657 
Varied Drug Resistance Profiles following First-line Regimen Failure in the South Africa Antiretroviral Program
Carole Wallis*1, I Sanne1, F Venter1, J Mellors2, and W Stevens1,3
1Univ of the Witwatersrand, Johannesburg, South Africa; 2Univ of Pittsburgh, PA, US; and 3Natl Hlth Lab Svc, Johannesburg, South Africa
Background:
Despite 570,000 adults and children estimated to be accessing treatment, data
on drug resistance are limited from the South African national roll-out program.
This study sought to determine the frequency and types of drug-resistance mutations
following first-line regimen failure in the South African program.
Methods: Plasma from 269 HIV-1 subtype C-infected adults
with virologic failure on first-line regimens (n = 147—stavudine
[d4T], lamivudine [3TC], efavirenz [EFV]; n = 22—d4T, 3TC, nevirapine [NVP];
n = 54—zidovudine [AZT], 3TC, EFV; and n = 3—AZT, 3TC, NVP) was used to determine
resistance patterns. Protease and RT sequences were analyzed for drug
resistance mutations using the International AIDS Society–USA mutation list and
the ViroScore database.
Results: The
majority of patients failing d4T-containing regimens had both the M184V/I
mutation and NNRTI mutations (70%), but 16% had a NNRTI mutation only (K103N or
K101E), and 13% had wild-type virus. K65R was detected in 8 (5%) and the 151M
complex in 6 (4%). Half of Q151M mutations were associated with the K65R
mutation. The most common TAM was D67N (19%) alone. More than 1 TAM occurred infrequently
(9%) with the 67N pathway predominating (70%) over the 41L pathway. The V106M NNRTI
mutation was more frequent with failure of EFV-containing regimens (34%) than
NVP (2%). Different NNRTI-resistance profiles were observed depending on the
NRTI backbone. In patients on NVP, a more limited set of NNRTI mutations
occurred with AZT-containing regimens (K103N, Y181C, and G190A) than with d4T-containing
regimens (V90I, A98G, K101E, K103N, V106M, V108I, Y181C, Y188L, G190A, P225H),
which could affect response to etravirine.
Conclusions: The varied resistance patterns after failure of first-line therapy
indicate that resistance testing is important to identify appropriate second-line
therapy. “Blind” regimen switches are unlikely to provide optimal second-line
treatment responses. Sentinel surveillance for drug resistance should be part
of national treatment programs to identify effective and low-cost regimens.
|
