Pre-treatment drug resistance as high as 17% with no impact on viral load suppression
‘Surprising’ research from Conference on Retroviruses and Opportunistic Infections (CROI) shows drug resistance does not impact test-and-treat roll-out in South Africa – at least in the short term.
Pre-treatment HIV drug resistance (PDR) in South Africa was found to have no impact on antiretroviral treatment (ART) outcomes in ART-naïve people enrolled in a ‘test-and-treat’ programme in KwaZulu-Natal, despite high levels (up to 17%) of drug resistance in this population.
At a session presented last week at the Conference on Retroviruses and Opportunistic Infections (CROI) in Seattle, study author Anne Derache presented findings from drug resistance testing completed on 1,337 newly and chronically infected people living with HIV in the South Africa Treatment as Prevention clinical trial (TasP).
Drug resistance presents a major threat to the roll-out of antiretroviral treatment programmes, particularly in low-resource settings where access to second- and third-line treatments are limited. The consequences of drug resistance include treatment failure and further spread of drug resistant HIV.
The findings show high levels of drug resistance in this study, with around 9% of the study population having HIV strains with mutations that were over 20% (majority variant) resistant to HIV, and 17% showing mutated strains that were between 2% and 20% (minority variant) drug resistant.
They found minimal differences in prevalence of HIV drug resistance between people who were newly infected (transmitted drug resistance) and chronically infected (pre-treatment drug resistance).
Of the people with PDR, the majority (88%) only had one or two drug-resistant mutations – largely associated with non-nucleoside reverse transcriptase inhibitors (NNRTIs) mutations. NNRTI resistance accounted for 73% of drug resistant cases, of which 61% were majority variant.
Resistance to nucleoside reverse transcriptase inhibitors (NRTI) and protease inhibitors were less common and occurred at low levels (the minority variant) – at 13.5% and 8.8% respectively.
Derache commented: “You can see that it is mostly the NNRTI class that is compromised and mostly driven by the K103N mutation, which is not really a surprise because we know that this mutation persists longer without antiretroviral pressure.”
The K103N mutation is shown to have high-level resistance to the NNRTIs efavirenz and nevirapine. Of people starting World Health Organization (WHO) recommended first-line treatment consisting of efavirenz (EFV) + emtricitabine (FTC) + tenofovir (TDF), at least 7% of those with highly drug resistant strains would be starting ART with only two of three drugs having full effect.
Derache continued: “if we look at the two other drugs, FTC and TDF, they were quite preserved and most of the people will be susceptible to those drugs.”
While the large majority of patients only had one drug mutation, 11% of the study cohort had three or more drug-resistant mutations. Sometimes they had up to seven different mutations – the likely result of previous undisclosed exposure to antiretroviral drugs.
The study also looked at treatment outcomes for patients on the fixed dose combination pill Atripla, (EFV + FTC + TDF) – as defined by viral load suppression.
To the surprise of researchers, pre-treatment resistance at baseline did not affect viral suppression – a development which had not previously been observed by any publication. At the median follow up time of 16 months, there was no difference between the group with PDR and the control group. The only predictors of poor viral suppression were high viral load at baseline and low adherence.
Derache added that the results contrasted with a range of other studies, notably one in Mexico released at the end of last year which showed that pre-treatment drug resistance significantly reduced the effectiveness of first-line ART regimens.
Speaking at a press conference, Derache explained that more research needed to be done, with long-term follow-up.
Efavirenz is a very effective drug even at a high resistance level, so it is possible to have successful treatment on it even when drug-resistant mutations present – even more so because the large majority of the patients only had one mutation present.
Derache also said that the setting of the study, conducted within a clinical trial environment with treatment adherence counselling, could have a positive effect on the viral load suppression that other studies have not been able to demonstrate.
The results beg the question of the need for expensive HIV drug resistance testing (genotypic and phenotypic testing) in low-resource settings. While this is the norm for high-income countries, access is severely limited across sub-Saharan Africa, and the WHO recommends other means for monitoring drug resistance in these contexts, such as monitoring of viral load and clinic visits, etc.
In any case, as drug resistant mutations are present in a large proportion of the patients, long-term follow-up is needed to ensure that viral suppression continues as they present a high risk for treatment failure.
Understanding drug resistance will be integral to the mass roll-out of treatment programmes in high-burden settings such as South Africa.
Community guidelines for comments can be found in our website T&Cs