Background: Tenofovir (TDF)-based 1st-line antiretroviral therapy (ART) is globally recommended by the World Health Organization (WHO). Effects of switching from prior non-TDF to TDF-based regimens on ART success are unknown. High rates of the TDF-associated K65R mutation in HIV-1 subtype C, but not B, are reported, associated with a poly-A template in reverse transcriptase (RT) positions 64-66. Such mechanisms in other HIV-1 subtypes are not well-defined.
Methods: We examined treatment failure and resistance in Kenyan patients on ≥6 months of TDF-based 1st-line ART at the academic model providing access to healthcare (AMPATH). We compared these outcomes among patients with (Prior-ART Group) and without (TDF-Only Group) prior non-TDF-based 1st-line ART. Pol sequences of patients with detectable VL (>40 copies/mL) were interpreted with Stanford Database tools using Fisher exact and Wilcoxon tests.
Results: Among 332 enrolled patients (55% female, median age 41 years, 63% XTC/TDF/nevirapine; 37% XTC/TDF/efavirenz; median CD4 336 cells/μL), detectable VL was in 17%, and VL>1,000 copies/mL (WHO cutoff) in 10%. Of those, 216 were in the TDF-Only Group (median 20 months on ART), and 116 in the Prior-ART Group (median 24 months on TDF-based ART and 47 months on prior ART). Detectable VL was in 23% of the TDF-Only Group and 7% of the Prior-ART Group (p<.001). TDF-only failing patients had lower CD4 values (p<.001) and higher WHO stage (p=.03). VL>1,000 copies/mL was seen in 15% of the TDF-Only Group and 1% of the Prior-ART Group (p<.001). In 35 available genotypes from the TDF-Only Group 69% were subtype A, 11% D, 12% C and 9% AD. RT resistance was in 89%, 89% to NNRTIs, 83% to NRTIs and 83% dual-class. K65R occurred in 69% (24/35) of patients; subtype C 100% (4/4), subtype A 71% (17/24) and subtype D 50% (2/4) (p=.22). Patients with K65R had lower CD4 values (p<.02), higher WHO stage (p=.004) and more RT resistance mutations (p<0.001). Examination of RT positions 64-66 from subtype A infected ART-naïve persons from the same Kenya clinic (n=32) and from the Stanford Database (n=3,903), demonstrated similarity to subtype B, without the poly-A chain that promotes K65R in subtype C.
Conclusions: In this Kenyan cohort, switching from non-TDF to TDF-based 1st line ART was successful. Higher failure rates were seen in patients with only TDF-based ART compared to those with prior 1st-line ART exposure. High observed K65R rates in subtype A were not explained by the poly-A template, suggesting alternative mechanisms.