing (NGS) to comprehensively define the virome and microbiome in BM samples collected during the pre-ART era in Kenya (2003-2005) from 53 HIV-infected women at 1 month postpartum: 30 women with CD4 <250 and 23 women with CD4 >500. Illumina sequencing was performed on Phi29-amplified nucleic acid (virome) and the 16S rRNA gene V4 region (bacterial microbiome). Quantitative real-time PCR (qPCR) was used to quantify select viral species. Results: Among 53 HIV+ women, BM bacterial microbiomes were highly diverse and shared a core bacterial microbiome composed of Streptococcaceae (18.1%), Staphylococcaceae (10.1%), Moraxellaceae (4.1%) and Eubacteriaceae (3.6%) families. There was no significant difference in the diversity of BM bacterial microbiome between women with CD4>500 compared to CD4<250 in terms of ecological measurements of richness (p>0.65), alpha-diversity (p>0.14) and beta-diversity (p>0.17). The BM virome was dominated by cytomegalovirus (CMV). The average proportion of CMV virome sequences did not differ between women with CD4 >500 and <250, with an average of 55.6% vs 69.4%, respectively (p>0.21). These NGS results were corroborated by qPCR measurements of CMV viral loads in BM (p>0.09). All women had a high abundance of bacteriophage families: Myovirdae (20.7%), Siphoviridae (11.6%) and Podoviridae (3.4%). Other eukaryotic viruses detected include papillomaviruses and anelloviruses. There was no significant difference in the BM virome richness (p>0.68), alpha- diversity (p>0.15) or beta-diversity (p>0.30) between women with CD4>500 compared to CD4<250. Conclusion: In this cohort of HIV+ Kenyan women from the pre-ART era, BM harbors a core bacterial microbiome and a diverse virome that is dominated by CMV. Diversity and richness of the BMmicrobiome and virome were not significantly influenced by immunosuppression at 1 month postpartum. 798 REDUCED BASAL GANGLIA AND TOTAL GREY MATTER VOLUME IN HIV- EXPOSED UNINFECTED NEONATES Catherine J. Wedderburn 1 , Nynke A. Groenewold 2 , Annerine Roos 3 , Shunmay Yeung 1 , Jean-Paul Fouche 2 , Andrea M. Rehman 1 , Diana Gibb 4 , Katherine L.Narr 5 , Heather Zar 2 , Dan Stein 2 , Kirsty Donald 2 1 London School of Hygiene & Tropical Medicine, London, UK, 2 University of Cape Town, Cape Town, South Africa, 3 Stellenbosch University, Stellenbosch, South Africa, 4 MRC Clinical Trials Unit at UCL, London, UK, 5 University of California Los Angeles, Los Angeles, CA, USA Background: Evidence suggests HIV-exposed uninfected (HEU) children have impaired early growth and development compared to HIV-unexposed (HU) children. However, little is known about the neurobiological mechanisms underlying adverse developmental outcomes in this population. We examined the effects of in utero exposure to HIV and ART on the neuroanatomy of uninfected neonates in a South African birth cohort. Methods: A subgroup of neonates in the Drakenstein Child Health Study (DCHS), born after 36 weeks’ gestation, without medical comorbidities or neonatal intensive care admission, had magnetic resonance imaging (MRI) at the Cape Universities Brain Imaging Centre, South Africa. Mother-child pairs received antenatal and postnatal HIV testing and ART per local guidelines. Acquired structural T2-weighted images were processed using statistical parametric mapping software. Subcortical regions-of-interest were defined using the automated anatomical labeling atlas and volumes were extracted from grey matter segmented images bilaterally. Subcortical and total grey matter volumes were compared between groups using multivariable linear regression adjusting for intracranial volume, infant age and sex. Results: 183 neonates in the DCHS had multimodal MRI between October 2012 and September 2015. Following quality control, 143 structural images were included (HEU n=39; HU n=104) (mean age 3.2 weeks, 51%male). All HEU infants were exposed to ART (87% to maternal triple ART). HEU infants had smaller caudate volumes bilaterally compared to HU (left hemisphere p=0.006, adjusted Cohen’s d effect size -0.50 [95% CI -0.87 to -0.13]; right hemisphere p<0.001, adjusted Cohen’s d -0.68 [-1.06 to -0.31]). There were no group differences in other subcortical volumes (all p>0.2). Total grey matter volume was also reduced in HEU infants (p=0.039, adjusted Cohen’s d -0.33 [-0.70 to 0.04]). The associations remained significant after further adjusting for maternal age and education, household income, and prenatal alcohol exposure.

Poster Abstracts

CROI 2020 296