An analysis of people who experienced treatment failure on their first-line HIV therapy with a viral load of over 400 copies/ml and who had taken either the triple-combination pill Atripla (tenofovir/emtricitabine/efavirenz), regimens containing its component drugs as separate pills, or regimens substituting lamivudine for emtricitabine has found that, although the rate of virological failure was no higher in people taking separate pills rather than Atripla, they developed virological failure faster and were more likely to develop drug resistance.
This study has implications for the replacement of combination pills like Atripla with individual generic drugs as a cost-saving strategy as HIV drugs come off patent.
Who was in the study
The study looked at data from eight prospective clinical cohorts in Europe and North America and four randomised controlled trials (RCTs). To be included in the analysis, participants had to be taking either Atripla, Truvada (tenofovir/emtricitabine) plus efavirenz, tenofovir plus emtricitabine plus efavirenz, or tenofovir plus lamivudine plus efavirenz (lamivudine has similar properties to emtricitabine and the same drug-resistance profile). They also had to have had a drug resistance test within three months of viral failure. Most participants were taking the combination as their first anti-HIV drug regimen, though people who had taken other drugs but never experienced virological failure were also allowed.
Sixty-four people out of 871 participants in the four RCTs met these criteria (an overall rate of 7% viral failure on these drugs) and 122 people out of an estimated 4800 in the much more varied cohorts. The RCTs all took place between 2000 and 2008 and the cohorts were followed between 2000 and 2010.
In total, 58 people were on Atripla and 128 on non-Atripla regimens.
In the cohort group, 58 were on Atripla, 52 on Truvada + efavirenz, 18 on tenofovir + emtricitabine + efavirenz and 58 on tenofovir + lamivudine + efavirenz.
In the RCT participants, 11 were on Truvada + efavirenz, 13 on tenofovir + emtricitabine + efavirenz, and 40 on tenofovir + lamivudine + efavirenz. Because Atripla was not licensed until late 2006 in the US and late 2007 in Europe, none of the RCT participants took the single pill because their protocols had been designed earlier, so comparisons of viral failure of Atripla versus multi-pill regimens was not possible in RCT participants.
Because of this, the researchers did a second analysis looking solely at failure and drug resistance rates in the cohort study participants.
Rates of treatment failure
In the cohort studies, there was no significant difference in the virological failure rates of the four regimens: the failure rate was 18% on Atripla, 22% on Truvada + emtricitabine, 19% on tenofovir + emtricitabine + efavirenz, and 20% on tenofovir + lamivudine + efavirenz.
There was, however, a notable difference in the time to failure. Over all participants, the time to virological failure was 350 days on Atripla and 211 days on the non-Atripla regimes, and this difference was statistically significant (p = 0.03).
Not everyone who experienced virological failure developed drug resistance but about two-thirds (68%) did. One-sixth developed resistance to tenofovir, one third to emtricitabine or lamivudine, and 61% to efavirenz.
There was a notable difference in the proportion who developed drug resistance between the people on Atripla and those on non-Atripla regimens: 53% in Atripla developed at least one drug-resistance mutation versus 74% on the other regimens, and this was highly statistically significant (p = 0.005). Because not all drug resistance mutations in HIV have the same cost in terms of restricting the choice of subsequent regimens, the researchers also calculated a Genotypic Sensitivity Score (GSS) which calculates the impact of a person’s HIV resistance on their choice of other drugs. The results were essentially the same: 52% on Atripla had significant resistance according to their GSS versus 72% on the other regimens (p = 0.01).
The researchers also did an analysis looking only at the cohort study participants and found almost exactly the same results: 52% on Atripla versus 75% on the other regimens had a GSS indicating significant drug resistance.
In terms of the likelihood of developing resistance to individual drugs, the individual numbers of participants were too low to detect any significant difference between Atripla and non-Atripla regimens, other than that in the cohort-only analysis participants on the non-Atripla regimens were significantly more likely to develop resistance to efavirenz (p = 0.01).
Emtricitabine versus lamivudine
There was a difference in the patterns of resistance seen in people on lamivudine versus any of the regimens containing emtricitabine. Although these two drugs are often regarded as almost identical in their properties, in fact emtricitabine reaches higher levels in the body and stays there longer.
There was no difference in the Atripla and other-regimen groups in the proportion of people who developed the M184V/I mutation, which confers resistance to lamivudine or emtricitabine. However, there was a significant difference in the proportion of people developing resistance to tenofovir. In people taking emtricitabine, one in eight developed resistance to tenofovir. In people taking lamivudine, it was one in four and this difference was statistically significant (p = 0.01). (The presence of one drug can cause resistance to another if it allows HIV to reproduce in the presence of the second drug.) In the cohort-only analysis the difference between the two drugs was larger, with one in six who took emtricitabine developing resistance to tenofovir but one in three who took lamivudine.
There was also a tendency in the cohort-only group for people on lamivudine to develop more resistance to efavirenz (78% versus 61.5% on emtricitabine) but this was not statistically significant.
Tenofovir resistance is potentially more serious than resistance to lamivudine or emtricitabine, as it may also cause resistance to some other drugs of the same type, the nucleoside reverse transcriptase inhibitors (NRTIs). In addition, a small proportion of participants (3% in the Atripla and 5% in the non-Atripla groups) developed resistance not only to efavirenz but to the second-generation drugs of the non-nucleoside reverse transcriptase inhibitor (NNRTI) family, etravirine and rilpivirine.
In conclusion, it is important to emphasise that there was no indication of a higher rate of treatment failure in people taking individual drugs rather than Atripla. However, if treatment did fail, it seemed to do so somewhat faster in multi-pill regimens, and was more likely to result in drug resistance.
The differences may be rather slight in terms of clinical effect; but the authors argue that reintroducing multi-pill regimens to people who have taken a single pill in order to save money may represent “a step backwards in the achievement of many therapeutic advances regarding patients’ quality of life, simplicity, adherence and efficacy.”
Blanco JL et al. Lower prevalence of drug resistance mutations at first-line virological failure to first-line therapy with Atripla versus tenofovir + emtricitabine/lamivudine + efavirenz administered on a multiple tablet therapy. AIDS 28, early online publication, doi: 10.1097/QAD.0000000000000424. 2014.