Factors Predicting Discordant Virological and Immunological Responses to Antiretroviral Therapy in HIV-1 Clade C Infected Zulu/Xhosa in South Africa

Boris Julg,Bruce D Walker,Danielle Poole,Henry Sunpath,Carmen Castilla,Richard A Murphy,Musie Ghebremichael,Marcus Altfeld

doi:10.1371/journal.pone.0031161

Abstract

Factors predicting suboptimal CD4 cell recovery have been studied in HIV clade-B infected US and European populations. It is, however, uncertain to what extent these results are applicable to HIV clade-C infected African populations. Multivariate analysis using logistic regression and longitudinal analyses using mixed models were employed to assess the impact of age, gender, baseline CD4 cell count, hemoglobin, body mass index (BMI), tuberculosis and other opportunistic co-infections, and frequencies of regimen change on CD4 cell recovery at 12 and 30 months and on overtime change in CD4 cells among 442 virologically suppressed South Africans. Despite adequate virological response 37% (95% CI:32%–42%) and 83% (95% CI:79%–86%) of patients on antiretroviral therapy failed to restore CD4 cell counts ≥200 cells/mm3 after 12 and ≥500 cells/mm3 after 30 months, respectively, in this South African cohort. Critical risk factors for inadequate recovery were older age (p = 0.001) and nadir CD4 cell count at ART initiation (p<0.0001), while concurrent TB co-infection, BMI, baseline hemoglobin, gender and antiretroviral regimen were not significant risk factors. These data suggest that greater efforts are needed to identify and treat HAART-eligible patients prior to severe CD4 cell decline or achievement of advanced age.

Highlights

The roll-out of combination antiretroviral therapy has reduced AIDS-related morbidity and mortality in South Africa [1,2,3,4]
Despite the potency of HAART, it has been observed that a significant proportion of patients – despite an adequate virological response – do not achieve CD4 cell recovery above critical thresholds associated with opportunistic infections
The need to better understand factors influencing CD4 cell restoration was underscored by a study from Cape Town, South Africa which found that CD4 cell count was the variable most strongly associated with mortality risk during HAART in South Africa and that a high cumulative mortality risk was associated with person-time accrued at low CD4 cell counts [2]

Summary

Introduction

The roll-out of combination antiretroviral therapy has reduced AIDS-related morbidity and mortality in South Africa [1,2,3,4]. Factors affecting CD4 cell restoration have been studied in Western cohorts it is uncertain to what extent these findings are applicable to African populations with HIV-1 clade C virus and distinct demographic as well as epidemiological characteristics [5,6,7,8,9,10,11,12]. In this study we retrospectively examined CD4 cell count trajectories of 442 HIV-1 clade-C infected Zulu/Xhosa in KwaZulu-Natal Province, South Africa, who presented with CD4 cell counts ,200 cells/mm at HAART initiation. Age, gender, hemoglobin, body-mass-index (BMI), baseline CD4 cell counts, presence of opportunistic infections (OI) and antiretroviral therapy regimen on overtime change in CD4 cell counts and CD4 cell restoration at 12 and 30 months Using mixed and logistic regression models we assessed the predictive value of baseline patient factors, incl. age, gender, hemoglobin, body-mass-index (BMI), baseline CD4 cell counts, presence of opportunistic infections (OI) and antiretroviral therapy regimen on overtime change in CD4 cell counts and CD4 cell restoration at 12 and 30 months

Methods

Results

Conclusion