CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Background: Women living with HIV give birth to ~1.5M infants each year, 80% of them exposed to ARVs in utero. Most ARVs can cross the placenta, but their safety has not been fully elucidated in the context of pregnancy. The Tsepamo study reported a signal for risk of neural tube defects in infants exposed to the InSTI dolutegravir (DTG) from conception. Many ARVs affect mitochondria, which could impact embryonic development. Our objective was to characterize and compare the effects of 14 ARV regimens on cultured human embryonic stem cells (hESCs), with respect to pluripotency, and cellular and mitochondrial health. Methods: CA1S hESCs were cultured (n=5 independent experiments) in the presence of 0.1% DMSO or 1X C max of the following regimens: DTG, raltegravir (RAL), bictegravir (BIC), cobicistat-boosted elvitegravir (EVG/COBI), or efavirenz (EFV) on a TDF/FTC backbone; DTG, RAL, BIC, EVG/COBI, or rilpivirine (RPV) on a TAF/FTC backbone; DTG, RAL, or ritonavir-boosted darunavir (DRVr) on an ABC/3TC backbone; cabotegravir (CAB)/RPV. After 3 days, cells were assessed via flow cytometry using markers for mitochondrial mass, intermembrane potential, reactive oxygen species (ROS), cell viability, and apoptosis. Two markers of pluripotency, specifically SSEA-3 (lost early in differentiation) and TRA-1-60 (a later marker), were also assessed. Regimens were grouped according to base ARV and compared to DMSO control using Kruskal-Wallis with Dunn’s correction. Results: HESCs exposed to DTG or BIC had a 3-fold reduction in cell counts (p≤0.005) compared to controls. BIC exposure resulted in a 5-fold decrease in viability (p=0.026) and a 6-fold increase in apoptosis (p=0.01). In regards to pluripotency, exposure to regimens containing DTG or CAB resulted in a >80% loss of SSEA-3 expression compared to controls (p≤0.02). There were no significant differences between regimens with respect to mitochondrial mass, intermembrane potential, ROS, or loss of TRA-1-60 expression. No effects were detected for the backbones, RAL, EVG/COBI, EFV, RPV, or DRVr. Conclusion: These data indicate that exposure to some ARV regimens at pharmacological concentrations, especially DTG or BIC, appear toxic to cultured hESCs. Our results further suggest that exposure to the InSTIs DTG and CAB can induce hESC differentiation. Given the increasing use of DTG and other InSTIs, it is imperative to investigate their long-term safety in the context of pregnancy and embryonic development.

respectively. After adding maternal self-reported adherence to risk definition, 67% of perinatally infected infants became high risk, increasing a 24% of high- risk patients the WHO classification, while low risk reduced to 33% (Figure 1) Conclusion: In our cohort, 24% HIV infants were miscategorized as low risk using the current WHO algorithm. Including self-reported adherence information can help to provide ePCP to all eligible infants

788 IMPROVED HEMATOLOGICAL OUTCOMES WITH NEVIRAPINE FOR INFANT HIV POSTNATAL PROPHYLAXIS Catherine Dollfus 1 , Jérôme Le Chenadec 2 , Laurent Mandelbrot 1 , Roland Tubiana 1 , Albert Faye 3 , Stephane Blanche 1 , Pierre Frange 1 , Josiane Warszawski 2 , for the ANRS CO1 et CO11 1 AP–HP, Paris, France, 2 INSERM, Le Kremlin-Bicetre, France, 3 Assistance Publique – Hôpitaux de Paris, Paris, France Background: With combination antiretroviral therapy (cART) in HIV-infected women, mother-to-child transmission rates declined to less than 1%. For post- natal infant prophylaxis, in situations of low risk of perinatal HIV transmission, high income countries use zidovudine (ZDV), whereas low income countries use either nevirapine (NVP) or ZDV. Given the low transmission risk and the concerns about the toxicities of ZDV in newborns, French national guidelines recommend since 2015 the use of NVP as an alternative to ZDV for post-natal prophylaxis in full term babies born to HIV1-infected mothers with suppressed viral load and no history of NVP resistance. We compared hematological outcomes between ZDV-exposed and NVP-exposed infants. Methods: In the French prospective national Perinatal Cohort Study, we compared hematological outcomes (blood cell counts and differentials) at birth, 1 and 3 months among the infants born in 2016-2017 , at >=37weeks gestation. We included only mothers treated with cART without ZDV, to exclude a potential impact of maternal treatment on infant outcomes. ZDV was prescribed for 4 weeks, NVP for 2 weeks; mothers did not breastfeed. Results: 137 infants were exposed to NVP and 251 to ZDV. None became infected.68% of mothers were born in sub-Saharan Africa (79.4% in NVP group, 62.9% in ZDV group) .Median hemoglobin levels were respectively 17.4 g/ dL vs 17.0 at birth (p=0.49),11.7 vs 11 g/dL at 1month (p=0.003) and 11.4 vs 11.2 g/dL at 3 months(p=0.02).Anemia grade >=2 was observed in 0.8% vs 1.7 % of infants at birth (p=0.66), 1.2% vs 9.4% at 1 month (p=0.014), 3.6% vs 7.3% at 3 months(p=0.40). Median neutrophil counts were similar, grade >=2 neutropenia was found in 4.2% vs 2.7% infants at birth(p=0.53),15.9% vs 13.1% at 1 month(p=0.56), and 12.2 vs 13.4% at 3 months(p=0.84). No difference was found in platelets counts. Conclusion: In this population of HIV-exposed uninfected infants, post-natal prophylaxis with NVP, compared to ZDV was associated with higher hemoglobin levels at 1 and 3 months and a 9-fold lower incidence of anemia at 1 month of age. These findings support the use of nevirapine as a first choice for single drug post-natal prophylaxis in low risk situations. 789 TOXICITY OF INTEGRASE INHIBITORS IN A HUMAN EMBRYONIC STEM- CELL MODEL Marie-Soleil R. Smith 1 , Hélène C. Côté 1 1 University of British Columbia, Vancouver, BC, Canada

Poster Abstracts

790 ESTIMATES OF PERICONCEPTIONAL EXPOSURES TO INTEGRASE INHIBITORS UNITED STATES Margaret A. Lampe 1 , Jeff Wiener 1 , Jennita Reefhuis 1 , Steven R. Nesheim 1 1 CDC, Atlanta, GA, USA Background: In 2018, an increased risk of neural tube defects among infants with periconceptional dolutegravir (DTG) exposure was reported from Botswana, triggering changes to global treatment guidelines for women and limiting access to DTG. Estimates of periconceptional integrase inhibitor (INSTI) exposures are needed to understand the potential to study the impact of INSTI use on pregnancy and birth outcomes in the United States. Methods: We estimated U.S. periconceptional INSTI exposures as follows. We used hospital discharge data from the Healthcare Cost and Utilization Project from 2007-2014 to predict the number of deliveries in 2015-2017 to women with diagnosed HIV using Poisson regression. We used National Vital Statistics Report estimates of proportions of all pregnancies resulting in live births (65%) and the

CROI 2020 292