CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

ml (blips), 3 for patient preference, 4 for discontinuing EFV (3 toxicity, 1 other), 1 for poor clinic attendance. Of the 13 CT with confirmed VL≥50c/ml, 7/13 resuppressed, 5 without a treatment switch and 2 with a switch to second-line. Over 144 weeks, 3 SCT vs 6 CT switched to second-line for viral failure (p=0.50); and 3 had new CDC B events (2 SCT, 1 CT). There were no significant differences between SCT and CT in grade 3/4 AEs (21 vs 20; p=1.00) or ART-related AEs (10 vs 17; p=0.48). Reports of missed ART in the week before a clinic visit (excluding days off in SCT) were similar in SCT and CT (6.6% v 7.8%, p=0.86). Conclusion: Sustainable non-inferiority of VL suppression in YP on EFV-based first line ART was demonstrated for SCT vs CT over almost 3yrs with >70% SCT patients remaining on the strategy; most YP experiencing viral rebound on SCT resuppressed on the same regimen on return to CT. SCT is a viable option for adherent HIV-1 infected YP on EFV-based first-line ART with 3-monthly VL monitoring. 814 COST-EFFECTIVENESS OF PREEMPTIVE SWITCH TO EFAVIRENZ (EFV) IN HIV-INFECTED CHILDREN Sophie Desmonde 1 , Simone C. Frank 2 , Ashraf Coovadia 3 , Désiré Lucien Dahourou 4 , Elaine J. Abrams 5 , Rochelle P. Walensky 2 , Kenneth Freedberg 2 , Louise Kuhn 6 , Valeriane Leroy 7 , Andrea L. Ciaranello 2 1 Univ of Bordeaux, Bordeaux, France, 2 Massachusetts General Hosp, Boston, MA, USA, 3 Empilweni Service and Rsr Unit, Johannesburg, South Africa, 4 Cntr de Rsr Intl Pour la Santé, Ouagadougou, Burkina Faso, 5 ICAP at Columbia Univ, New York, NY, USA, 6 Columbia Univ, New York, NY, USA, 7 INSERM, Toulouse, France Background: Pre-emptive switching to EFV for HIV-infected children suppressed on lopinavir/ritonavir (LPV/r) can simplify ART while preserving LPV/r for second-line treatment. The South African NEVEREST3 trial found this switch strategy to be non-inferior to remaining on LPV/r. The MONOD ANRS 12206 trial in Côte d’Ivoire showed similar 6-month outcomes, although limited by low event rates. We evaluated the cost-effectiveness of a preemptive switch to EFV. Methods: We used the Cost-Effectiveness of Preventing AIDS Complications (CEPAC)-Pediatric model to simulate South African children suppressed on LPV/r. The base case used NEVEREST3 data to examine 3 strategies: 1) Switch: switch to EFV-based ART (initial 24-week RNA suppression [<1000c/ml]: 98%; virologic failure (VF) after 24-week suppression: 0.15%/month), followed by return to 2nd-line LPV/r at observed VF; 2) WHO guidelines: Remain on LPV/r (VF after initial suppression: 0.22%/month), with switch to 2nd-line NNRTIs at observed VF; and 3) LPV/r only: Remain on LPV/r (VF after initial suppression: 0.22%/month) with no 2nd-line regimen, reflecting practice in many settings. We projected life expectancy (LE) and HIV-related healthcare costs. Sensitivity analyses varied all key clinical and cost parameters. A secondary analysis used data fromMONOD, which also reported EFV 24-week suppression of 98%, but higher monthly VF after initial suppression for EFV (0.72%) than for LPV/r (0.34%). Results: With NEVEREST3 data, the switch strategy led to the longest projected LE (21.38y) and lowest projected lifetime cost ($32,720; Table). Results were not affected by plausible variations in ART or healthcare costs, loss to follow-up rates, or age at time of LPV/r initiation or ART switch (unless this affected virologic outcomes on ART). Results were modestly sensitive to initial suppression rates: if 24 week RNA<1000c/ml was <90% for children switching to EFV (base-case value:98%), WHO guidelines was preferred. Results were most sensitive to risk of VF after initial suppression. This risk for children switching to EFV needed to be equal to or lower than that for children remaining on LPV/r for the switch strategy to remain the most effective and cost-saving. With MONOD data, for example, the WHO guidelines strategy maximized LE (20.21y) and minimized cost ($33,570; Table). Conclusion: For children similar to NEVEREST3 participants and suppressed on LPV/r-based ART, pre-emptive switching to EFV can improve clinical outcomes and be cost-saving.

Poster and Themed Discussion Abstracts

815 PREDICTORS OF SWITCH TO SECOND-LINE ART IN HIV-POSITIVE CHILDREN: A GLOBAL ANALYSIS Kara K. Wools-Kaloustian 1 , Colette J. Smith 2 , Ruth Goodall 2 , Intira J. Collins 2 , Elaine J. Abrams 3 , Mary-Ann Davies 4 , Mary Paul 5 , Jorge Pinto 6 , Russell B. Van Dyke 7 , for the CIPHER Durability of First Line ProjectTeam 1 Indiana Univ, Indianapolis, IN, USA, 2 Univ Coll London, London, UK, 3 ICAP at Columbia Univ, New York, NY, USA, 4 Univ of Cape Town, Cape Town, South Africa, 5 Baylor Coll of Med, Houston, TX, USA, 6 Univ Fed de Minas Gerais, Belo Horizonte, Brazil, 7 Tulane Univ, Metairie, LA, USA Background: Data on durability of first-line antiretroviral therapy (ART) in children across settings are needed to understand key factors associated with, as well as appropriateness of, switch to second-line ART. Methods: Data were pooled from 12 cohort networks within CIPHER. Children aged <18-years initiating combination ART (≥2 nucleoside reverse-transcriptase inhibitors [NRTI] plus non-NRTI [NNRTI] or boosted protease inhibitor [PI]) were included. Switch to second-line was defined as: (i) change of ≥1 NRTI plus either change in drug class (NNRTI to PI or vice versa) or PI change; (ii) change from single to dual PI; or (iii) addition of a new drug class. Factors associated with time to switch were explored using proportional hazards regression models with death and loss to follow-up (LTFU) as competing risks in (a) everyone, and (b) cohorts with routine CD4 monitoring. The effect of WHO immunological status for age was explored in model (b). Results: Of 93,213 children included, 22%were from Southern Africa (South Africa, Lesotho, Botswana), 68% from the rest of sub-Saharan Africa and 10% from other regions. At ART start, median [IQR] age was 3.9 [1.6, 6.9] yrs; CD4% 15% [10, 22%], 89% and 11% initiated NNRTI- and PI-based ART respectively. Median duration of follow-up from ART start was 27 [9, 54] mo; 1% died, 25% LTFU and 20% transferred out. Overall, 3979 (4.3%) switched to second-line at a median of 35 [19, 57] mo after ART start. At switch, median CD4% (n=2892) was 20% [11, 28%] and 12% (n=2287) had VL<400 copies/mL. The cumulative incidence of switch at 3 yrs after ART start was 1.4% (95% CI 1.3, 1.6%) in low- to 7.0% (6.6, 7.4%) in upper/upper-middle-income countries. In model (a) males, children who initiated ART at older ages, started an NNRTI-based regimen, were in settings with routine CD4 and VL monitoring and higher income countries had a significantly shorter time to switch (p<0.0001, Table). In model (b) (n=32,989), severe immunodeficiency at ART start was also associated with shorter time to switch (p<0.0001).

CROI 2017 352