CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

Results: 54 HIV-infected infants (61% girls) enrolled from 11 countries; 81.5% breast-fed. Median study enrollment age in Cohort 1 was 22 hrs and 8 days in Cohort 2. For Cohorts 1 and 2 median age at ART initiation was 7.3 (1.8, 21.0) and 33 (0.4, 40.1) hrs, and median earliest VL was 4.9 (4.0, 5.3) and 4.1 (3.2, 5.2) log10 cps/mL, measured at a median of 1 (0,1.0) and 6.5 (2.0, 8.0) days of age; loss to follow-up was 3% and 15%. Estimated probability of sustained viral suppression through 52 wks on Step 2 was 50% (32%, 66%) and 67% (37%, 85%) in Cohorts 1 and 2, respectively; 47/52 who started LPV/r met virologic criteria to stop NVP at median age 29.4 (25.0, 37.7) wks. Grade 3 or 4 related events that were reversible occurred through 52 wks in 15 (44%) and 7 (35%) infants from Cohorts 1 and 2, and were mostly hematologic. There was one death in each cohort, neither related to Study ART. Among infants followed through wk 84, 5/8 and 4/5 in Cohorts 1 and 2 are HIV-seronegative. Conclusion: VE-ART for infants with in-utero HIV-infection results in moderate rates of strict virologic control through 52 wks. More effective VE-ART regimens are needed to achieve high rates of sustained virologic suppression in infants with in-utero HIV infection

33% infants showed pre ART control. Cell-associated viral DNA (vDNA) in PBMCs were comparable at 6 wpi in infant (Mean: 1.5e7 copies/million CD4+ T cells) and adult (Mean: 2.1e7 copies/million CD4+ T cells) RMs. Upon initiation of ART, plasma vRNA suppressed below levels of detection within 2-4 wk. Interestingly, ART-suppressed RMs showed similar frequencies of cell-associated vDNA in naïve, Tfh and memory CD4+ T cell populations in LNs. (Range: undetectable- 2.6e4 copies/million CD4+ T cells). Upon ART interruption, 5/6 infants and 3/6 adult RMs rebounded to >150 vRNA copies/ml of plasma. Plasma VL at ART initiation was a predictor of time to viral rebound, and infants had more variability in time to rebound. While all the adults controlled systemic virus within 3-4 wk of rebound, only 3/5 infants demonstrated post-rebound viral control. Oral LNs were a primary site of vRNA and infectious virus in both adults and infant RMs. Conclusion: Using a RMmodel of postnatal infection, we have characterized SHIV.C.CH505 reservoirs and rebound kinetics, which can inform correlates of viral rebound, and design immune-based interventions to reduce pediatric HIV reservoirs. 801 HIV-SPECIFIC ANTIBODY LEVELS CORRELATE WITH TFH MATURATION IN EARLY TREATED INFANTS Julie Mitchell 1 , Thanyawee Puthanakit 2 , Kenneth Dietze 1 , Marta Massanella 3 , Pope Kosalaraksa 4 , Thitiporn Borkird 5 , Suparat Kanjanavanit 6 , Piyarat Suntarattiwong 7 , Thidarat Jupimai 8 , Panadda Sawangsinth 9 , Mark de Souza 9 , Nicolas Chomont 3 , Jintanat Ananworanich 1 , Lydie Trautmann 1 , for the HIVNAT209 Study group 1 US Military HIV Research Program, Silver Spring, MD, USA, 2 HIV–NAT, Thai Red Cross AIDS Research Centre, Bangkok, Thailand, 3 Université de Montréal, Montreal, QC, Canada, 4 Khon Kaen University, Khon Kaen, Thailand, 5 Hat Yai Hospital, Songkhla, Thailand, 6 Nakornping Hospital, Chiang Mai, Thailand, 7 Queen Sirikit National Institute of Child Health, Bangkok, Thailand, 8 Chulalongkorn University, Bangkok, Thailand, 9 Thai Red Cross AIDS Research Center, Bangkok, Thailand Background: Though they have a developing immune system, infants develop potent HIV-specific antibodies. To understand the B cell response that develops after early ART, we measured the circulating T follicular helper cell (cTfh) frequencies and HIV-specific antibody production in early treated infants. Methods: Eighty-two Thai infants living with HIV were included. Samples were taken from viremic infants at baseline (median 2.3 mo old, range 0.8-6.6 mo; n=59), and virally-suppressed infants after a median 12.6 months of ART (range 10.8-16.6 mo; n=50) or a median 24.5 months of ART (range 20.7-26.4 mo; n=25). CD19+CD20+ B cells and CD4+ T cells were analyzed by flow cytometry. Plasma Env-specific IgG and IgM levels were measured by ELISA. Plasma levels of CXCL12, CXCL13, and soluble CD40 ligand (sCD40L) were measured by Luminex. Results: At baseline, very low Env-specific IgM levels were detected in the plasma of infants with HIV (median 1.4µg/mL). Env-specific IgM levels correlated with viral load (r=0.75, p<0.001), as well as plasma levels of the stimulatory molecule sCD40L (r=0.43, p=0.03) and CXCL13, a biomarker of germinal center activity (r=0.63, p<0.001). Though infant Env-specific IgG levels could not be measured due to the presence of maternal antibodies, the frequency of cTfh (CXCR5+CD45RA–CD4+ T cells) correlated with the frequency of IgG+ B cells (r=0.55, p=0.04) at baseline. Baseline frequencies of IgG+ B cells and cTfh in the blood negatively correlated with plasma levels of CXCL12 (r=-0.62 p=0.03; r=-0.66 p=0.01), suggesting these cells are exiting the blood through high endothelial venules. The levels of Env-specific IgM increased after 1 year of ART (median 2.4µg/mL, p<0.001). Env-specific IgG levels decreased as maternal antibodies waned, and remained stable between 1 and 2 years of ART (median 2.4 vs 1.5µg/mL). After 1 year of ART, Env-specific IgG levels correlated with the frequency of Th1-biased cTfh expressing CXCR3 (Fig 1A; r=0.48, p=0.03). In contrast, after 2 years of ART Env-specific IgG levels correlated with the frequency of CXCR3–PD-1+ cTfh, which provide better B cell help (Fig 1B; r=0.64, p=0.03). Conclusion: During early infection, viremic infants produce low levels of Env-specific IgM. Though early treated infants had low levels of Env-specific IgG after ART, antibody levels correlated with the maturing cTfh populations. These data suggest that effective elicitation of HIV-specific antibodies in infants will depend on therapeutic targeting of proper cTfh populations.

Poster Abstracts

800 SHIV CH505 T/F RESERVOIR AND REBOUND IN INFANT AND ADULT RHESUS MACAQUES Ria Goswami 1 , Ashley N. Nelson 1 , Joshua J. Tu 1 , Maria Dennis 1 , Riley J. Mangan 1 , Cameron Mattingly 2 , Amit Kumar 1 , Maud Mavigner 2 , Ann Chahroudi 2 , George Shaw 3 , Katharine J. Bar 3 , Kristina De Paris 4 , Koen Van Rompay 5 , Sallie Permar 1 1 Duke Human Vaccine Institute, Durham, NC, USA, 2 Emory University, Atlanta, GA, USA, 3 University of Pennsylvania, Philadelphia, PA, USA, 4 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 5 University of California Davis, Davis, CA, USA Background: Each year >150,000 infants get infected with HIV, ~50% infections occurring during breastfeeding. While lifelong ART results in effective viral suppression, these infants are predisposed to long-termmetabolic consequences and development of drug-resistant viral strains. Therefore, a functional cure is required to attain an ART-free life of sustained viral remission. The primary barrier for a cure is the ability of HIV to establish persistent viral reservoirs, immediately upon infection. Therefore, strategies to reduce viral reservoirs are urgently needed to delay viral rebound, and attain prolonged viral remission. For that, it is critical to monitor establishment of HIV reservoirs and kinetics of viral rebound. Methods: In this pilot study, 6 SHIV.C.CH505 T/F infected infants and 6 adult rhesus macaques (RMs) were used to characterize viral replication and establishment of reservoirs upon ART initiation at 12 wpi. After 8 wk of ART, the kinetics and anatomic distribution of viral rebound upon ART interruption was measured using ddPCR and co-culturing mononuclear cells with Tzm-bl reporter cells. Results: Plasma viral RNA (vRNA) in infants and adults peaked at 2 wpi (infant mean: 6.9e6 vRNA copies/ml; adult mean:4.2e6 vRNA copies/ml). While both groups demonstrated similar viral load kinetics until ART initiation at 12 wpi,

CROI 2019 311