Foundations of ART Management in Heavily Treatment–Experienced Patients

The current recommendations for selecting a new regimen suggest including ≥2 fully active agents, 2 being adequate if ≥1 of them has a high barrier to resistance, such as DTG, boosted darunavir (DRV), or bictegravir (BIC).¹ These are drugs for which resistance mutations are relatively rare in those who are failing therapy. In some cases, virologic suppression can be achieved with a fully active drug combined only with partially active drugs.

If no fully active drug with a high barrier to resistance is available, then every effort should be made to include 3 fully active drugs. The definition of “fully active” is that there exists no predicted resistance based on treatment history or resistance testing. Those drugs with a novel mechanism of action or those in classes to which the patient has not been exposed should also be fully active.

There are certain drugs in classes for which the person may have underlying resistance that may still be fully active, such as etravirine, which was developed to have activity in people who had failed earlier-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs). DRV was similarly developed to be active in select people who had previously failed with resistance to other PIs. Dolutegravir (DTG), an integrase strand transfer inhibitor (INSTI), is a third agent that has this property. 

The other important thing that we have learned from the early history of sequential monotherapy is to not add only 1 drug to a failing regimen. This puts the patient at risk for sequentially accumulating increasingly resistant virus.

DTG is a second-generation INSTI, developed to have activity even in patients with virologic failure to regimens including the first-generation INSTIs raltegravir (RAL) and elvitegravir. RAL and elvitegravir had high potency and good tolerability, but low barriers to resistance, meaning that when patients failed regimens with these drugs, integrase resistance was common.

VIKING-3 studied DTG 50 mg given twice daily—as opposed to the usual once-daily dosing—given with an optimized background regimen (OBR) to people who had previously failed first-generation INSTIs with integrase resistance.⁵

In the overall population, 69% and 63% achieved virologic suppression at Weeks 24 and 48, respectively, with OBR plus twice-daily DTG. It was learned from VIKING-3 that the group with underlying integrase resistance that benefited most were those who did not have the Q148 mutation. It was further noted that there was partial benefit in those who had the Q148 and 1 additional secondary mutation, but not in those with Q148 and ≥2 secondary mutations.

DRV was originally developed as a salvage PI. It was given boosted with ritonavir (RTV) twice daily to patients in whom previous-generation PIs had failed. We then learned that in patients who had no underlying resistance, DRV/RTV could be given once daily—and that’s how it’s more commonly used today.^1,6

DRV resistance mutations have been defined and are shown on this slide.⁶ We know that if patients have experienced virologic failure on a PI-based regimen and their resistance tests do not show these mutations, once-daily boosted DRV should be active.

If, however, these mutations are present, twice-daily boosted DRV still may be partially or fully active. These 2 examples—DTG and DRV—show that some drugs from classes previously used may still be active in select patients.

Several studies in low-income countries investigated whether we can truly use “recycled” drugs. The study on this slide, EARNEST, looked at patients with clinical failure (by World Health Organization criteria) of a first-line regimen consisting of 2 nucleos(t)ide reverse transcriptase inhibitors (NRTIs) plus an NNRTI.⁷ At the time this study was done, HIV-1 RNA testing was infrequent and resistance testing was often unavailable in this setting.

The study population in EARNEST was randomly assigned to receive lopinavir (LPV)/RTV (a boosted PI) plus 2-3 NRTIs (selected by the physician without the benefit of resistance testing), LPV/RTV plus RAL, or LPV/RTV monotherapy. At the time of this study, all of the participants would have been PI and INSTI naive.

Regardless of how many active NRTIs were determined to be in the regimen (by subsequent baseline resistance testing), the rate of virologic response was similar in those receiving LPV/RTV plus NRTIs. That is, approximately 80% of those who received LPV/RTV plus any number of NRTIs achieved virologic suppression that lasted through 144 weeks.

The LPV/RTV plus RAL arm was equally active.

The LPV/RTV monotherapy arm fell short, however, with only 60% maintaining virologic suppression at 96 weeks. 

What we must conclude is that, even when the NRTIs were not found to be fully active, they drove the success of the PI-based regimen. The group that received the PI alone had a higher rate of virologic failure over the course of the study.

What we learned from this and several other similar studies is that if one includes a fully active boosted PI with a high barrier to resistance and any other active drug with it—in this case RAL or NRTIs—resulting virologic response rates in these second-line regimens will be very high. And even if the NRTIs included are not fully active, they will be associated with greater activity than no NRTI. 

The DAWNING study asked a similar question but using the second-generation INSTI DTG to compose the second-line regimen.⁸ This study looked at patients with first-line virologic failure with a regimen of 2 NRTIs plus 1 NNRTI, again in a low-income country. In this case, however, resistance testing was available in real time. Participants were randomized to receive either DTG or LPV/RTV, both with 2 NRTIs. One of the NRTIs had to be fully active based on genotypic resistance testing.

This study was stopped early because of superior efficacy of the DTG-based regimen. We learned from this study that DTG—just like boosted PIs—can be associated with a high rate of viral suppression when given with 2 NRTIs, but in this case, ≥1 of the NRTIs had to be fully active.

An important observation made in this study is that 2 of the patients who received DTG developed emergent INSTI resistance over the course of the study. That is different than what was observed with LPV/RTV, where no patients had emergent resistance. DTG clearly has a higher barrier to resistance than many drugs, but maybe not quite as high as a boosted PI.

These are data looking at virologic response among those who had the M184V/I mutation (associated with resistance to lamivudine [3TC] and emtricitabine [FTC]) with or without other mutations at baseline.⁹ The rate of virologic response was very high in those who received 3TC or FTC and who had M184V (with or without other reverse transcriptase [RT] mutations), similar to those who did not have M184V/I or those with mutations who received other NRTIs than 3TC and FTC. 

This emphasizes the fact that DTG appears to have very high suppressive activity even in those who have underlying NRTI resistance. But in this case, ≥1 of the NRTIs was active; in EARNEST there was a high virologic response rate even in those who had no fully active NRTIs in the regimen.