ABSTRACT
Antiretroviral therapy (ART) of HIV infection has been one of the most successful public health interventions in
history, but the emergence of HIV drug resistance is a threat to ending the HIV epidemic. The next phase of the
global HIV response will rely on integrase strand transfer inhibitors (InSTIs), particularly dolutegravir (DTG), a
potent antiretroviral with a high genetic barrier to resistance. DTG is recommended for all lines of treatment (1st,
2nd- and 3rd-line) by the World Health Organization, and drugs from the same class are developed for pre-
exposure prophylaxis. There is limited understanding of the spectrum of DTG-selected mutations and their
effects on phenotypic susceptibility, particularly in non-B subtype viruses, and of the importance of mutations
outside of the integrase gene. It is critical to establish robust systems for early detection of InSTI resistance in
HIV treatment programs, and develop an understanding of where and how resistance to DTG occurs in real
world cohorts. We propose to do this in the NIH-funded International epidemiology Databases to Evaluate AIDS
(IeDEA) and the Antiretroviral Therapy Cohort Collaboration (ART-CC). Specifically, we aim i) to determine the
patterns and spectrum of InSTI drug resistance mutations in adults and adolescents with virologic failure on
DTG-based ART by regimen and HIV-1 subtype; ii) to identify risk factors for virologic failure and resistance in
adolescents and adults on DTG-based ART; and iii) to investigate correlations between novel resistance
genotypes and phenotypic InSTI resistance across HIV-1 subtypes. In total, 46 IeDEA sites and 7 ART-CC
cohorts from 32 countries will participate in this study. Ten sites and cohorts will contribute existing data (Sanger
sequences and clinical data), rather than prospectively recruiting patients; 19 sites and cohorts will contribute
pairs of samples or sequences at switch and failure, and 32 will contribute data on adolescents. We will include
adults (=18 years) and adolescents (10-17 years) with virologic failure (viral load =1000 copies/mL) on any DTG-
based ART. Analyses will be based on about 4600 patients failing DTG-based ART and about 490 patients with
DTG drug resistance mutations. We will perform whole genome sequencing using the Illumina MiSeq next-
generation sequencing platform to detect integrase mutations and mutations outside integrase (e.g. in the 3'
polypurine tract region of the nef gene). We will combine the data with existing sequence data from North
American and European cohorts. We will compare the prevalence of InSTI DRMs at different frequency
thresholds (20%, 5%, and 2%) by ART regimen and HIV-1 subtype. We will identify predictors of virologic failure
and drug resistance, and characterize the phenotypic effect of novel resistance genotypes by performing
phenotypic testing on selected specimens across the range of HIV-1 subtypes. Our proposal has three key
strengths: i) the pooling of clinical and sequence data across a vast global network to explore the determinants
of DTG resistance in people on 1st, 2nd-, and 3rd-line ART; ii) inclusion of multiple HIV-1 subtypes across
geographic areas and iii) combining expertise from clinical, epidemiological, biological, and computational fields.
