CROI 2020 Abstract eBook

Abstract eBook

Oral Abstracts

Results: Between Mar 2016 and Nov 2018, we randomized 4,000 HIV-exposed infants at a median age of 6 (IQR 6-7) weeks to POC EID (1,989) or Offsite EID (2,011). Most mothers (94%) reported ART for PMTCT. Eighty-one (2.0%, 95%CI 1.6-2.5%) infants were diagnosed with HIV. Every infant in the POC arm received a same-day result, while the median time to diagnosis in the Offsite armwas 27 (IQR 22-30) days. The majority of infants randomized to Offsite EID relied upon the trial’s diagnostic safety net (Fig. 1a). ART initiation was high in both arms (Fig. 1b), but adverse outcomes were common. Among 81 HIV-infected infants, there were 15 (19%) deaths, 15 (19%) follow-up losses, and 30 (38%) virologic failures (1 viral load result pending). By 12 months, only 20 (25%, 95%CI 16–34%) HIV-infected infants were alive, in care, and virally suppressed and this did not differ by study arm: 13 (30%, 95% CI 16-43%) infants in the POC arm vs. 7 (19%, 95% CI 7-32%) in the Offsite arm (RR 1.5, 95% CI 0.7-3.4). Conclusion: Despite the success of Zambia’s PMTCT program, adverse outcomes are high among HIV-infected infants. POC EID eliminated diagnostic delays and resulted in rapid ART initiation but this did not translate to treatment success at 12 months. As countries consider whether to adopt POC EID, they will need to weigh the costs of new technology against the costs of improving existing EID services. Most importantly, substantial investments are needed to strengthen pediatric HIV treatment programs.

and were randomized to: 1) intervention: immediate ART, annual population level testing, and patient-centered HIV care (including welcoming staff, flexible hours, and facilitation between antenatal care and HIV clinic); or, 2) control: HIV-care per national guidelines. Pregnant women were offered immediate ART in both arms. After 3 years, we repeated population-level testing including children <3 years and ascertained births and deaths. In pre-specified analyses, we compared HIV-free survival (% of infants alive and HIV uninfected) and vertical transmission (% of living infants with HIV infection) between study arms among infants born to a) all women with known HIV+ status by year 3; and, b) the subset of women with known HIV+ status at baseline using cluster- level targeted maximum likelihood estimation. Results: There were 1,417 births to 1,332 women with known HIV+ status by year 3; outcomes were ascertained in 76% of infants in intervention and 78% in control. The proportion (95%CI) with HIV-free survival was higher and vertical transmission was lower in the intervention versus control: 3.3% (1.0- 5.6%) in the intervention died or became HIV-infected by year 3 versus 6.4% (4.7-8.0%) in the control (Relative risk:1.03; 95%CI:1.00, 1.06; p=0.04). Vertical transmission was 1.8% (0.2-3.3%) in the intervention versus 4.4% (2.7-6.1%) in the control (p=0.04). Of 1,230 births to 1,158 women with known HIV+ status at baseline, vertical transmission was 0.5%(0-1.3%) in the intervention, compared to 3.7% (2.4-5.1%; p<0.001) in the control. Conclusion: Universal testing and a patient-centered care delivered via government clinics reduced 3 year population-level HIV infection/mortality among infants by over 50% and reduced vertical transmission to 0.5% among women with known HIV, progress toward the elimination of vertical transmission. PREDICTORS OF THE PERSISTING VIRAL RESERVOIR IN VERY EARLY TREATED INFANTS Maria Paximadis 1 , Bianca Da Costa Dias 1 , Sizanani Mncube 1 , Renate Strehlau 2 , Yanhan Shen 3 , Stephanie Shiau 3 , Faeezah Patel 4 , Megan Burke 2 , Karl Technau 2 , Shayne Loubser 1 , Elaine J. Abrams 3 , Caroline Tiemessen 1 , for the LEOPARD Study Team 1 National Institute for Communicable Diseases, Johannesburg, South Africa, 2 Empilweni Service and Research Unit, Johannesburg, South Africa, 3 Columbia University Medical Center, New York, NY, USA, 4 University of the Witwatersrand, Johannesburg, South Africa Background: The size of the persisting viral reservoir while receiving antiretroviral therapy (ART) has consistently been shown to be smaller when ART is initiated at a younger age in perinatally-acquired HIV infection. However, there are only limited data on predictors of the proviral DNA reservoir in very early treated infants. Methods: Sixty-three confirmed HIV-infected neonates recruited at Rahima Moosa Mother and Child Hospital in Johannesburg, South Africa, who had been identified <48 hours after birth were included. Viably- preserved PBMCs collected pre-treatment, and at 1, 3, 6 and 12 months after ART initiation, were tested if sufficient sample was available. To quantify the proviral DNA reservoir, a semi-nested real-time quantitative hydrolysis probe (TaqMan) PCR assay was designed to detect and quantify total HIV-1 subtype C proviral DNA. The assay was designed to target the integrase gene of HIV-1 subtype C. We conducted six replicates to allow detection to a level of one copy/9.1x105 cells. Multivariable Generalized Estimating Equation (GEE) regression models were used for statistical analysis. Results: Thirty-one (49.0%) infants initiated ART <48 hours of birth and the remaining 32 infants at median of 7 days (all received daily nevirapine prophylaxis prior to ART start). Three-quarters were infected despite their mothers having received ART during pregnancy and, for 25%, mothers had received no ART prior to delivery. At all post-ART time points, infant HIV-1 DNA was significantly associated with concurrent HIV-1 RNA levels (viral load [VL]) (Spearman correlation=0.645, p<0.0001). If VL was target not detected, the median HIV-1 DNA was 1.56 log copies and 23.1% had <10 DNA copies detected. Whereas, at VL <50, 51-399, 400-999 and >1000 RNA copies/ml, median DNA log copies (and %with <10 copies) were 1.83 (17.7%), 2.38 (10.8%), 2.83 (10.0) and 3.15 (0%), respectively. In multivariable analysis, starting ART <48 hours after birth (p=0.03), having been born to a mother who did not receive ART during pregnancy (p<0.0001), and pre-treatment infant CD4+ T-cell percentage >30 (p<0.0001) predicted lower HIV-1 DNA log copies in the first year post-treatment (Table).

Oral Abstracts

135

134LB POPULATION-LEVEL HIV-FREE INFANT SURVIVAL IN THE SEARCH TRIAL Shalika Gupta 1 , Jane Kabami 2 , Gabriel Chamie 3 , Norton Sang 4 , Dalsone Kwarisiima 2 , Douglas Black 3 , Laura B. Balzer 5 , James Ayieko 4 , Craig R. Cohen 3 , Elizabeth A. Bukusi 4 , Moses R. Kamya 6 , Diane V. Havlir 3 , Maya L. Petersen 3 , Theodore Ruel 3 1 University of California Berkeley, Berkeley, CA, USA, 2 Infectious Diseases Research Collaboration, Kampala, Uganda, 3 University of California San Francisco, San Francisco, CA, USA, 4 Kenya Medical Research Institute, Nairobi, Kenya, 5 University of Massachusetts Amherst, Amherst, MA, USA, 6 Makerere University College of Health Sciences, Kampala, Uganda Background: Universal test and treat (UTT) strategies could reduce vertical transmission of HIV by diagnosing women living with HIV earlier and improving care delivery. We evaluated the effect of universal HIV testing and a patient- centered HIV care model on vertical transmission and HIV-free survival in the SEARCH Trial. Methods: At baseline, 32 communities in rural Uganda and Kenya (total population ~350,000) received population level HIV testing (90% coverage)

45

CROI 2020