Just over half of a sample of Nigerian patients on failing antiretroviral therapy had extensive cross-resistance to all nucleoside analogues, including tenofovir, according to findings from Nigeria’s national treatment programme presented on Wednesday at the 2008 HIV Implementers’ meeting in Kampala, Uganda.
The findings provide a sobering warning of what can happen in settings where largely unrestricted private sector prescribing of antiretroviral drugs is accompanied by a lack of viral load testing to detect cases of treatment failure, and could have important implications for the efficacy of second-line treatment in some developing countries.
If antiretroviral therapy fails to control HIV replication, viruses will gradually emerge with mutations that confer resistance to the drugs being used in the regimen, and the longer uncontrolled replication persists in the presence of these drugs, the higher the level of resistance that will emerge. The drug resistance mutations that develop may also cause resistance to drugs in the same class. This is a particular problem in the nucleoside analogue class, where mutations that develop during treatment with d4T can compromise the effectiveness of AZT, abacavir, ddI and tenofovir, depending on the pattern of mutations that emerges. A similar cross-resistance phenomenon can occur as a result of AZT failure.
Tenofovir failure, on the other hand, tends to result only in potential cross-resistance with abacavir and perhaps ddI, leaving AZT or d4T active in second-line treatment.
There are still few data on resistance patterns or the extent of resistance in people on failing antiretroviral treatment in resource-limited settings, due in part to lack of resistance testing but also because of the complex laboratory equipment and training needed to carry out resistance testing.
Antiretroviral treatment has been available in Nigeria through a government-funded treatment programme and private sources since at least 2005, but use was relatively unregulated, said Dr James Shepherd of the University of Maryland’s ACTION programme. The programme has been funded by the US President’s Emergency Plan for AIDS Relief to provide antiretroviral therapy through sites in a number of states in Nigeria, and is now providing treatment to 40,000 people.
The PEPFAR-funded programme initially offered free first-line treatment with either d4T or AZT plus 3TC and nevirapine or efavirenz, but in 2007 replaced d4T and AZT with tenofovir in order to reduce the risk of adverse events.
Patients on treatment were initially monitored for the emergence of clinical symptoms or declines in CD4 cell count as potential indicators of treatment failure. Viral load testing was introduced in November 2006 in order to spot treatment failure earlier: when viral load rises above 400 copies/ml it may be advisable to switch regimens in order to limit the risk that drug resistance will emerge, although some treatment programmes using viral load will wait until viral load has risen above 1,000 copies/ml.
In the Nigerian ACTION programme viral load testing was targeted at patients who had taken antiretroviral drugs prior to enrollment in the programme and at those with poor adherence, since it was thought that these patients had the highest risk of maintenance on failing treatment.
Resistance testing (genotyping was carried out in all patients identified with a viral load measurement above 400 copies/ml where a sample was available with viral load above 1000 copies/ml (genotyping is often impossible in blood samples in blood samples with lower viral concentrations).
Of 150 viral load tests carried out in patients who met the criteria, 101 were found to have detectable levels of HIV, and 79 had a viral load above 1,000 copies/ml.
Of the 79 samples, 73% had taken at least one antiretroviral drug prior to enrolment in the ACTION treatment programme, for a median of 636 days, and the patient group as a whole had received a median of 601 days of ART in the ACTION programme. Some patients had therefore been potentially exposed to long periods of sub-optimal treatment and viral replication despite triple combination ART, circumstances in which a high level of drug resistance might be expected to emerge.
Among the 79 patients who received resistance testing 78% had taken d4T, 12% AZT and 10% tenofovir, and the median viral load was 4.6 log. The median CD4 count prior to viral load testing was 112 cells/mm3.
Thirty per cent had at least four nucleoside analogue (NRTI)-related mutations, the mean number being 3.35, while 45% had at least two non-nucleoside reverse transcriptase inhibitor (NNRTI)- related mutations (mean 2.08).
A greater number of days on ART within the ACTION programme was strongly correlated with intermediate or high level resistance to all NRTIs, including tenofovir (p=0.02), and 53% of patients had no active NRTI available to them, while 30% were resistant to two NRTIs. Ten per cent were resistant to tenofovir, although this number should be treated with caution since the total number of patients studied was small.
Given these finding, the second-line antiretroviral options for a substantial proportion of these patients will be very limited, with over half having no active nucleoside analogues to pair with a boosted protease inhibitor (the World Health Organization’s recommended second-line option). Only 22% of the patients might be susceptible to etravirine, a new second-line NNRTI recently licensed in the United States, due to the large number of NNRTI-related mutations they had accumulated.
“We think there is a place for targeted viral load testing as a targeted strategy for reducing the accumulation of drug resistance mutations, “ said Dr Shepherd. “We feel that a new patient who has any history of prior drug treatment should undergo a timely resistance test.”
He suggested that after few months on an antiretroviral regimen, viral load testing would indicate whether the patient was likely to respond to the drug combination.
The findings are likely to add to the debate over when viral load testing should be used in resource-limited settings, and how widely it needs to be made available to safeguard the prospect of effective second-line treatment.
Shepherd J et al. High levels of class-wide NRTI resistance among HIV-positive patients failing first-line antiretroviral regimens in Nigeria. 2008 HIV Implementers’ Meeting, Kampala, Uganda, abstract 796.