Athens Roundtable: Views on the 8th ECCATH meeting from BHIVA members

During last week's Eighth European Conference on Clinical Aspects and Treatment of HIV Infection, Drs Caroline Sabin (Department of Primary Care & Population Sciences, Royal Free & University College Medical School), Mike Youle (Royal Free Hospital) and Martin Fisher (Royal Sussex County Hospital, Brighton) talked to Anna Poppa for aidsmap.com about the highlights of the meeting. This is an edited transcript of that interview, which took place on Tuesday 30th October.

Caroline: I had a look yesterday at the posters and one of the trends I noticed coming out was switching from PI HAART, usually to efavirenz, sometimes to abacavir-containing HAART, and generally the results tended to be positive. There was one study where the virological response was the same but the CD4 response was actually better in the group the remained on PIs, but there was only one study saying that.

Mike: My issue with those studies is that I think that just about every study has shown some benefit from switching as opposed to not switching, but what it doesn't say is whether you switch and then stay on that regimen, or switch back at some point. I would be interested in triple switching, and single class switching, i.e. to take all non-nukes and then all proteases and then go back to square one and keep it going.

Caroline: One of the questions I had was - I mean obviously some of it's about switching to a combination which is easier - but is it also something to do with the process of switching, so you perhaps re-generate some interest in tablets, and so on; something new to think about?

Martin: Absolutely. Certainly most of the data from the efavirenz switch studies suggest that a lot of it is driven by adherence rather than any difference in virological control, but what all of the switch studies continue to show, and again the BEST study [where participants either remained on their three times daily indinavir-containing HAART or switched to twice daily indinavir/ritonavir] showed it here, is that when changing in someone who's otherwise been well, you're going to run the risk of toxicity, which really isn't what we want.

Mike: Yes, but there again perhaps with the non-nukes maybe the toxicity which has been reported before is less in people who are already suppressed. I think that's certainly true of nevirapine - the rash is less. So you may be giving the best case scenario to the new drug.

Martin: Andrew Phillips presented some data from the Royal Free and Frankfurt cohorts which looked at people on their first HAART regimen and their time to virological failure, and showed that it was related very significantly to previous nucleoside analogue experience - but that you really didn't need very much experience at all; it could be less than two months and that significantly affected your subsequent outcome. Now, clearly for people starting therapy now that's not of a great deal of relevance apart from in the case of pregnancy, and in some cases PEP, where current guidelines are still sometimes recommending monotherapy or dual therapy. It looks like you might be setting that person up for massive difficulties in the future.

Caroline: And all of these people had gone undetectable at some stage, everyone had had an initial response so it looked very promising, but the differences in subsequent rebound rates were actually quite big. I know he took account of all the other factors such as the calendar year, viral load and CD4 at baseline, but they were very big differences. It's almost as if, if you take a small dose of treatment at some stage, then really your future treatment options are very limited.

Mike: I thought VEGA was fabulous. VEGA was the first controlled study that shows that injections of New Fill improved people's facial appearance, improved people's facial structure, and improved their quality of life by visual analogue score. So that's a very positive short-term message. The question is what the long-term data will be, though there are cohort data from the same unit which go out to almost three years now, and of course the technique has been used in non-HIV patients for a period of time.

Anna: What are you providing at the Royal Free?

Mike: At the Royal Free we can't do anything because we can't find a plastic surgeon who'll do it. Apparently at St Mary's they're going to be offering it on the NHS in a limited way. Chelsea and Westminster have got no provision at the moment but clearly after this I think they might.

Anna: Haven't they got a trial?

Mike: No, that's recruited.

Martin: We have a plastic surgeon who has done the procedure already, and we're putting in a bid to our health authority in an attempt to provide it on the NHS. At the least we would hope that if the individual did have to pay, they would only pay for the agent.

Anna: How many people do you think you'd be talking about?

Martin: That's clearly a moot question. We've been running with a figure of about 20% of people on therapy would have severe lipodystrophy at two years, and that they would need between three and five injections, with a top-up two years later.

Mike: What was interesting was that [the VEGA study's author] was asked if people had had a deterioration and she said over the first nine months they were not seeing deterioration; that they'd expected to do more top-ups. I think the other thing is that of the people who have this problem, clearly a proportion will not necessarily want this procedure. I think that's important. I've got someone with very severe lipodystrophy and I said "Well aren't you thinking about…" and he said "No, I actually quite like the way I look".

Martin: To return to your point about provision across the UK, I think there's a danger that if certain units are set up and others aren't we'll be back where we were two or three years ago with people having to move treatment centres to get the latest service.

Mike: Well that's why I thought we should try to get a meeting with all the providers, plastics, dermatologists, to come up with some kind of guidelines for what we might do. I'm not sure how easy that will be though.

Martin: For the cost of £400, if you can maintain someone on their therapy who would otherwise have discontinued, which the data in Buenos Aires suggested, then logically that should be beneficial. Unfortunately the NHS doesn't always work that way, and to complicate matters it's classed as a medical device rather than a drug, and that means the whole approval process is different to the one we're all familiar with. It's a completely different purchasing system, which means you can't get it through pharmacy, which is what we're used to with combination therapy.

Caroline: With an interest in HCV co-infection, there were a couple of things which I though were noteworthy. There was one study which showed that in people who were co-infected, if you treated the hep C for six months, you reduced the discontinuation rate when people went onto HAART quite dramatically - discontinuations due to hepatotoxicity. It was reduced from something like 54% discontinuations in the control group to 13%, 16%, something like that, so a big difference. The interesting thing about that was that it actually didn't really matter what the hepatitis response was, so even in patients who'd had an initial hep C response but it wasn't sustained, they still seemed to get the benefit when they then went on to start HAART, so I thought that was interesting.

There were then a couple of studies looking at the CD4 response when people start HAART and comparing the response in people who were co-infected with hep C and people who were just infected with HIV alone, and very similar to the Swiss Cohort Study, patients who are co-infected do seem to have this delayed immunological response, for whatever reason.

Martin: I think part of the worrying thing now is that the data on the success of interferon and ribavirin seems to be suggesting that you need to have a reasonably good CD4 count to get a response. So given that most people in the UK are still presenting relatively late, you're in a bit of a double whammy. If your hep C treatment won't work and your HIV treatment won't get your CD4 count up enough for your hep C treatment to work, even if you can tolerate your HIV treatment… So, we need to see more a few more data on what you can do for those people who are currently below the hep C treatment threshold.

Mike: GIGHAART I think is important for two reasons. One is that it's the first controlled study of treatment interruptions [versus no interruption in people with heavy drug experience, about to start a multi-drug regimen]. We've had the same intimation from the Royal Free data and from Veronica Miller's data, that having a treatment interruption was virologically beneficial for you in the short-term.

In GIGHAART, 68 patients were followed, and these were very late stage patients, I mean 26 CD4s versus 28 and they had a high-ish viral load. They'd all been multiply experienced, and quite clearly, on whatever analysis you took, having an eight week interruption was virologically beneficial. The viral load drop was 0.4 log drop versus 1.9 log drop. It's kind of difficult to argue against these data. The question is, is that data going to be the same out to 24 weeks, and clearly Christine [Katlama, the study investigator] intimated that it would.

The next thing I'd say is what are the implications of the data for the OPTIMA study? I kind of think either the OPTIMA study has to ignore those data, which I think is difficult because the study was not stopped by the investigators, it was stopped by the DSMB [Data and Safety Monitoring Board] which means that it was considered unethical to continue because of the high benefit; or the OPTIMA study has to say well, eight weeks has been shown to be beneficial, perhaps we should randomise people to do an eight week interruption and then go back on therapy, or do as long an interruption as you wanted. I certainly think that needs to be considered by the OPTIMA steering committee fairly quickly because it's very early in its recruitment phase. One difficulty may be a difference in the UK opinion and the US and Canada's opinion because I think they're very different treating sites.

In terms of where we move with GIGHAART itself, there was a clear issue about what the sensitivity of the virus was under the genotypic testing, and I think one of the questions becomes, and that is being answered in OPTIMA, do you need all those drugs once you go back on therapy?

The final thing that I think links that with the 3D study [which investigated the use of ddI/d4T/efavirenz with or without hydroyurea], is that certainly in the experienced patients in the 3D final 48 week data from Rob Murphy, you had a benefit to hydroxyurea over the naïve patients, and that may be related to an immune activation issue. In GIGHAART seventy-odd per cent had hydroxyurea, and in our salvage data in our cohort at the Royal Free we had a number of people on hydroxyurea, some of whom stopped, blipped, and went back on hydroxyurea and re-suppressed. Now, I would like to look at those data in a much more controlled way because I think the role of hydroxyurea is not yet adequately explained, especially in late stage patients.

Anna: A lot of people dropped hydroxyurea in the GIGHAART study.

Mike: But that may be a good thing, and maybe you only need it for a short period of time. The problem with the 3D study, which really crippled that study is that firstly, people didn't start it immediately, and secondly, it was a continuous dose. I think that quite clearly after a long period of time on hydroxyurea you do get toxicity.

Martin: Another thing that was quite interesting with GIGHAART was that the rationale for performing the study was that people would revert to wild-type virus and that was why they would do well, whereas in actual fact it seems to be the treatment break itself rather than just the switch. Although the people who reverted to wild-type did slightly better, those who had no change in their resistance pattern also did much better than the people who didn't have an interruption. So clearly we don't understand what's going on.

Martin: I think tenofovir is worth a mention. Clearly quite impressive data both in terms of efficacy and safety, though the data are really only to one year, so some concerns about bone mineral density remain, but it looks like the kidneys are given the all clear with tenofovir as opposed to adefovir, and that it appears to be a very safe drug. What I think is going to be interesting is to see what niche it fits and to see whether clinicians in the UK will be recommending it up front or further down.

Mike: The other thing that I thought was important treatment wise was MaxCMin, which is basically saquinavir/ritonavir versus indinavir/ritonavir, and I think somewhat to people's surprise there wasn't a huge difference between the two groups except in the toxicity side of it. There was more toxicity - although it remains to be seen whether it was statistically significant - in the indinavir group. In terms of efficacy it looks like there was good efficacy in both arms. So I think that's important.

Martin: I think the other thing to factor in, in relation to the dose of indinavir was that the BEST study showed that 800/100 is probably the wrong dose to be comparing.

Mike: Merck are a bit slow but I think they are going to do a dose-controlled study where you can change to whatever dose you want to based on [drug level monitoring], which I think is the right way to do it.

Anna: Tell me what you think about atazanavir.

Mike: Atazanavir is quite a difficult issue because they do seem to have reached a level of tolerability in that they can't push the dose up much more. The head-to-head study with efavirenz which is ongoing will be interesting, but clearly the data that they've presented so far shows that it's as efficacious as nelfinavir. Of course nelfinavir has quite a checkered history, but it's actually quite an effective drug. So once again I think you're in the setting of the drug works if you can take it. What would be nice is a bit more data about people who didn't do well on it, and clearly more studies. What's interesting about it is that it's once a day and this lipid issue, which appears to be much much better, so I think that even if on a head-to-head basis it's not stronger than ritonavir-boosted PIs, it may be part of a "lipid-sparing" - that's a new concept - regimen which includes a protease inhibitor.

What's interesting of course with BMS joining with Du Pont is that 3D is going to be a pivotal study for them. There was good suppression data from ddI/d4T/efavirenz. What they quite do with atazanavir at that point… Julio Montaner presented the sustained release d4T, and it was exactly the same [as twice daily] so BMS are quids in with once a day therapy.

Anna: The virological data on atazanavir isn't there though, is it?

Martin: Well it's comparable with nelfinavir.

Mike: I have to say that nelfinavir is not bad. The vast majority of people stay well on nelfinavir.

Martin: And it's still the most widely prescribed protease inhibitor in the world.

Mike: Exactly, and one has to ask why that is. It's because actually it's pretty good, and the durability is good - I mean we haven't got durability data on lopinavir yet.

Anna: Andrew Philips also had some interesting data on the frequency of CD4 counts at higher CD4s.

Mike: Basically what Andrew did [with data on patients at the Royal Free Hospital] was look at all the patients with a CD4 count over 500, and say, if you have an undetectable viral load did your CD4 ever go below 350, and the answer was no. There were a couple that went down to, I think, 350 temporarily but their percentages didn't drop. The benefit of that is that you don't have to do lots of blood tests, but you still have to do the viral load so it doesn't make any difference in terms of patient attendance. The danger, I feel, of presenting this data too widely in front of funders is that they'll say "Oh well we don't have to pay for that".

Having said that, there's going to be a very interesting late breaker tomorrow which is Ilesh Jani's data, where they've looked at doing cheap CD4s using red diode lasers versus [commonly used] blue diode lasers, and the correlation is 99.9%. So the issue is that you don't need to spend half a million on a machine and £25 on a test, and that you can actually do CD4 counts for about £2.50. So we may be getting few things, diagnostics wise, which may be much cheaper in the future.

Martin: Of course it would be nice to say that you take that saving and spend it on New Fill, but unfortunately things don't work like that.

Anna: Though these things obviously have implications for the developing world.