July 15, 2013
This past March, the 20th Conference on Retroviruses and Opportunistic Infections (CROI) brought an unprecedented flurry of mainstream interest in potential HIV cure strategies. TheBodyPRO.com's Nelson Vergel sat down with leading HIV cure research activist Richard Jefferys for an update on current important aspects, and controversies, in HIV cure research. Jefferys, who coordinates the Basic Science Vaccines and Prevention Project at Treatment Action Group in New York City, is at the forefront not only of gathering data and information from different studies, but also of educating other activists on the new language of immune-based therapies. Read Part Two of this conversation.
Richard, we just attended the 2013 Conference on Retroviruses and Opportunistic Infections (CROI) in Atlanta in March. Could you let us know a little bit about what's been going on with cure research? With the reports from the conference about the baby that was cured, which got a lot of media exposure, people are a little confused, so let's start with that and then expand into the different cure studies presented at the conference.
That was obviously the story that dominated the meeting, although there were other interesting presentations.
I think it was a confusing case. The media coverage has not been all that easy to follow and it's also unclear whether the case has relevance to the adult situation or not. I think they're hoping to shed more light on that in the future. But right now the only thing that they seem to be able to say clearly is that there was good evidence that the infant was infected in utero, so before birth, and received very early treatment with antiretroviral therapy. Then, when there was an unscheduled stop of therapy around 18 months later, there was no rebound. Then the child and the mother were lost to care for several months. When they returned for care, there was no detectable HIV viral load, despite the fact that the child had been off treatment.
That's what led the clinician to involve other scientists in the case who looked for evidence of virus in the infant's body and could only find some bits and pieces of possible DNA, but no replication-competent virus, no active virus and no immune responses. Perhaps the very early institution of therapy, rather than a preventive regimen, prevented the formation of a reservoir of latently infected cells and that meant that there was no rebound in virus when treatment was stopped.
Some people say that other babies have cleared the virus in the past due to the natural tendencies for a baby's immune system to basically regenerate itself after birth. Can you explain this? I think you wrote a piece in your blog about this issue.
There were isolated case reports in the scientific literature during the 1990s of infants who appeared to be infected with HIV based on PCR testing for HIV DNA, but later showed no evidence of infection. It was proposed that they may have cleared the infection on their own.
What happened, though, is that there was a research group who investigated many of those cases looking at the genetic sequences of the virus to see if it was connected to the mother's virus. In all the cases they looked at, there seemed to be other explanations for what happened -- including cases of sample mix-ups where the sample wasn't actually from the child, and apparent false-positive PCR readings.
Although not every case has been formally ruled out, the paper that was published on this observation was in Science in 1998 with Lisa Frenkel, M.D., as lead researcher. They set out some criteria for proving transient infection in infants and so far, as far as I know, there haven't been any reports since then. So, although it's theoretically possible that this child was going to clear the infection on her own, the evidence for that happening in the past is not convincing and so this does seem to be an exceptional case.
Another study that has been mentioned for a few years and is getting new attention is the VISCONTI study. Can you explain this study to people who are not familiar with it? I understand that it included some patients who started HIV treatment early in their infection who later stopped treatment. After they stopped HIV medications, some of these people seemed to continue to have very low or undetectable viral loads for years.
The term "post-treatment controllers" is being used more often because of this study and some other individual case reports. The interesting thing about the VISCONTI study is that it does seem more directly relevant to adults with HIV infection.
Essentially, they've identified 14 people that were treated early and stayed on treatment for quite some time -- I think around three years on average -- and they have been able to maintain extremely low levels of virus after stopping treatment, in some cases up to 10 years now off treatment, which is obviously very exciting.
The terminology that's being used in cure research now has become a little bit confusing. Researchers are reluctant to refer to this group necessarily as being functionally cured because they're not sure how long the effect of treating early is going to last, so they are cautiously calling this a type of "drug-free remission."
I think what it really calls for is a more systematic effort at a search for cases like these, where people have stopped treatment for some reason and been able to maintain control. Because although the VISCONTI cohort represents probably the largest number of people with the longest-term follow-up, there have been a number of case reports over the years since that have some similarities. The most famous probably is the original "Berlin patient." It was not Timothy Brown, but a different patient in Berlin in the late '90s, who was treated for acute infection and maintained viral load less than 50 copies off treatment for many years.
There are a few conference presentations and case reports of similar individuals. Giuseppe Pantaleo, M.D., and Jan van Lunzen, M.D., have each presented a case. There's just a lack of any kind of systematic analysis and I think it would be productive for researchers to try to come together to assess how common this is and to try and do a more comprehensive job of identifying factors that may predict somebody's ability to safely go off treatment.
[Editor's note: Dr. van Lunzen presented data at IAS 2013 on a functional cure for a German patient who had started treatment within a month after being infected.]
The interesting thing about this study was the fact that they could not predict or find out what made those people special. Why were these 14 people able to control their virus post interruption while many others who started treatment early did not? How do these "post-treatment controllers" differ from elite controllers, who have never taken HIV medications but remain with very low viral loads for years and years without any antiretrovirals?
One of the differences is that most elite controllers have strong immune responses to HIV that are quite easy to detect, particularly CD8 cell responses that are responsible for identifying and killing virus-infected cells. Those responses seem to be important in a lot of elite controllers. One of the interesting differences with the VISCONTI cohort members is that, in general, they don't seem to have particularly strong responses.
Also, there are immune response genes that you inherit, and some of them seem to be particularly good in terms of allowing the immune system to recognize and respond to HIV. Elite controllers often have these kind of good immune response genes and I think only three of the 14 VISCONTI cohort members have these kind of genes. In fact, quite a few of them had immune response genes that in untreated infection have been associated with faster disease progression rather than slower. If there's something to that, in terms of an interaction between their immune response genes and the long-term treatment that they were on, it's not clear yet, but it's a line of inquiry that the researchers are pursuing.
The other missing piece of the puzzle is that one of the immune responses that seems key in elite controllers is not the CD8 response, but the CD4+ T cell (T-cell helper) response targeting HIV. There's been no information presented on CD4+ T-cell responses to HIV in the VISCONTI cohort members. So whether there is something different going on there is not known yet.
Do we have data that shows how fast HIV populates the different reservoirs of the body after infection? What would be the optimum or longest time to wait to start treatment to attempt to be one of these post-treatment controllers? When is the point of no return when we have so much virus in different reservoirs that we would be unable to control virus replication after treatment interruption?
That's a really good question. There was a study presented at CROI by a Thai researcher, Jintanat Ananworanich, M.D., which involves probably one of the largest cohorts of people treated early that has ever been recruited. They break down stages of acute HIV infection into these different categories called Fiebig categories based on research on defining different stages of infection. Ananworanich's study has a lot of people in Fiebig 1, which is where there's virus detectable, but no beginnings of an antibody response, and those individuals are being treated. There are people in Fiebig 1 through 4, so various time points after acute infection, all being treated with antiretroviral therapy. There are plans to do an interruption in therapy in that cohort. So that really should help answer the question if there's a point early enough that will predict prolonged control because the formation of the reservoir has been inhibited.
That's probably also the closest adult parallel to the infant case in that people in Fiebig stage 1 are very, very soon after infection. It might be an opportunity to see whether the formation of a significant reservoir can be prevented in adults.
In terms of animal studies, there were some conducted by a researcher named Jeff Lifson in the '90s who administered tenofovir [Viread] to macaques very shortly after a challenge with a pathogenic SIV -- the simian cousin of HIV. If the drug was administered within 24 hours after the challenge, most of those animals showed prolonged control or even an absence of signs of infection after drug withdrawal; so there may be reason to think that very, very early treatment can lead to prolonged control if treatment is later interrupted.
I am very concerned about the fact that if someone presumes that they have been infected with HIV it is very difficult for them to get early treatment after they get tested. Personally, if I was to get infected right now with HIV, after reading all this information, I would be really compelled to not wait too long and to not allow my reservoirs to be populated by HIV. The lucky ones who accomplish this may have a better chance to be cured when the day comes.
Regarding the effects of early treatment on preventing HIV's spread into body compartments, I'm also following some of the work done in measuring how well some HIV antiretrovirals penetrate the different reservoirs. The VISCONTI cohort was exposed to older HIV drugs that may not penetrate as well, so I wonder if we would have seen more post-treatment controllers if they were provided drugs in the integrase class, which seems to penetrate well and quickly into reservoirs. Could you explain what we're seeing so far?
So, one of the big controversies in this research is the extent to which ongoing viral replication may contribute to virus persistence despite treatment. There is some evidence that there can be ongoing replication in different tissues where drugs are not penetrating so well. Probably the best evidence is that protease inhibitors may not be penetrating into the lymph tissue and into the gastrointestinal [GI] tract as well as other drugs. Recent studies are suggesting that the integrase inhibitor raltegravir [Isentress] penetrates very well into the GI tract. There was a study presented by Hiroyu Hatano, M.D., from UCSF, at CROI that added raltegravir to antiretroviral therapy in an intensification strategy. She did report some evidence that people that received that strategy showed additional declines in markers of inflammation and that viral replication was being inhibited. Similar to some other studies, it seemed that participants on a protease inhibitor-containing regimen had this additional suppression with raltegravir.
There is a large, ongoing study looking at different drugs and tissues being conducted at the University of Minnesota. It involves several different research teams: Timothy Schacker, M.D., looking at the tissues; Courtney Fletcher, Pharm.D., looking at drug levels; and Mario Stevenson, Ph.D., looking at viral measures. Some preliminary results from that ongoing study have been presented, but I think probably the full publication of the results will shed significantly more light on that issue.
Going back to VISCONTI's 14 patients that were shown to control virus after they had stopped treatment: I was very surprised to see that some of those patients only took nucleosides or even dual therapy and got to that point of control after interruption.
It really struck me that maybe the type of treatment that they start at the beginning is not as important as starting treatment of some kind. I'm interested in further research that's hopefully going to be available soon, so that we can have a better understanding of what treatments to start with to treat early infection, and things that can actually change the current guidelines. They don't really specify what type of treatment to start with, just basically any of the first-line therapies that are available right now. That's something that I think is very interesting. Obviously, people that are reading and have friends or family that may unfortunately find out that they have been exposed to HIV will at least have a higher awareness of the fact that it's very important to start treatment early, even if they don't really know their CD4+ and viral load numbers.
Are there any other updates from CROI that I haven't mentioned when it comes to reservoir testing? I think there are some other indications that current testing is underestimating HIV reservoir size. Can you explain the data presented at CROI on testing of the reservoirs?
Because in most people the reservoir is relatively small, there's a needle-in-the-haystack aspect of trying to find and quantify how much of a reservoir is there, how much of it is actually made up of virus that can replicate and might rebound when treatment is stopped. It's technically just a very challenging thing to do.
I think one of the encouraging things is that oftentimes in research it's very hard to get different groups to collaborate with each other. In terms of measuring the reservoir, there has been a big collaboration that's seen multiple laboratories with different kinds of tests compare their different tests to see how well they perform in comparison with each other and whether they correlate with each other. Labs would all like to believe that their test is great, and so there is considerable compromise involved in engaging in that collaborative process, which they've all done.
The results are not great, in that each test may be showing something useful, but they don't seem to link very well together. HIV DNA is one thing that can be measured, but a lot of the HIV DNA, it turns out, is kind of dead-end defective virus that can't really replicate; so it's not clear whether you really need to measure all of it.
The gold standard is trying to do what they call viral outgrowth, or measuring how much virus can actually replicate in a sample. It involves taking quite a lot of cells from somebody and it's quite difficult to do, but that still seems to be the best guide to how much virus is there.
The other kinds of tests don't seem to correlate too well, although there is one test developed by Una O'Doherty, M.D., at the University of Pennsylvania, that looks for HIV DNA that's integrated into the genome of the cell. And that does seem to correlate better with how much replication-competent virus can be detected. I think there's work going on now to try to streamline those tests so that you can get a clearer and more reliable sense of what impact an intervention is having on the size of the reservoir, because that's obviously crucial in terms of deciding whether you want to look at things like treatment interruption.
Bob Siliciano, M.D., who has really pioneered a lot of the latency research, did present a new model suggesting that, if you can push down the levels of the reservoir by several logs, it would predict at least a delay, possibly of several years, in the rebound of virus. That's obviously something that needs to be further confirmed as interventions develop which reduce the reservoir.
This transcript has been lightly edited for clarity.
Read TheBodyPRO.com's multi-author blog, HIV Care Today.