CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

1 Univ of California, Los Angeles, Los Angeles, CA, USA, 2 McGill Univ, Montreal, Quebec, Canada Background: Approximately 6.2 million individuals are living with HIV infection in South Africa, where only half are receiving antiretroviral treatment. The non-nucleoside reverse transcriptase inhibitor Efaverinz (EFV) has been widely used in first-line therapies since 2007, which has led to high levels of acquired and transmitted resistance; the predominant mutation being K103N. Testing for resistance strains in South Africa is seldom offered upon virological failure. In resource-rich countries, Dolutegravir (DTG) a new integrase inhibitor, has replaced EFV in first-line regimens. DTG has the highest clinical barrier against resistance of any antiretroviral, however it can occasionally lead to the mutation R263K if used in second-line therapies. We model the impact of (i) introducing resistance testing upon HIV diagnosis and during treatment, and (ii) switching first-line treatment from EFV to DTG, on reducing HIV drug resistance in South Africa. Methods: We use a transmission model, parameterized with epidemiologic data from South Africa to reconstruct treatment history, including acquired and transmitted resistance. We conduct an uncertainty analysis using Latin Hypercube Sampling and a multivariate sensitivity analysis based on response hypersurface modeling. Input variables are treatment rates for different CD4-stages of infection, drug adherence/efficacy and the frequency of testing for resistance. Results: If current treatment conditions are maintained, incidence will decrease from 300,000 in 2016 to 170,000 in 2030, but the number of individuals with the K103N mutation will increase to 1,400,000. Introducing resistance testing will prevent 330,000 new K103N infections by 2030 (pink curve). Introducing testing and switching to DTG will prevent 860,000 new K103N infections (blue curve) and also ~1,200,000 cases of acquired resistance. Notably, only a small number of individuals with the R263K mutation will be seen by 2030, and DTG will prevent the greatest number of new infections compared to current conditions and introducing testing. Conclusion: Introducing resistance testing into South Africa will substantially reduce the transmission of drug resistant strains, but have no effect on acquired resistance. Introducing testing and switching to DTG-based first-line therapies, reduces TDR even further and essentially prevents the development of acquired resistance, indicating that DTG has the potential to eliminate drug resistance in South Africa. 490 IMPACT OF PRETREATMENT HIV-DRUG RESISTANCE ON VIROLOGIC OUTCOME OF FIRST-LINE ART Ingrid A Beck 1 , Molly Levine 1 , Ross Milne 1 , Isaac So 1 , Nina Andersen 1 , Michael Watling 2 , Michael H. Chung 2 , James Munyao 3 , Catherine Kiptinness 4 , Lisa Frenkel 2 1 Seattle Children’s Rsr Inst, Seattle, WA, USA, 2 Univ of Washington, Seattle, WA, USA, 3 Coptic Hope Cntr, Nairobi, Kenya, 4 Univ of Washington, Nairobi, Kenya Background: Pre-treatment HIV-drug resistance (PDR) to WHO-recommended 1st-line antiretroviral treatment (ART) is increasing in low-resource communities. Data on the risk that specific PDR mutations confer for virologic failure (VF) of ART could inform policy makers in their management of HIV resistance. We assessed the risk that PDR at single or multiple codons conferred for VF during EFV- or NVP-based ART. Methods: Kenyans initiating 1st-line ART within studies in 2006, 2010 and 2014 were evaluated for PDR mutations K103N, Y181C, G190A, M184V and K65R by an oligonucleotide ligation assay (OLA) sensitive to 2%mutant within a subject’s viral population. Consensus sequencing and 454-pyrosequencing or Illumina assays confirmed OLA results in those with PDR and/or VF. VF was defined as plasma viral load ≥ 400 HIV RNA copies/mL at month-12 of NNRTI-ART. Rates of VF at month-12 were compared (Fisher’s exact) between ART regimens and between mutants and wild-type (WT) genotypes during EFV- or NVP-based ART. Results: A total of 1228 individuals given NNRTI-ART had month-12 virologic outcome and PDR data for mutations K103N, Y181C, G190A and M184V, and a subset of 922 for K65R. 30/518 (5.8%) subjects on EFV-ART had VF compared to 97/710 (13.7%) on NVP-ART (p<0.0001). PDR was detected in 63/1228 (5.1%) individuals; among these, 23.8% (5/21) given EFV-ART had VF vs. 69% (29/42) given NVP-ART (p=0.001). No VF was observed in subjects with low frequency mutants (2-9%) in the EFV-ART group (0/5) compared to 63.6% (7/11) in the NVP-ART group (p=0.034). Among subjects with PDR, single NNRTI mutations were detected in 39 (61.9%), single NRTI mutations in 3 (4.8%), and multiple NNRTI/ NRTI mutations in 21 (33.3%). In subjects given NVP-ART, those with single or multiple mutations had higher rates of VF compared to those with WT genotype (p<0.0001). In contrast, in subjects given EFV-ART, those with multiple mutations had increased rates of VF compared to those with WT (p=0.003), but those with a single NNRTI mutation had rates of VF similar to those with WT HIV (p=0.165) (Fig 1). Conclusion: The lower rate of VF among subjects receiving EFV-ART compared to NVP-ART, regardless of whether infected with WT or DR HIV, emphasizes the greater potency of EFV-ART. A lower risk of VF conferred by single NNRTI mutations during EFV-ART suggests that use of assays to detect and manage PDR could maximize viral suppression and extend the use of EFV-ART in low-resource settings.

Poster and Themed Discussion Abstracts

491 RESPONSE TO FIRST-LINE ART IN ADULTS WITH DRUG RESISTANT HIV, ANRS 12249 TASP TRIAL

Collins C. Iwuji 1 , Anne Derache 2 , Kathy Baisley 3 , Siva Danaviah 4 , Tulio de Oliveira 5 , François Dabis 6 , Kholoud Porter 1 , Deenan Pillay 2 , for the ANRS 12249TasP Study Group 1 Univ Coll London, London, UK, 2 Africa Hlth Rsr Inst, Mtubatuba, South Africa, 3 London Sch of Hygiene and Trop Med, London, UK, 4 Africa Cntr for Pop Heath, Mtubatuba, South Africa, 5 Univ of KwaZulu-Natal, Durban, South Africa, 6 Univ de Bordeaux, Bordeaux, France Background: Mathematical models suggest that high levels of transmitted drug resistance (TDR) could compromise ART programmes. We assessed the impact of pre-treatment drug resistance (PDR) on viral suppression (VS) [viral load [VL] <400 copies/mL] in adults who initiated 1st-line ART within the cluster randomised TasP trial in rural KwaZulu-Natal Methods: HIV-positive adults enrolled fromMarch 2012–June 2016 and initiated ART with VL data available were eligible. HIV whole genome sequences (WGS) were generated on Illumina MiSeq in recently-infected (RIn) adults with available plasma samples and chronically-infected (CIn) ART-naïve adults. WGS were assembled using Geneious software;

CROI 2017 205