CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

TB/HIV co-infected children who are treated with the maximized weight-based efavirenz dosage during anti-TB treatment will be comparable to concentrations in HIV-infected children receiving ART alone. Methods: ART-naïve HIV-infected children aged 3 – 14 years old were enrolled and given ART regimen consisting of efavirenz (10-13.9kg – 200 mg; 14-24.9kg – 300mg; 25-39.9kg – 400 mg and > 40Kg – 600 mg) per WHO recommended weight-band dosing, plus zidovudine 180 - 240 mg/m2 and lamivudine 4 mg/ kg twice daily. For the TB/HIV co-infected patients, anti-TB treatment using the newWHO recommended TB drug dosages for children was started immediately upon TB diagnosis and ART started within 2 to 8 weeks of TB therapy. Blood samples were collected at times 0, 2, 8, 12 and 24 hours post-dose after 4 weeks of ART in both arms. Efavirenz concentrations in plasma were measured using validated LC/MS/MS assays and pharmacokinetic parameters calculated using noncompartmental analysis. Pharmacokinetic parameters were compared by rank sum test. Results: Of the 72 patients, 38 (53%) had TB coinfection. Children with TB coinfection compared to those with HIV infection alone were younger, had lower body weight and height but received a higher efavirenz dose (median, 15 mg/kg vs. 13 mg/kg, P = 0.008). TB/HIV co-infected patients had significantly lower efvairenz Cmax, Cmin and AUC0-24h compared to those with HIV alone (see table below). The proportion of children with efavirenz Cmin < 1 µg/ mL (considered subtherapeutic) was also higher among those with TB/HIV coinfection than those with TB alone (47.4% vs. 17.6%, P = 0.008). Conclusion: This is the first study to investigate effect of first-line anti-TB drug regimen using new higher drugs dosages on efavirenz pharmacokinetics in children. Unlike the findings of adults and prior pediatric studies, 4-drug anti-TB therapy in the co-infected children was associated with significant reduction in efavirenz plasma exposure and trough concentrations. The effect of anti-TB treatment on long-term HIV treatment outcome in TB/HIV co-infected children need to be evaluated.

836 MARKERS OF PRETERM DELIVERY IN HIV+WOMEN; ROLE OF PROTEASE INHIBITORS AND VITAMIN D Adriana Weinberg 1 , Yanling Huo 2 , Deborah Kacanek 2 , Kunjal Patel 2 , Heather Watts 3 , Diane Wara 4 , Jelena Klawitter 1 , Uwe Christians 1 1 University of Colorado Anschutz Medical Campus, Aurora, CO, USA, 2 Harvard University, Cambridge, MA, USA, 3 National Institute of Child Health and Human Development, Bethesda, MD, USA, 4 University of California San Francisco, San Francisco, CA, USA Background: HIV+ women have increased risk of spontaneous preterm delivery (SPD). We identified plasma biomarkers associated with the risk of SPD in HIV+ women and their correlations with factors that were previously shown to affect the risk of SPD, such as HIV suppression, vitamin D deficiency, and use of PI in pregnancy. Methods: Plasma was obtained from 103 HIV+ women with SPD (≤35 wg) and 205 controls with term delivery (≥37 wg) matched to cases 2:1 by race and gestational age at blood draw. Women with obstetric risk factors for SPD were excluded. Plasma levels of inflammatory markers TNFα, IFNγ, IL6, IL8, IL1β, IL18, IL17, GCSF, MCP1, IP10, sIL2Rα, sCD14, VEGFA, MCSF, GROα and MMP9; anti- inflammatory IL10, TGFβ and sCTLA4; and eicosanoids were compared between cases and controls using conditional logistic regression, adjusted for HIV- unrelated risk factors of SPD and inflammation. Weighted linear regression was used to evaluate associations of viral suppression, vitamin D deficiency (<20 ng/ mL), and PI use in pregnancy with plasma biomarkers correlated with SPD. Results: Among 308 women (mean age=29y, BMI=31; 58% black, 33% Hispanic), 80%were on PI and 19%were on ARV without PI. At plasma collection, 68%were in the 3rd trimester of pregnancy, 63% had CD4>350 cells/µL and 76% had VL<400 HIV RNA c/mL. Demographic and HIV disease characteristics were similar in the two groups. In adjusted analyses, higher levels of sIL2Rα were significantly associated with increased odds of SPD (aOR=2.97, p=0.01). Higher levels of sCD14, GCSF, PGF2α and 5-HEPE were marginally associated with greater odds of SPD. Women who initiated PI before or during the 1st trimester, but not later in pregnancy, had higher levels of GCSF (Estimated Difference=0.15, p=0.03) and 5-HEPE (Est. Diff.=0.26, p=0.01) compared with women on ARV without PI. Vitamin D deficiency was associated with higher sCD14 (Est. Diff.=0.06, p=0.04), higher PGF2α (Est. Diff 0.16, p=0.02), and lower 5-HEPE (Est. Diff.=-0.25, p=0.002). No associations were found with HIV suppression. Conclusion: The best plasma predictor of SPD in HIV+ women was sIL2Rα, a marker of T cell activation. Monocyte activation (sCD14; GCSF), PGF2α (uterine prostaglandin; promotes contraction), and 5-HEPE (epoxide; induces regulatory T cells) were also associated with the risk of SPD. Vitamin D deficiency and use of PI in early pregnancy may increase the risk of SPD by modulating monocyte activation and/or the metabolism of eicosanoids, hypotheses that warrant further testing. 837 EFFECT OF ANTITUBERCULOSIS THERAPY ON THE PHARMACOKINETICS OF EFAVIRENZ IN CHILDREN Awewura Kwara 1 , Hongmei Yang 2 , Sampson Antwi 3 , Anthony Enimil 3 , Fizza S. Gillani 4 , Anima M. Sarfo 3 , Antoinette Ortsin 3 , Albert Dompreh 3 , Lubbe Wiesner 5 , Charles A. Peloquin 1 1 University of Florida, Gainesville, FL, USA, 2 University of Rochester, Rochester, NY, USA, 3 Komfo Anokye Teaching Hospital, Kumasi, Ghana, 4 Brown University, Providence, RI, USA, 5 University of Cape Town, Cape Town, South Africa Background: Efavirenz-containing antiretroviral therapy (ART) is the preferred regimen in children older than 3 years receiving rifampin-containing antituberculosis (anti-TB) therapy. To date, there is limited data on the drug- drug interactions between efavirenz and 4-drug anti-TB therapy in children. We hypothesized that at the population level, efavirenz plasma concentrations in

Poster Abstracts

838 PHARMACOKINETICS OF 8-HOURLY LOPINAVIR/RITONOVIR IN CHILDREN ON RIFAMPICIN Helena Rabie 1 , Holly Rawizza 2 , Peter Zuidewind 1 , Jana L. Winckler 1 , Heather Zar 3 , Annelies Van Rie 4 , Jennifer Norman 3 , Helen Mcilleron 3 1 Stellenbosch University, Tygerberg, South Africa, 2 Brigham and Women’s Hospital, Boston, MA, USA, 3 University of Cape Town, Cape Town, South Africa, 4 University of Antwerp, Antwerp, Belgium Background: In children requiring lopinavir/ritonavir (LPV/r) and rifampicin, doubling the dose of LPV/r-4:1 fails to achieve lopinavir Cmin>1mg/L in 60%. Adding ritonavir to a 1:1 ratio LPV/RTV-1:1 achieves lopinavir exposures comparable to those in children not receiving rifampicin, but this strategy is complex and ritonavir alone is not widely available. Pharmacokinetic modeling predicts that adjusted 8-hourly dosing of LPV/r -4:1 could achieve adequate Cmin in >95% of children receiving rifampicin. This study tests that hypothesis. Methods: This pharmacokinetic study evaluated lopinavir concentrations among South African children treated with co-formulated LPV/r-4:1 oral solution and rifampicin-based TB treatment. Children where switched from LPV/RTV-1:1 dosing (South African standard of care) to 8-hourly LPV/r-4:1 dosed according to weight, with children in the 3-5.9, 6-9.9, 10-13.9 and 14-19.9kg weight bands receiving approximately 27, 21, 20, and 18mg/kg lopinavir per dose, respectively. Two weeks after starting 8-hourly LPV/r-4:1, 2-10 hours sampling after the morning LPV/r-4:1 dose was performed, where after children were switched back to standard LPV/RTV-1:1. Alanine Aminotransferase (ALT) were performed to assess safety. Rifampicin was dosed according to the WHO guidelines.

CROI 2018 316