People with HIV who already have resistance to a non-nucleoside reverse transcriptase inhibitor (NNRTI) when they start first-line antiretroviral treatment containing the integrase inhibitor dolutegravir were less likely to achieve viral suppression, a large clinical trial in South Africa has reported.
NNRTI resistance has not been shown to affect responses to dolutegravir-based treatment previously. The findings are reported in Nature Communications by an international team associated with the ADVANCE study.
Antiretroviral treatment containing the integrase inhibitor dolutegravir is recommended as the preferred option for everyone starting treatment by the World Health Organization. Most countries have introduced combination pills including dolutegravir to replace older drugs, including combinations containing the NNRTIs efavirenz or nevirapine.
Even if people with HIV have not taken a three-drug antiretroviral regimen in the past, they may have NNRTI resistance. Viruses with NNRTI resistance mutations are sexually transmitted. In the past, nevirapine alone was given to women around the time of delivery to prevent mother-to-child transmission of HIV but this practice often led women to develop resistance to the drug.
Research in South Africa carried out in 2017 suggests that around one in six people already had resistance to at least one antiretroviral drug before starting treatment and 15% had NNRTI resistance. A systematic review published in 2017 found high rates of NNRTI resistance in untreated people in eastern and southern Africa. The review also found that the rate of resistance in untreated people was growing by as much as 23% each year in southern Africa.
Findings of this sort led the World Health Organization to recommend that NNRTI-based treatment should be phased out and everyone should start treatment with an integrase inhibitor. Dolutegravir is the preferred option owing to its high barrier to resistance, low cost and lack of drug interactions.
Data from ADVANCE
The ADVANCE study compared two dolutegravir-based regimens to standard NNRTI-based treatment containing efavirenz in South Africa. The study was designed to inform the roll-out of dolutegravir in South Africa, which has the largest antiretroviral treatment programme in the world.
The trial showed that dolutegravir-based treatment was just as effective as efavirenz-based treatment. Researchers also carried out an analysis of study outcomes based on pre-treatment resistance profiles, expecting to find that pre-existing NNRTI resistance compromised treatment responses in people treated with efavirenz, but not dolutegravir.
Instead, they found that pre-existing drug resistance affected responses to treatment regardless of treatment regimen.
Study participants were tested for drug resistance before starting treatment (991 of 1,053 study participants provided samples for testing and virus could be successfully sequenced in 83% of these samples). Fourteen percent of samples had pre-treatment drug resistance and all but two of these samples had NNRTI resistance. Resistance to drugs in the nucleoside reverse transcriptase inhibitor class such as lamivudine was rare, detected in only 2% of those tested. Samples were not tested for resistance to integrase inhibitors, but previous studies have shown it to be extremely rare in South Africa.
Overall, 83% of study participants achieved viral suppression (<50 copies/ml). People with pre-treatment drug resistance were less likely to achieve viral suppression (65% vs 85%, p< 0.001) and this pattern was evident in people receiving efavirenz (60% vs 86%, p=0.002) and in those receiving dolutegravir (66% vs 84%, p<0.001).
People with pre-treatment drug resistance were around 60% less likely to achieve virological suppression (adjusted odds ratio 0.38, 95% CI 0.21-0.61) after adjusting for demographic factors and adherence.
The impact of pre-treatment drug resistance on treatment outcome was still evident when the researchers used a more stringent definition: persistent virologic failure (two consecutive viral loads above 200 copies/ml) (85% vs 94%, p=0.001 in the dolutegravir group, 68% vs 93%, p<0.001 in the efavirenz group).
However, pre-treatment drug resistance did not affect the extent of viral load reduction 12 weeks after starting treatment in the dolutegravir group. Whereas people with pre-treatment drug resistance in the efavirenz group had a significantly smaller reduction in viral load by week 12 of treatment compared to those without resistance, the same pattern was not detected in the dolutegravir group.
Explanations and implications
Dr Mark Siedner and colleagues say that theirs is the first study to find an effect of NNRTI resistance on responses to integrase inhibitor treatment. They point out that studies of dolutegravir monotherapy in France and the Netherlands have shown that dolutegravir treatment failure may take time to emerge, consistent with the finding in the ADVANCE study that pre-treatment drug resistance did not compromise early viral load responses to dolutegravir.
“NNRTI resistance might be a marker for people who are likely to experience adherence problems rather than the cause of treatment failure.”
A behavioural explanation cannot be ruled out, say the researchers, even though adherence (measured through self report and pill counts) didn’t affect the risk of poorer treatment response in multivariate analysis. It’s possible that people with pre-treatment drug resistance in this study may have dropped out of treatment in the past, did not disclose a previous history of antiretroviral treatment on joining the trial and may be more prone to adherence problems, the researchers speculate. In dolutegravir-treated patients, NNRTI resistance might be a marker for people who are likely to experience adherence problems rather than the cause of treatment failure.
They say further research is needed to find out if there is a resistance mechanism that explains the result or if it is behavioural. Also, the findings from ADVANCE need to be confirmed in similar cohorts.
Either way, the researchers say that regular virological monitoring for people taking dolutegravir is essential and that vigilance is needed when people switch to dolutegravir-based treatment. Testing for integrase inhibitor resistance needs to become part of resistance surveillance in resource-limited settings and more thought needs to be given to how to construct second- and third-line regimens in people after failure of dolutegravir-based treatment – especially those with pre-treatment NNRTI resistance.