Factors Associated with the Development of Drug Resistance Mutations in HIV-1 Infected Children Failing Protease Inhibitor-Based Antiretroviral Therapy in South Africa

Abstract

Objective: Limited data are available from the developing world on antiretroviral drug resistance in HIV-1 infected children failing protease inhibitor-based antiretroviral therapy, especially in the context of a high tuberculosis burden. We describe the proportion of children with drug resistance mutations after failed protease inhibitor-based antiretroviral therapy as well as associated factors.

Methods: Data from children initiated on protease inhibitor-based antiretroviral therapy with subsequent virological failure referred for genotypic drug resistance testing between 2008 and 2012 were retrospectively analysed. Frequencies of drug resistance mutations were determined and associations with these mutations identified through logistic regression analysis.

Results: The study included 65 young children (median age 16.8 months [IQR 7.8; 23.3]) with mostly advanced clinical disease (88.5% WHO stage 3 or 4 disease), severe malnutrition (median weight-for-age Z-score -2.4 [IQR -3.7;-1.5]; median height-for-age Z-score -3.1 [IQR -4.3;-2.4]), high baseline HIV viral load (median 6.04 log10, IQR 5.34;6.47) and frequent tuberculosis co-infection (66%) at antiretroviral therapy initiation. Major protease inhibitor mutations were found in 49% of children and associated with low weight-for-age and height-for-age (p = 0.039; p = 0.05); longer duration of protease inhibitor regimens and virological failure (p = 0.001; p = 0.005); unsuppressed HIV viral load at 12 months of antiretroviral therapy (p = 0.001); tuberculosis treatment at antiretroviral therapy initiation (p = 0.048) and use of ritonavir as single protease inhibitor (p = 0.038). On multivariate analysis, cumulative months on protease inhibitor regimens and use of ritonavir as single protease inhibitor remained significant (p = 0.008; p = 0.033).

Conclusion: Major protease inhibitor resistance mutations were common in this study of HIV-1-infected children, with the timing of tuberculosis treatment and subsequent protease inhibitor dosing strategy proving to be important associated factors. There is an urgent need for safe, effective, and practicable HIV/tuberculosis co-treatment in young children and the optimal timing of treatment, optimal dosing of antiretroviral therapy, and alternative tuberculosis treatment strategies should be urgently addressed.

Fig 1. Protease inhibitor (PI) drug resistance mutations (black bars denote major PI mutations, white bars denote minor PI mutations).

Fig 2. Kaplan-Meier curve of time to virological suppression among children with or without use of ritonavir as single protease inhibitor (RTV-sPI) at the time of protease inhibitor (PI) initiation.

Groups: (A) No RTV-sPI exposure or history of exposure to TB therapy; (B) No RTV-sPI exposure, on TB treatment at time of PI initiation; (C) History of RTV-sPI exposure, and on TB treatment at time of PI initiation.