Participants
Margaret Johnson – Director of the Royal Free Hospital Centre for HIV Medicine, London
Martin Fisher – Consultant in HIV Medicine,Royal Sussex County Hospital, Brighton
Caroline Sabin – Statistican, Royal Free and University College Medical School, London
Today's topics
- When to start – treatment guidelines revisited
- HIV testing policy
- HIV/hepatitis C co-infection
- New antiretrovirals and strategies
- Therapeutic drug monitoring
- Post-exposure prophylaxis and sexual risk-taking
Abstracts can be viewed at http://www.retroconference.org
Individual abstract numbers link to the abstracts online at http://www.retroconference.org.
Where session titles are underlined, this link takes you to the list of all abstracts presented in that session.
Where presenter names are underlined, this presentation can be viewed online, and the link will take you to the list of sessions which can be viewed online.
When to start – treatment guidelines revisited
KA: What’s your reaction to the news that the Americans are to bring their guidelines into step with what the British HIV Association has been recommending for the last two years in terms of starting treatment?
MJ : We’re pleased, in a way!
[General laughter]
CS: I was just thinking about this the other day. I think it’s just so ironic that things have gone around in a complete circle from when I started [in HIV epidemiology], when 200 CD4 cells was the point at which you started treatment. Then of course in America it went to 500, and now it’s going back down again.
MF: It will be interesting to see how low it goes with the advent of the new BHIVA guidelines.
KA: Do you think it can go lower?
MF: I think it can go lower than 350.There was data in Glasgow suggesting that treatment is only compromised below 200 CD4 cells or viral load above 100,000 copies. There is some supportive data coming up later on at this meeting that suggests the 200 mark may be reasonable, certainly as far as treatment success is concerned. Whether that is a guide as to when to start or not is a different question.
MJ: I think clearly, as the Robert Siciliano presentation today has shown, viral reservoirs are there, even in primary HIV infection. So that, again, makes it logical that what we think was going to happen is the way it has panned out now, isn’t it?. There’s no evidence that you are going to do better in terms of viral reservoirs by starting earlier, to date.
MF: Realistically, a lot of clinicians in the UK and probably the US as well have been more conservative than the guidelines, in their routine practice, anyway. So whether the change in the guidelines will translate into a change in practice is a separate question.
MJ: I think in terms of starting therapy, they’ve a similar problem to what we’ve got in the UK - which is late presentation. There’s a poster up there which looks at the need for earlier diagnosis. Exactly the problem that we have in the UK - which is that patients have medical problems that could be HIV-related and clinicians don’t ask about risk factors for HIV and don’t bring up HIV testing. They’re trying to address that in the same way as we should be in the UK.
HIV testing policy
KA: These were the sort of issues that Kevin de Cock was raising yesterday in his plenary where he was talking about the need for more routinisation of HIV testing both in the UK and in the developing world.
MF: I thought his most important comment was that there had been a confusion of confidentiality and anonymity that had resulted in secrecy. While clearly he was referring to sub-Saharan Africa, I think that rings bells at home as well, in terms of the lack of normalisation of HIV testing. The fact, for example, that all antenatal women are being tested yet gay men attending STI clinics aren’t - it’s illogical.
CS: The antenatal testing issue in the UK is a perfect example of where pediatricians and the Department of Health wanted it to be more normalised to increase the uptake because we were about the only European country that was offering antenatal testing to such a low proportion of women. I guess they have tried to normalise it, in the same way as Kevin de Cock was suggesting.
MJ: I think there is no reason for that not to be the case. Clearly consent is such a big issue in every medical problem. I think none of us should be taking blood and sending people off for tests without telling the patient what you are doing and why you are doing it and getting their permission. I think what we should be doing is trying to have good practice across all tests and that HIV then doesn’t step out of line, and it becomes something that healthcare workers will feel much happier bringing up.
HIV/hepatitis C co-infection
KA: There seems to me that there has been much more discussion of coinfection with HIV and hepatitis C at this meeting already than at previous Retroviruses meetings.
CS: Also the number of people in the hall! [The afternoon session on Co-infections was jammed to such an extent that people were queuing to get in]. People have realised over the last one or two years that coinfection is real problem and they are going to have to deal with it whereas, in the past, we always had problems getting anything published. No one was interested at all.
MF: I think the problem in the UK is that people are dabbling and maybe what we need are regional specialist units in HIV/HCV coinfection. Or to develop some particular expertise in that, rather than just a passing interest.
MJ: And very simple advice, that there is such clear evidence, that patients do so much worse if they have a high alcohol intake [should be given]. I think even to begin with this is very simple advice that every patient should be getting and I think every patient then, as Martin says, needs both the liver input and an HIV input. It really should be on a more regional pattern.
MF: ...Alcohol, hepatitis A, vaccination, all simple things that we should be doing, and I don’t know how well we are doing.
CS: One of the speakers said was that there are an awful lot of studies out there about hepatitis C and natural history. And there are a lot of studies but, generally, there are very few good natural history studies and very few longitudinal studies. Instead, the studies out there are cross sectional studies looking at the effect of coinfection on liver function markers, but actually very few longitudinal studies. And I think that people working in hepatitis C and coinfection are starting to realise now that we need to follow HIV and do some longitudinal studies about what happens to patients infected with hep C and what happens to the long-term coinfected.
KA: So have there been any useful data on coinfection at this conference?
CS: One of the talks today was from the Haemophilia Growth and Development study, in which they showed that every tenfold increase in baseline HIV RNA was associated with a 72% increase in HCV viral load, and that CD4 cell count was inversely correlated with HCV viral load (abstract 35). Almost all of the haemophiliac individuals are coinfected. The difference with the cohort studied here is that they’re all children who have been followed over 15 to 20 years. So they’re probably very different [from other cohorts which have been reported eg (abstract 34]. They were probably infected with HIV and hepatitis C at around the same time. Whereas a lot of the other patients, either the hepatitis C will have come first, and in the adult haemophiliac population the hepatitis C probably came 10 years before HIV. So they’ve been [the adult haemophiliac population] infected with hep C for probably 25 years-30 years, which is quite a different story in terms of the natural history than a group who have been infected with hep C for maybe only 15 years, or in the case of drug users it may even be for shorter periods of time. So I think there are quite big differences between the risk groups in terms of their infection and its natural history.
MJ: The State of the Art lecture on management of co-infection discussed possible strategies for using pegylated interferon with ribavirin but it mainly highlighted the fact there is very little data
CS: I think the reason being that a lot of those trials didn’t include coinfected patients
MF: The nucleoside analogue/ribavirin issue[anatagonism between certain nucleoside analogues and ribavirin] has been on the back burner for several years but it seems to be making no great advance at the moment and it is often used as an excuse not to treat.
MJ: There was some suggestion, though, that it might also augment mitochondrial toxicity so there might be the potential for that toxicity to be increased in our group of patients.
CS: I think that the point that you raised, Martin, is very important in terms of what do you treat first? Do you treat the HIV or do you treat the hep C? That’s crying out for a trial. And given that we’ve only just started to realise that coinfected patients should be treated as well, and that is only just starting to happen [there's a long way to go].
MJ: Speakers in the co-infection session were saying anecdotally that patients with low CD4 counts appeared not to respond or didn’t clear HCV with the alpha interferon and ribavirin and therefore perhaps what you had to do was to treat their HIV and get their CD4 counts up. But then the issue also was - the antivirals might make their hep C worse and give them fibrosis and therefore cirrhosis. I think Caroline’s right that it is very important that we have proper trials to address that question.
CS: I think that the other difference in terms of the treatment is the hep C treatment is a short term thing – 6-months-to-a-year of treatment generally – whereas HIV treatment is ongoing and potentially lifelong treatment. So that may potentially have an impact on which you do first.
MF: There is provocative data on continuous interferon in the absence of clearance of hepatitis C but as a maintenance interferon therapy, so it may be that, for some people, long-term treatment is the right option.
MJ: There was also a discussion about the role of liver transplantation [abstracts 577 and 578]. It was pointed out that very few transplants have actually occurred in patients who are coinfected. Perhaps we’ve got worse outcomes, but again, another very important area and we need to investigate further.
CS: I know that early pre-HAART studies which looked at transplantation in haemophilia patients with coinfection showed that survival was very poor and you wouldn’t expect anything else at that time. Now things have changed so dramatically with the availability of HAART that we need to re-assess whether transplantation is beneficial.
New antiretrovirals and strategies
KA: What about the antiretroviral studies presented today? Have you seen anything today that is likely to change your clinical practice or any compounds on the horizon that you think are particularly interesting?
MF: The interesting compounds were the Dupont compounds, DPC 681 and 683, both protease inhibitors, but clearly they are all at early stages of development [abstract 11]. And the somewhat disappointing data for FTC and the BMS protease makes one less enthusiastic about such preliminary findings. The AI424-007 study showed equivalence between various doses of BMS 232632 and nelfinavir, with around 60% in each arm below 400 copies at 24 weeks [abstract 15].
MJ: The thing about the BMS data was that even in comparison with nelfinavir, the percentage of patients with a viral load response below 50 copies was very low, around 30-35% at 24 weeks, and I don’t quite know why but it was....
MF: Surprisingly poor.
MJ: ....surprisingly low even given we’ve seen nelfinavir shown to be less effective lately in a head to head study with Kaletra.
KA: And in the Combine study comparing nelfinavir and nevirapine as well.
MJ: Yes. That sort of surprised me about the BMS 232632 study.
MF: And it was three times daily nelfinavir dosing versus once daily BMS 232632 dosing. And clearly, therefore, once daily dosing may not be as a good as we all want it to be and I think there’s a danger that we all want it to be simple when perhaps it ain’t going to be, and similarly with the FTC simplification data – it’s not the finding everybody expected to see [abstract 8].
Editorial note: Data from two randomised studies of 150 mg 3TC BID compared with 200 mg FTC once daily in triple therapy combination regimens were reported.
In study FTC-303, 440 individuals with viral load below 400 copies/mL on a 3TC-containing regimen were randomized to switch to FTC or continue on 3TC. No differences in the rate of viral rebound was observed between the two arms up to week 48 (8% in each arm).
However, a second study, FTC-302, showed more disappointing results. This study was a randomized, double blind comparison of FTC and 3TC in 468 treatment naïve patients also receiving d4T and either nevirapine (if viral load below 100,000 copies/mL) or efavirenz (if viral load above 100,000 copies/mL). Viral rebound after initial suppression below 400 copies occurred in 12% of the FTC recipients compared to 6% of 3TC recipients (p
MF: The result was partly driven by the toxicity but overall... it was a bit of a hotch potch of two different studies actually looking at quite different things. One was a naive study comparing FTC and 3TC and the other was a 3TC to FTC switch. And in the switch study, like most of the switch studies, there was an increased drop out rate in the switch arm due to new toxicity. So again, moving to a once daily dosing regimen may not be the right thing to do in somebody who is otherwise well.
MJ: And the liver toxicity in the FTC study, although it was all put down to nevirapine, clearly that was also concerning [abstract 19].
MF: Yes, the actual rate of hepatotoxicity was similar to the bulk of other nevirapine studies with the exception of the 1090 study (which always stands out in numerous analyses looking at different questions - one wonders how neutral that data is). Again, the explanation was put down to the fact that it was largely a population of black women - both of which [geneder and ethnicity] are known to be risk factors for nevirapine toxicity. Where I don’t entirely follow the logic is the recommendation for very intensive monitoring of liver function tests which covers everybody’s backs but it doesn’t seem to predict who is going develop severe hepatotoxicity and who won’t.
MJ: And then there was the switch study for patients undetectable on a protease inhibitor, randomised to either continue their protease inhibitor or switching to efavirenz and showing better outcomes for those switched to efavirenz in terms of ongoing viral control [abstract 20].
KA: And that was 48 week follow-up.
MJ: Yes it was. So, I think, again, that that’s important. And most of that was actually down to adherence and raises the question of continuing to think about adherence with every patient even when they’re undetectable. It really is important to say ‘look there are other things we can do if adherence is a problem, perhaps we think about a simpler regimen’, in order to try to avoid failure. Do you feel that, Martin?
MF: Yes, obviously with the switch studies what one is looking for is better or at least equal virological suppression without a big drop out due to toxicity and the efavirenz study appeared to achieve both. So it was notable that there were 25% with CNS toxicity. And if we believe the Australian data, then maybe 48 weeks isn’t the right time to be looking at the long-term significance of that.
MJ: I am always very sceptical about CNS toxicity with efavirenz and I don’t think any study properly defines what CNS toxicity is. I mean, how do you measure depression, are they using validated methodologies? It’s not my clinical experience that 25% of patients have significant CNS toxicities that I need to do anything about. Do you think that’s the case Martin?
MF: I agree with you that there is no validated measure. It’s disappointing we have come this far and we are using so much of the drug without really knowing what we are all talking about.
MJ: Absolutely.
MF: That being said, it’s rather coincidental that so many patients report remarkably similar complaints for it to be nothing real. That being said, most people seem to be able to put up with it, at least in the short to medium term.
MJ: There was also T1249, another entry inhibitor from Trimeris/Roche in phase II studies in humans - fourteen day data. It appears to be a once daily dose and seems to have less of an injection site reaction than T20, and is active against T-20 resistant virus in the test tube. About 1.2 log viral load reduction at 14 days [abstract 14].
MF: And the possible synergy with other entry inhibitors looks very exciting.
MJ: And that was backed up also by our State of the Art lecture by Robert Doms of the University of Pennsylvania. He talked about CCR4 and CCR5 antagonists, entry inhibitors and other strategies for preventing HIV entry and fusion, and the way in which a CCR5 inhibitor may improve the ability of entry inhibitors like T-20 to inhibit gp41 by reducing the amount of CCR5 expression on cells and thus increasing the amount of time that HIV must hang around, exposing gp41 regions while it looks for extra CD4 receptors to bind to, in the absence of CCR5 co-receptors. Several combinations are already being looked at and this is clearly a very promising area.
CS: Were you interested in cyclophosphamide? [a study which looked at the utility of the cancer drug in purging reservoirs of latently infected cells][abstract 16].
MF: It was a disproof of concept study. It fits with previous interleukin-2 data. It looks like you can’t purge latent reservoirs – they’re here to stay and we’ve got to live with it and adapt. It’s so important that people present negative results. They are so often hidden away at the back of the hall.
MJ: - and presented in a positive light. And I thought that was quite refreshing to see that.
Therapeutic drug monitoring
KA: The last session on therapeutic drug monitoring – particularly a prospective study - was provocative.
MF: The Pharmadapt study [abstract 260b) took a rather unusual slant on TDM in that it took patients going into what seemed like a second or third line therapeutic regimen. Patients were randomized to receive genotypic resistance tests or therapeutic drug monitoring and a genotypic test at week 4, and adapted their proteases on the back of that at week 8. They found no difference in viral load outcome at week 12 - 2.51 log in the TDM arm versus 2.3 log in the genotyping arm. Which really begs the question – does that suggest that TDM doesn’t work or is it that they are doing it at the wrong point in time and adapting too late? It seems to me that they were shutting the stable door after the horse had bolted in this study.
MJ: So they were doing that to improve efficacy, not to manage toxicity?
MF: It was purely to increase doses to get above the IC 50 compared to wild type. Which again seems a rather unusual target for patients who presumably don’t have wild type virus anyway. Another problem with the study is, I think, that around 60% of the patients were already on ritonavir boosted proteases, so it may be better to analyse the non-ritonavir patients to prove the concept. But I think most of us would be using ritonavir enhancement in that patient group anyway. So maybe it’s telling us that TDM won’t fulfill what we want it to and may be better situated to look at toxicity rather than efficacy.
Post-exposure prophylaxis and sexual risk-taking
MF: There were some interesting posters on behaviours, and I know it comes up again later during the meeting. One was on post-sexual exposure prophylaxis; one poster presentation which basically seemed to suggest complete anarchy over who gets it, why they get it, etc [abstract 226].
KA: - that was the French example -
CS: - Few actually follow it through -
MF: - Very few finish, very few have follow up,very few have subsequent antibody testing. And there appeared, across several posters, to have been no seroconversions during post exposure prophylaxis but...
MJ: - And it was a million dollars per QALY![Quality Adjusted Life Year]-
MF: ....no coherence about service provision at all. Which makes one worry about the UK where it seems to be rather an unspoken subject and one wonders to what extent the same anarchy is occurring but hasn’t been publicised.
MF: And there were a few posters again raising the issue of whether people’s sexual practices are changing as a response to different perceptions of HIV severity [abstracts 211-219]. I think the work here today seems to suggest that some people are relaxing their sexual behaviour. Whether that will all change as treatment guidelines become more conservative and everybody becomes more pessimistic about treatments again, I think we will have to wait for this time next year.
CS: I think there are a lot of large studies now showing very worrying trends about sexual behaviour, increases in other STDs, which might potentially give us increases in HIV over the next few years.
MJ: There was a San Francisco poster[abstract 224], they were looking at post sexual exposure prophylaxis and showing that they seemed to think it had no effect in terms of increasing unsafe sex. They thought it was a good opportunity, even if it wasn’t entirely followed through, to get through safer sex messages.