Abstract Body

Background: Resistance mutations may vary by viral subtype, although data to date are limited, especially for non-B subtypes. Understanding the common mutational patterns is important for determining the optimal standardised regimens for the public health approach to ART in settings without resistance testing or individualised therapy.

Methods: Genotypes were obtained from stored baseline samples from 792 patients aged ≥12 years who met WHO treatment failure criteria after >12 months on NNRTI-based first-line ART in the EARNEST trial in 4 sub-Saharan African countries. Subtype and drug susceptibility were determined by REGA and Stanford algorithms respectively. Presence of specific mutations and intermediate-high level resistance was modelled using multivariable logistic regression including subtype, ART exposure at time of first-line failure, ART prior to the failing regimen, years on first-line ART, and WHO 4 events, CD4 and VL at failure.

Results: Patients had advanced treatment failure (42% VL≥100,000 c/ml, 63% CD4<100 cells/mm3). Viral subtypes were A1 (40%; Uganda, Kenya), C (31%; Zimbabwe, Malawi) and D (25%; Uganda, Kenya) with 4% recombinants/unclassified. One or more major NRTI or NNRTI mutations were found in 774 (98%) and mutations to both classes in 747 (94%). In adjusted analyses, 4 NRTI and 7 NNRTI mutations differed significantly across the three subtypes (p<0.05; Table 1); most of these 11 mutations were significantly more common in subtype C than in A and/or D. Intermediate/high level TDF resistance (seen in 57%) was independent of subtype (adjusted p=0.38). Intermediate/high level ZDV resistance (in 71% overall) was marginally more common in subtype C (78% (adjusted) than D (76%, p=0.05) and A (76%, p=0.06). Intermediate/high level resistance to ETR or RPV (in 51% and 62% respectively overall) was more common in subtype C (adjusted 63% and 76% respectively) than D (51% and 59% respectively, p<0.02) and A (47% and 60% respectively, p<001). Just 46% patients would be switched to a second-line regimen with predicted ≥1 active NRTI using the WHO algorithm (based on first-line NRTI history), but genotyping would identify a more active regimen in only an additional 35 (4%).

Conclusions: We found differences between subtypes in resistance mutations at first-line failure. The impact on residual NRTI drug susceptibility was modest. However, subtype C developed higher rates of ETR and RPV resistance, limiting their potential utility in salvage regimens in resource-limited settings.

viewer?TAG_ACTION=SECURE_FILE_DOWNLOAD&V
“Prevalence of resistance mutations by subtype”