CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

Methods: Children aged under 18-years at initiation of combination ART, in cohorts where ≥5% of children were born abroad (defined as migrants) were included. Follow-up was from ART start to death, last visit in paediatric care or 21st birthday. Hazard of first AIDS event or death among children AIDS-free at ART start was assessed by migrant status, using multivariable Cox models adjusting for sex, mode of transmission, age, weight-for-age z-score, WHO severe immunosuppression for age (based on CD4% or count) and calendar year at ART start. Missing baseline values were imputed. Sensitivity analysis assessed new/recurrent AIDS or death in all children, including those with AIDS at ART start. Results: Of 2,284 children from 11 European countries, 55%were migrants (of whom 85%were from Africa), the proportion of migrants varied by country, from 5% in Poland to 97% in Sweden. At ART start, migrant children were older than domestic-born (median 8.1[IQR 4.0,11.7] vs 1.8[0.3,7.2] respectively; p<0.001), and more likely to be severely immunocompromised (48.3% vs 42.3%, p=0.013) and TB co-infected (2.2% vs. 0.7%; p=0.003), but fewer had AIDS (14.5% vs. 18.7%; p=0.012). Median follow-up after ART start was 5.4[2.3,8.7] and 7.6[3.1,11.1] years, respectively (p<0.001). Of 1,901 children AIDS-free at ART start, 103 (5.4%) had ≥1 AIDS event and 14 (0.7%) died. The rate of AIDS/death was 1.02 {95% CI 0.85, 1.23}/100 person-years. Cumulative probability of AIDS/death at 5 years after ART start was 6.0% {4.6,7.7} in migrant vs. 5.0% {3.7,6.7} in domestic-born children (p=0.17). After adjustment, the hazard of AIDS/death was not significantly higher among migrant children (adjusted hazard ratio (aHR) 1.45 {0.90,2.33}, p=0.129). In sensitivity analysis including all children, 141 (6.2%) had ≥1 AIDS event and 36 (1.6%) died, cumulative probabilities of AIDS/death at 5 years were 7.1%migrant vs. 7.3% domestic (p=0.75), again with no effect of migrant status (aHR 1.29 (0.88,1.88), p=0.199). Conclusion: After adjusting for characteristics at ART start there was no increased risk of AIDS/death in migrant compared to domestic-born children in Europe. This may be partly due to the rarity of events or selection bias of long- term survivors among migrant children, or may indicate equality in care. 854 BETTER TREATMENT OUTCOMES FOR RWANDAN CHILDREN AFTER “TREAT ALL” Stephen M. Arpadi 1 , Matthew R. Lamb 1 , Isaie Nzeyimana 1 , Greet Vandebriel 1 , Gloria E. Anyalechi 2 , Marcia Wong 1 , Rebecca Smith 1 , Emilia M. Rivadeneira 2 , Eugenie Kayirangwa 3 , Samuel S. Malamba 3 , Canisious Musoni 3 , Emilia Koumans 2 , Sabin Nsanzimana 4 1 ICAP at Columbia University, New York, NY, USA, 2 CDC, Atlanta, GA, USA, 3 CDC, Kigali, Rwanda, 4 Rwanda Biomedical Centre, Kigali, Rwanda Background: In 2012, Rwanda expanded criteria for universal ART in HIV- infected children from 18 to 60 months of age regardless of CD4 count or clinical status. We compared antiretroviral therapy (ART) initiation and outcomes, before and after this change. Methods: We conducted a nationally-representative retrospective study of children 18-60 months enrolled in care between June 2009-December 2011 (Before Treat All cohort [BTA]) and July 2012-April 2015 (Treat All cohort [TA]). Probability proportional to size sampling identified 100 health facilities. We extracted medical records up to 14 months after enrollment for all eligible children. Differences in frequencies of health problems (reported symptoms or opportunistic infection), median and mean outcomes were compared using chi-square, Wilcoxon rank sum, and t-tests. We compared outcomes (loss to follow-up or death) with competing-risk cumulative incidence functions. Results: There were 374 children enrolled; 227 in the BTA and 147 in the TA. Mean (SD) age (3 years [1]) and WAZ (-2 [2]) at enrollment were similar across cohorts. Among the BTA, 59% initiated ART within one year, vs. 89% in the TA cohort. Median time to ART initiation was 68 days (interquartile range [IQR] 14-494) for the BTA and 9 days (IQR 0-28) for the TA (p<0.001); 9 children (5%) in the BTA had same-day initiation compared to 50 (37%) in the TA (p < 0.001). Prior to ART initiation, 59% of children in the BTA reported at least one health problem compared to 35% in the TA cohort (p < 0.001). Infectious diseases (BTA: 41%, TA: 24%, p = 0.001), diarrhea (BTA: 19%, TA: 10%, p = 0.02), and wasting syndrome (BTA: 5%, TA: 1%, p = 0.02) were more frequent in the BTA compared to the TA. After ART initiation, reported conditions were similar between cohorts. Viral load was done for 13% of the BTA compared to 44% of the TA on ART. Of 56 children on ART for > 6 months in the TA, 84% had a viral load < 1,000 copies/ml. Four children (2%) died prior to ART initiation in the BTA, compared to 1 (1%) in the TA (p=0.4). Overall loss to follow-up was similar

between cohorts (BTA: 13%, TA: 8%, p = 0.2) while loss to follow-up before ART initiation was higher in the BTA cohort (8% versus 2%, p = 0.02). Conclusion: Nearly 90% of Rwandan children 18 to 60 months old started ART within one year of enrollment; most within 1 month, with greater than 90% retention following implementation of 2012 guidelines. The ‘Treat All’ strategy is also associated with lower morbidity and better retention prior to ART. 855 AFRICAN MULTI-SITE 2-YEAR STUDY OF NEUROCOGNITION IN HIV INFECTED/AFFECTED CHILDREN Michael J. Boivin 1 , Miriam Chernoff 2 , Barbara Laughton 3 , Bonnie Zimmer 4 , Celeste Joyce 5 , Linda Barlow-Mosha 6 , Mutsa Bwakura-Dangarembizi 7 , Mmule Ratswana 8 , Lee Fairlie 8 , Portia Kamthunzi 9 , Katie McCarthy 10 , Patrick Jean- Philippe 11 , Avy Violari 8 , Mark Cotton 12 , Paul Palumbo 13 1 Michigan State University, East Lansing, MI, USA, 2 Harvard University, Cambridge, MA, USA, 3 Stellenbosch University, Cape Town, South Africa, 4 Frontier Science & Technology Research Foundation, Inc, Amherst, NY, USA, 5 Chris Hani Baragwanath Hospital, Johannesburg, South Africa, 6 Makerere University–Johns Hopkins University Research Collaboration, Kampala, Uganda, 7 Harare City Health Department, Harare, Zimbabwe, 8 Wits Reproductive Health and HIV Institute, Johannesburg, South Africa, 9 Malawi Department of HIV and AIDS, Lilongwe, Malawi, 10 FHI 360, Durham, NC, USA, 11 NIAID, Bethesda, MD, USA, 12 Stellenbosch University, Tygerberg, South Africa, 13 Dartmouth–Hitchcock Medical Center, Lebanon, NH, USA Background: We compared cognitive outcomes at weeks 0, 48, and 96 for HIV- infected (HIV), HIV-exposed uninfected (HEU), and HIV-unexposed/uninfected (HUU) cohorts of children at 6 sub-Saharan sites. Methods: IMPAACT P1060 compared Nevirapine (NVP) versus Lopinavir/ Ritonavir (LPVr)-based ARV in children (HIV+) 6 to 35 months of age. They were later enrolled for neurocognitive follow-up at 5 to 11 yrs, evaluating them at 3 annual time points, compared to age-matched HEU and HUU controls. Most HIV children in the NVP arm had been switched to 2nd line cART prior to the present study and 96%were virally suppressed (WHO Stage I=38 (15%); Stage II=58 (24%), Stage III=137 (56%); Stage IV=13 (5%)). They were tested with the Kaufman Assessment Battery for Children, 2nd ed. (KABC-II) cognitive ability, Tests of Variables of Attention (TOVA) attention/impulsivity, Bruininks- Oseretsky Test of Motor Proficiency (BOT-2), and parental Behavior Rating Inventory of Executive Function (BRIEF). Cohorts were compared using linear mixed models adjusted for site, child’s age and gender. Results: 611 (246 HIV+, 183 HEU, 182 HUU) of the 615 enrolled at 6 sites (South Africa [3], Zimbabwe, Malawi, Uganda), were compared across 3 assessment time points (weeks 0, 48, 96). 603 children completed week 48 and 588 completed week 96 visits. Mean age at enrollment was 7.2 years, 47%were male, and 69%were in school. 94% of caregivers were biological mothers (85% for HIV group), 32% completing high school, 22% on social grants, 38% in urban areas, 29% reporting sufficient family income. Cohort comparisons were consistent across time points, with the HIV cohort significantly worse than the HEU and HUU cohorts at all KABC-II, TOVA, BOT-2 global outcomes (P<0.001) (Figure 1, center and right). The HUU and HEU cohorts were comparable (Figure 1, left). On the BRIEF (parent ratings), the HIV cohort performed similarly or better at week 48 & 96, perhaps from biased caregiver perceptions. The magnitude of neurocognitive deficits among cohorts was consistent across the 3 assessments, except for HIV children for the KABC-II planning subtests, where they showed less improvement than the HUU/HEU groups. Conclusion: Despite 56% being Stage III at diagnosis, HIV children had excellent clinical care and robust virological suppression. Still, the HIV group had poorer neurocognitive function at all 3 assessment points, especially in terms of executive function across time. Such deficits pose a serious risk as these children age into adolescence.

Poster Abstracts

CROI 2018 323