Virologic versus immunologic monitoring and the rate of accumulated genotypic resistance to first-line antiretroviral drugs in Uganda

Steven J Reynolds,Andrew Kambugu,Kevin Newell,Yukari C Manabe,Hakim Sendagire,Thomas C Quinn,Immaculate Nankya,Moses Kamya,Barbara Castelnuovo

doi:10.1186/1471-2334-12-381

Abstract

BackgroundViral load monitoring (VLM) to identify individuals failing antiretroviral therapy (ART) is not widely available in resource-limited settings. We compared the genotypic resistance patterns between clients with VLM versus immunological monitoring (IM).MethodsBetween 2004–2008, 559 ART naïve clients were enrolled in a prospective cohort, initiated on ART, and monitored with viral load (VL) and CD4+ cell counts every 6 months (VLM group). From February 2008 through June 2009, 998 clients on ART for 36–40 months (corresponding to the follow-up time of the VLM group) at the same clinic and monitored with CD4+ cell counts every 6 months were recruited into a cross sectional study (IM group). Samples from VLM clients at 12, 24 and 36 months and IM clients at 36–40 months with VL > 2000 copies/ml underwent genotypic drug resistance testing.ResultsBaseline characteristics were similar. Virologic failure (VL > 400 copies/ml) at 12, 24 and 36 months in the VLM group were 12%, 6% and 8% respectively, and in the IM group 10% at 36–40 months. Samples from 39 VLM and 70 IM clients were genotyped. 23/39 (59%) clients in the VLM group (at 12, 24 or 36 months) compared to 63/70 (90%) in the IM group, (P < 0.0001) had at least 1 non-nucleoside reverse transcriptase mutation. 19/39 (49%) of VLM clients had an M184V mutation compared to 61/70 (87%) in the IM group (P < 0.0001). Only 2/39 (5%) of VLM clients developed thymidine analogue mutations compared to 34/70 (49%) of IM clients (P < 0.0001).ConclusionsRoutine VL monitoring reduced the rate of accumulated genotypic resistance to commonly used ART in Uganda.

Highlights

Viral load monitoring (VLM) to identify individuals failing antiretroviral therapy (ART) is not widely available in resource-limited settings
Some VLM clients had more than one time point sent for genotyping resulting in a total of 45 unique genotypes for this group; among clients with >1 genotype, the first available result was used to compare VLM and immunological monitoring (IM) clients as this was considered to be the time point when a switch of therapy would be considered
The most common class of resistance was to the non-nucleoside reverse transcriptase inhibitors (NNRTIs) ranging from 50% (8/16) at 12 months, 69% (9/13) at 24 months and 60% (6/10) at 36 months

Summary

Introduction

Viral load monitoring (VLM) to identify individuals failing antiretroviral therapy (ART) is not widely available in resource-limited settings. We compared the genotypic resistance patterns between clients with VLM versus immunological monitoring (IM). Antiretroviral treatment (ART) programs have scaled up to reach over 5.2 million HIV-infected individuals in need of life-saving treatment in low and middle income countries by the end of 2009 [1]. Most of these individuals live in settings where laboratory monitoring is limited and treatment failure is determined using either clinical or immunological criteria. The objective of our study was to compare genotypic resistance patterns (with particular attention to TAMs) among a group of clients on first-line ART at an urban clinic in Kampala, Uganda, who were monitored both clinically, immunologically and virologically as part of a research cohort study to a similar group of clients attending the same clinic who were monitored only clinically and immunologically

Objectives

Methods

Results

Discussion

Conclusion