CROI 2017 Abstract e-Book

Abstract eBook

Oral Abstracts

Results: To date we have identified 100 HIV-infected infants as part of the birth PCR testing program (~1.5% of HIV-exposed infants tested). Of these, 68 HIV-infected infants have been enrolled into a clinical trial tracking their response to ART. Half of these infants started ART within the first 2 days of life (n=34), 25% started 3-7 days (n=17), 15% 8-15 days (n=10) and 10% 16-104 days (n=7). Three infants died in the first months of life, all of whom had started ART within the first 2 days. To examine viral response, we restricted analysis to infants who started ART 0-15 days and were alive and still in follow-up at 6 months. 11% of treated infants have persistently high HIV RNA levels, 35% have a declining trajectory that has not yet reached LDL, and 54% have declined to LDL or target not detected (TND) of the assay. Of those who achieved suppression by 6 months, 30% have changed to having a negative diagnostic HIV PCR during follow-up. Figure 1 shows the viral dynamics in the first year of life of the 34 infants who started ART 0-2 days. Conclusion: There is variability in virologic response to early ART among HIV-infected infants identified at birth. Many clinical and social challenges affect engagement in care of this high risk group of infants. Follow-up is on-going and updated data will be provided.

Oral Abstracts

28 RAPID DECLINE OF TOTAL HIV DNA IN CHILDREN STARTING ART WITHIN 8 DAYS OF BIRTH Kirsten A. Veldsman 1 , Jean Maritz 2 , Shahieda Isaacs 1 , Mary Grace K .Katusiime 1 , Heleen la Grange 1 , Anita Janse van Rensburg 1 , Barbara Laughton 2 , John Mellors 3 , Mark Cotton 2 , Gert U. van Zyl 1 1 Stellenbosch Univ, Parow, South Africa, 2 Stellenbosch Univ, Cape Town, South Africa, 3 Univ of Pittsburgh, Pittsburgh, PA, USA Background: Early infant antiretroviral combination therapy (ART), initiated in the first 2 months of life, reduces HIV-1 infected and transcriptionally active cells (van Zyl, JID 2015). In the case of the Mississippi baby, very early ART, initiated shortly after birth, resulted in delayed viral rebound after therapy interruption, probably due to a very small pool of infected cells. Although there are reports from resource rich settings of undetectable or very low levels of HIV DNA in children who started ART shortly after birth, data from resource limited settings are very limited. Methods: Eleven children diagnosed (at least 2 positive HIV nucleic acid tests) through a public health sector birth diagnosis programwere initiated on ART between 0 and 8 days after birth (median 3 days). Peripheral blood mononuclear cells (PBMCs) and plasma were processed at 3 monthly visits. HIV-1 total DNA was measured with a sensitive quantitative PCR assay (Hong, JCM 2016), adapted for HIV-1 subtype C, targeting a conserved region in HIV-1 integrase (iCAD; limit of detection 3 copies/million cells). Plasma HIV-1 RNA was quantified by Roche COBAS® AmpliPrep/COBAS® TaqMan® HIV-1 v2. Results: The initial ART regimen consisted of AZT/3TC/NVP, with NVP replaced by LPV/r after 2 weeks of age. Median baseline plasma HIV-1 RNA was 4.0 (range 2.4-4.7) log10 copies/ml. One child had ongoing viremia. All other children achieved plasma HIV-1 RNA <100 copies/ml after a median of 4 months, but two had subsequent single viremic episodes. Four children had no detectable HIV-1 DNA in ≥ 500,000 PBMCs assayed when first sampled at 9 days, 3.8 months, 4.9 months and 8.2 months after starting ART. Stored dried blood spots from before ART initiation were found for 3 of these children: 2 had detectable HIV-1 DNA by iCAD and one child with a low baseline plasma HIV-1 RNA load of 265 copies/mL had undetectable iCAD. In the other 6 children, excluding the child with ongoing viremia, there was progressive decline in HIV-1 DNA with 4 of 6 reaching < 10 copies per million cells within 13 months of ART initiation. Conclusion: Early ART initiation within a few days of birth can suppress viral replication, limit the initial number of HIV infected cells and result in their subsequent decay to undetectable levels. However, this rapid decay and limited sample amount require more robust HIV-1 molecular diagnostics to detect HIV persistence on ART and to prevent misdiagnosis of HIV infection in uninfected children. 29LB LOPINAVIR/RITONAVIR 1:1 SUPER-BOOSTING OVERCOMES RIFAMPICIN INTERACTIONS IN CHILDREN Helena Rabie 1 , Paolo Denti 2 , Janice Lee 3 , Mhleli Masango 4 , Ashraf Coovadia 4 , Sandy Pillay 5 , Afaaf Liberty 4 , Helen Mcilleron 2 , Mark Cotton 1 , Marc Lallemant 3 1 Stellenbosch Univ, Cape Town, South Africa, 2 Univ of Cape Town, Cape Town, South Africa, 3 Drug for Neglected Diseases Initiative, Geneva, Switzerland, 4 Univ of the Witwatersrand, Johannesburg, South Africa, 6 Enhancing Care Foundation, Wentworth Hospital, Durban, South Africa Background: Lopinavir/ritonavir 4:1 (LPV/r) is important in 1st-line antiretroviral therapy for infants. In high-burden settings, rifampicin based co-treatment for tuberculosis (TB) is often needed, but causes significant drug interactions. Superboosting LPV/r with ritonavir for a 1:1 ratio is considered effective, based on pharmacokinetic (PK) studies in 15 children. Methods: In an open-label, prospective study at 5 South African sites, we studied super-boosted LPV/r (1:1) during rifampicin co-treatment and LPV/r (4:1) thereafter in children weighing 3-15 kg. PK was studied on 3 occasions in each child with blood drawn at baseline and 1, 2, 4, 6 and 10 h post-dose: PK1 and PK2 after 2 and 5 months of rifampicin co-treatment (LPV/r 1:1); PK3 2-4 weeks after stopping TB therapy (LPV/r 4:1). Population PK modelling was used to interpret the data. PK1 data was used to develop a structural PK model for LPV, which was applied to PK2 and PK3 data to estimate all PK parameters. The uncertainty of PK parameters was obtained through a nonparametric bootstrap (n=500) and used for simulating 10 000 in silico patients, assuming a 30% decrease in clearance overnight to address known diurnal variation. The percentages of model-simulated (M-PK) C0 below 1 mg/L at PK2 and PK3 were compared for non-inferiority using a 10% delta threshold.

12

CROI 2017