Big differences in virological outcomes among people taking first-line HIV therapy in resource-limited settings

Michael Carter
Published: 16 October 2013

Virological outcomes vary enormously among people taking antiretroviral therapy in different resource-limited settings, according to research published in the online edition of Clinical Infectious Diseases. An international team of investigators monitored outcomes in patients in five African and two south-east Asian countries 12 and 24 months after starting HIV therapy. World Health Organization (WHO) approaches for the delivery and monitoring of antiretroviral therapy were used in all seven countries. Rates of virological failure differed significantly between settings, from a low of 3% to a high of 26%. 

“Treatment outcome using the [WHO] public health approach varies significantly from one setting to another,” comment the authors. “The differences in VF [virological failure] cannot be explained by study populations, ARV [antiretroviral] regimes or biological monitoring which were similar at all sites…other factors should therefore be considered, and local programme factors or local management of clinics and healthcare centres are most likely associated with the heterogeneity of virological outcomes.”

Access to antiretroviral therapy in resource-limited settings is expanding rapidly. This has in part been possible due to the adoption of the WHO model for the delivery and monitoring of treatment. This takes into account limited financial resources of low- and middle-income countries and the logistical constraints of their health systems. WHO recommends viral load monitoring where possible, but this is not feasible in most settings.

Studies conducted soon after the introduction of ARV rollout programmes in low- and middle-income countries showed that patient outcomes were at the very least equal to those observed in richer countries. However, subsequent research has highlighted alarming high rates of virologic failure accompanied by the development of drug-resistant virus.

An international team of investigators therefore analysed 12- and 24-month outcomes among patients who started HIV therapy in seven low- and middle-income countries in Africa (Burkina Faso, Cameroon, Ivory Coast, Senegal and Togo) and south-east Asia (Thailand and Vietnam).

Blood samples were taken at routine clinic visits and sent for viral load testing. A viral load above 1000 copies/ml was the threshold for virologic failure. Samples with a viral load above this figure were subjected to genotypic resistance testing.

Viral load for approximately 4000 patients taking first-line theray was sampled between October 2009 and December 2011. Women represented 70% of patients from African countries, whereas men were in the majority in the south-east Asia.

Data on 12-month outcomes were analysed for 2060 patients and 24-month outcomes for 1875 individuals.

Treatment regimens included d4T (stavudine, Zerit) or AZT (zidovudine, Retrovir) with 3TC (lamivudine, Epivir) and either efavirenz (Sustiva or Stocrin) or nevirapine (Viramune). A very small number of patients received tenofovir (Viread), but none were treated with a protease inhibitor.

Overall, 11% of patients experienced virological failure at month 12 and 12% at month 24.

Detailed analysis, however, revealed that outcomes varied significantly between countries.

Very low rates of treatment failure at both month 12 and month 24 (3 to 5%) were observed in Burkina Faso (group A countries). The frequency of treatment failure after one or two years of treatment varied between 9 and 14% in Cameroon, Senegal and Vietnam (group B). In group C countries (Ivory Coast and Togo), between 18 and 20% of patients had a viral load above 1000 copies/ml after twelve months of treatment. This increased to between 14 and 26% at month 24.

Between 10 and 50% of patients had a viral load between 1000 copies/ml and 5000 copies/ml, “indicating that using the current WHO threshold for failure at 5000 copies/ml could lead to significant misclassification of failing patients”.

Almost three-quarters (71%) of patients experiencing failure at month twelve developed strains of drug-resistant virus, and there was an 86% prevalence of drug resistance among patients with virological failure after two years of therapy.

The proportion of patients with two or more resistance mutations varied between 7 and 24%. Most of the mutations concerned the drugs used in first-line therapy, but a few patients developed cross-resistance to similar drugs.

The investigators were encouraged by the low prevalence of resistance observed in group A countries, which they note was “unexpected” and “similar, if not better than those obtained from countries where patients obtain adequate treatment monitoring, including viral load assessment, generally industrialised countries or well monitored cohort studies”.

The authors stress that the differences in outcomes between countries cannot be explained by differences in study populations or HIV treatment regimens. Rather, they suggest, “Other factors should…be considered, and local programme factors or local management of clinics and healthcare centres is most likely associated with the heterogeneity of virological outcomes.” The investigators call for “better programme management and implementation of actions to improve factors as patient adherence, drug stock-outs and lost to follow-up”.

The author of an accompanying editorial does not believe that the high failure rates in observed in some countries can be ascribed to lack of routine viral load monitoring and stresses that the drug-resistance testing is not a feasible option in most resource-limited countries.

“Debate will continue about balancing the cost of drugs, monitoring and heath systems…for countries struggling against poverty and high burdens of HIV,” he concludes.

Reference

Aghokeng AF et al. Extraordinary heterogeneity of virological outcomes in patients receiving HAART and monitored with World Health Organization public health approaches in sub-Saharan Africa and south-east Asia. Clin Infect Dis, online edition, 2013.

Katzenstein DA et al. HIV RNA and genotype in resource limited settings. Can we do better? Clin Infect Dis, online edition, 2013.