EACS: Kaletra monotherapy may need excellent adherence to work

Two studies of using lopinavir/ritonavir (Kaletra) in a one-drug-only anti-HIV regimen (monotherapy) have both found that the success of this strategy depends on excellent adherence. In both cases, poor adherence was the biggest single risk factor associated with monotherapy failure.

In the French MONARK Study, which randomised patients to receive either Kaletra alone or Kaletra plus nucleosides from the start of therapy, monotherapy was somewhat less successful than conventional combination therapy.

In the Spanish OK Study, which randomised patients either to change to monotherapy after at least six months of Kaletra+ NRTIs combination therapy or to stay on combination therapy, better results were achieved with monotherapy than with conventional therapy when an ‘on treatment’ analysis was used, which did not count as failures those patients who had to reintroduce their NRTIs as long as they re-suppressed their HIV. When an ‘intent to treat’ analysis was used, which did regard those patients as failures, monotherapy and combination therapy were equally successful.

Philippe Flandre of the French research institute INSERM presented the 48-week results of the MONARK trial. In this, 136 patients were randomised to receive either Kaletra alone or Kaletra and AZT/3TC (Combivir).

The 48-week results, in brief, showed that by intention-to-treat analysis where any discontinuation or change of therapy was classed as ‘failure’, 67% of patients on Kaletra alone and 75% of patients on Kaletra plus Combivir achieved viral loads under 50 copies/ml. On an ‘on-treatment’ analysis, where everyone who achieved an undetectable viral load on the regimen was a success even if they had to stop or change for other reasons, 84% on monotherapy and 98% on triple therapy – all but one patient – achieved a viral load under 50 copies/ml.

Flandre’s principal interest in the EACS presentation was in the fact that the response to monotherapy seemed to vary by viral subtype, with a smaller proportion of patients with subtypes other than B responding to monotherapy.

Eighty-seven per cent of patients with subtype B who were given monotherapy achieved a viral load under 50 copies/ml compared with 64% with non-B virus. Subtype B is the type found in the developed world, and if there is a biological reason why people with non-B virus respond less well to monotherapy it would be significant, as this is likely to be tried as a money-saving second-line option in poor countries.

One interesting aspect of the study was that for the first six months the non-B patients had a better response than the subtype B patients to monotherapy. By week four 59% of subtype B patients but 70% of non-B patients had a viral load under 400 and this advantage persisted until week 16. After that, however, the proportion of patients who had undetectable viral load was higher in those with subtype B.

The strongest predictor of whether patients would succeed on monotherapy was the viral load at week four. Patients who had a viral load under 400 copies/ml by week four were 7.6 times more likely to be responders at 48 weeks than patients who were still above 400 by this time.

It appears this may have been something to do with adherence. Nineteen per cent of responders to monotherapy but 31% of non-responders admitted they had missed doses, and Flandre showed a slide which indicated that adherence in patients with non-B subtype, which had been similar to the rest of the group, worsened after 24 weeks. He told aidsmap he could not speculate on whether this was due to the psychological effect of less frequent monitoring visits in the latter phases of the study.

In the other study, non-adherence and the nadir (lowest-ever) CD4 count were also the strongest predictors of virological failure on lopinavir/ritonavir monotherapy, and the main reason for changing from monotherapy was low-level but detectable HIV in the blood.

However the lead investigator of the study, Dr José Arribas, told the conference that this was outweighed by discontinuations due to drug toxicity in patients who stayed on nucleosides. The result was that after 96 weeks Kaletra monotherapy was at least non-inferior and nearly superior to Kaletra plus two NRTIs.

The crucial difference between MONARK and OK is that OK is an ‘induction-maintenance’ study. Patients – who mustn’t have failed on protease inhibitors before – had to be on stable combination therapy comprising Kaletra plus any two NRTIs for at least six months. After this half of them were randomised to Kaletra monotherapy while the other half stayed on their previous regimen. All discontinuations for whatever reason were counted as ‘failure’.

At the end of 96 weeks, on a strict intention-to-treat analysis, monotherapy was just as good as combination therapy.

Seventy-seven per cent of patients on monotherapy had viral loads under 50 compared with 77.6% on combination therapy.

Kaletra monotherapy looked better in an ‘on treatment’ analysis. Here, patients who experienced low-level HIV rebound were not counted as ‘failures’, despite resuming NRTI treatment, as long as they ended up with undetectable viral load again.

On this basis, 94.4% of monotherapy patients (including those who’d started NRTIs again) versus 86.4% of patients who’d taken triple therapy throughout had viral loads under 50 copies/ml at 96 weeks. This difference was not quite, but nearly statistically significant (p=0.06), so Arribas couldn’t say monotherapy was ‘superior’, but it was ‘non-inferior’.

The main reason for the difference was drop-outs due to NRTI toxicity. Thirteen out of 100 patients on monotherapy failed in the on-treatment analysis, and most of these (eight patients) were lost to follow-up. There were only three virological failures (two with the appearance of protease inhibitor resistance), and one regimen change for toxicity reasons. One person died, not of an HIV-related cause.

In the combination therapy arm, 22 out of 98 patients failed therapy. Five developed virological failure and protease inhibitor resistance – more than on monotherapy – and nine were lost to follow-up. The main difference was that eight dropped out due to NRTI toxicity.

However this toxicity was almost entirely due to drugs that have fallen out of favour and are now being used less. Six of the eight were taking d4T (stavudine, Zerit), and the other two ddI (didanosine, Videx). There was one case of kidney failure in someone taking ddI plus tenofovir – no longer a recommended combination. So, with less toxic NRTIs, monotherapy and combination therapy would have looked pretty much the same.

“This study supports the long-term efficacy and safety of LPV/r [lopinavir/ritonavir] monotherapy”, Arribas said. He showed a slide that indicated a somewhat higher proportion of patients on monotherapy experienced intermittent low-level HIV viremia in the blood (under 500 copies). For instance, in the first 24 weeks of the study three patients on monotherapy experienced viral load ‘blips’ (none above 200) compared with none on triple therapy. Whether this has implications for longer-term failure and resistance remains to be seen.