CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

Results: Among primiparous women under the age of 30 years, the proportion of PHIV women increased from 4.1 % to 17.7 % (p<0.001) over the study period. At the time of this first pregnancy, PHIV and non-PHIV women had similar living conditions, with about half not living with a partner, 40% being jobless, and less than 10% having psychoactive substance abuse. PHIV women were significantly more likely to have been born in France (73.9% vs 15.9%, p<0.001), and to be on combined antiretroviral treatment at the time of conception (71.7% vs. 54.1%; p=0.02). In women on treatment at conception, the proportion of uncontrolled viral load (VL) was higher for the PHIV than for the non-PHIV group during the first trimester (44.8% vs. 21.4%; p=0.007), as well as at delivery (26.7% vs. 10.6%; p=0.03). For those born outside France, virological failure at delivery remained more frequent in PHIV women, even after adjusting for initial VL (adjusted odds ratio: 8.3; 95% CI: 1.6-42.8; p=0.01). The prevalences of obstetric complications and neonatal outcomes were very similar in the two groups. No case of MTCT occurred in the PHIV group (upper CI =0.1%), versus three cases (0.7%; CI: 0.1-1.9) in the non-PHIV group. Conclusion: Pregnancy and neonatal outcomes were similar in PHIV and non-PHIV women. However, despite similarities in marital and employment status, uncontrolled viral load was more frequent at conception and delivery in pregnant PHIV women, possibly due to difficulties in maintaining treatment adherence for long periods since childhood. Strategies needed to support adherence to treatment in PHIV women of childbearing age should be reinforced during all the course of pregnancy. 776 ANTENATAL ANTIRETROVIRAL THERAPY AND ADVERSE BIRTH OUTCOMES: THE PROMISE TRIAL Benjamin H. Chi 1 , Mona Farhad 2 , Maxensia Owor 3 , Tsungai Chipato 4 , Min Qin 2 , Sufia Dadabhia 5 , Devasena Gnanashanmugam 6 , Nahida Chakhtoura 7 , Mary Glenn Fowler 5 , Jeffrey S. Stringer 1 1 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 Harvard Univ, Boston, MA, USA, 3 Makerere Univ–Johns Hopkins Univ Rsr Collab, Kampala, Uganda, 4 Univ of Zimbabwe, Harare, Zimbabwe, 5 The Johns Hopkins Univ, Baltimore, MD, USA, 6 DAIDS, NIAID, Rockville, MD, USA, 7 NICHD, Bethesda, MD, USA Background: The PROMISE trial found that antiretroviral therapy (ART) in pregnancy reduced mother-to-child transmission, but also increased the frequency of several adverse birth outcomes (ABOs) compared to antenatal zidovudine alone. Methods: PROMISE randomized HIV-infected women ≥14 weeks gestation and not in labor to receive 1 of 3 antenatal regimens: ZDV only (Arm A), ZDV+3TC+LPV/r (Arm B), or TDF+FTC+LPV/r (Arm C). In the early versions of the protocol, women could be randomized only to Arms A and B unless they tested HBsAg+; in version 3.0, women were randomized with equal probability into all 3 arms. We studied the association between antiretroviral regimen and ABOs: delivery <37 weeks (PTD); infant birthweight <2500g (LBW); composite of PTD, LBW, stillbirth (SB), and spontaneous abortion (AB); delivery <34 weeks (VPTD); infant birthweight <1500 g (VLBW); and composite of VPTD, VLBW, SB, and AB. Gestational age at delivery was estimated primarily by Ballard score. We adjusted for baseline factors (maternal age, BMI, HIV viral load, CD4, alcohol use, country, gestational age at entry) and obstetrical complications across all multivariable models. If there were zero counts in specific strata, the variable as a whole was not included in the model. In primary analysis, we included data from all participants; in sensitivity analysis, we restricted data to only those enrolled in version 3.0. Results: 3423 women who enrolled and delivered in PROMISE were allocated to Arm A (n=1507), Arm B (n=1497), or Arm C (n=419). When we considered outcomes with PTD and/or LBW, women on ZDV+3TC+LPV/r (Arm B) and TDF+FTC+LPV/r (Arm C) each had higher risk for ABO compared to ZDV alone (Arm A). When analysis was restricted to severe outcomes (i.e., VPTD, VLBW), the risk associated with Arm C remained elevated. In head-to-head comparisons between the two ART regimens, Arm C had a higher risk of severe ABO such as VPTD (AOR: 2.55, 95%CI:1.46-4.44) and VLBW (AOR: 3.06, 95%CI:1.23-7.59). Findings remained consistent with protocol version 3.0 data alone. Conclusion: LPV/r-containing ART was associated with a significantly elevated risk of ABO after adjustment for multiple obstetrical and clinical factors. For severe outcomes, this risk was higher among women on TDF-FTC compared to ZDV-3TC. Further study is needed to determine whether this is an independent effect of TDF-FTC, a result of drug-drug interactions with LPV/r, or due to other factors.

Poster and Themed Discussion Abstracts

777 RISK FACTORS FOR LOW BIRTH WEIGHT AND PRETERM DELIVERY IN THE PROMISE TRIAL

Dorothy Sebikari 1 , Min Qin 2 , Maxensia Owor 1 , Jeffrey S. Stringer 3 , Mona Farhad 2 , Nahida Chakhtoura 4 , Benjamin H. Chi 3 , Karin Klingman 5 , Gerhard Theron 6 , Mary Glenn Fowler 7 1 Makerere Univ–Johns Hopkins Univ Rsr Collab, Kampala, Uganda, 2 Harvard Univ, Boston, MA, USA, 3 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 4 NICHD, Bethesda, MD, USA, 5 DAIDS, NIAID, Bethesda, MD, USA, 6 Stellenbosch Univ, Cape Town, South Africa, 7 The Johns Hopkins Univ, Baltimore, MD, USA Background: Although antiretroviral therapy (ART) in pregnancy can reduce vertical HIV transmission to <1%, it may also increase the risk of low birth weight (< 2500g, LBW) and preterm delivery (<37 weeks, PTD), conditions that confer significant morbidity and mortality to newborns in resource-limited settings. In the multi-site PROMISE trial, we previously reported an increased risk of LBW and PTD among women initiating protease inhibitor (PI)-based ART during pregnancy, when compared to ZDV alone. We further describe obstetrical and clinical risk factors for LBW and PTD among study participants. Methods: Within the antepartum component of PROMISE, we assessed baseline clinical and obstetrical risk factors associated with LBW and PTD. Risk factors with p-value <0.15 in univariate logistic regression were included in multivariate backward logistic regression models. We also adjusted for treatment arm, gestational age (GA) at entry, and country. Results: Birth outcomes were available for 3423 HIV-infected women delivering between 4/2011-11/2014 across 14 sites in Africa and Asia. Among the 3333 women delivering at least one live born infant, median maternal age at enrollment was 26 years (IQR 22–30); 661 (20%) were primiparous, and 110 (3.3%) reported at least one prior PTD. Median birth weight was 2900g (IQR 2600–3200); and 558 (17%) infants weighed <2500 g. Median GA at birth was 39 weeks (IQR 38–40); 557 (17%, 95%CI: 16.1%-18.9%) were born prior to 37 weeks. In univariate analyses, clinical factors including maternal age 18-<21 year and entry RNA ≥20,000 copies were significant for PTD but not LBW; however, maternal age 18-<21 dropped out in the backward logistic model. In the final multivariate models, adjusted for country and GA at entry, obstetrical risk factors for LBW and/or PTD included BMI, multiple gestation, prior PTD, pregnancy or chronic hypertension, IUGR, placental abruption, preterm labor, oligohydramnios, PROM, and antenatal ART were significant risk factors (table).

CROI 2017 336