Palm Project Interview Series: Steven Deeks
February 8, 2007
Steven G. Deeks, MD is a highly respected clinician and scientist whose work encompasses both clinical care and research into the pathogenesis of HIV infection. Deeks became widely known among treatment activists for his work on individuals with multi-drug resistant HIV who remain clinically and immunologically healthy despite the fact that their antiretroviral therapies fail to fully control HIV replication. Deeks is now involved in many different projects and collaborations but maintains a particular focus on translational research, which aims to translate advances in basic science into clinically relevant therapies and treatment strategies.
Q: What are your research priorities right now?
In the last ten years, I've focused in on trying to understand the pathogenesis of drug resistance; focusing on the virus host interactions, and how reductions in viral fitness can lead to improved immunologic outcomes. We are still working on these issues, focusing on antigen-specific T cell responses, HIV-specific T cell immunity; T cell activation; T cell turnover; T cell trafficking, and so forth.
But in the past few years, it's become quite clear to us that we really can't fully understand how drug resistance and viral fitness impact outcome without a complete understanding of the normal response to HIV; and since there is no clear consensus on the immunologic correlates of virus control, and as there are no really validated standard assays to measure this stuff, we -- in the past couple years -- have become very interested in the study of untreated people.
And in particular -- and this really has led to a complete shift in our approach -- we became very interested in a subset of individuals who we think are essentially completely controlling the virus due to the immune response. About two years ago, we began to aggressively recruit these individuals who have no measurable virus in the absence of therapy -- these so-called elite controllers. And so while we're all still focusing on drug-resistance and viral fitness, we're devoting more and more time to characterizing these elite controllers, and we're doing this in collaboration with a number of investigators across the country.
Ultimately, what we'd like to do -- and what we're starting to do -- is to use lessons learned from this type of work to perform immune-based therapeutic studies. Our ultimate goals with those studies are to, one, confirm what we understand about pathogenesis; and two, to come up with novel strategies for individuals who really cannot respond or don't want to go on therapy.
Is the NIH making it easier to conduct clinical research?
This is a complex question. From one perspective, the answer is yes, as nearly all of our funding comes from the NIH, and we as a group are currently doing ok. However, most of our funding is either indirect and is used to support our cohort infrastructure or the salaries of faculty and research fellows. This funding allows us to support out laboratory-based collaborators by providing them access to unique and well-characterized patient specimens, and allows us to perform some of our observational work.
What we prefer to do is to use this established infrastructure to perform focused, pathogenesis oriented clinical trials. This type of research-clinical trials has always been hard to fund via the NIH, but has become particularly challenging as the NIAID will no longer fund clinical research via the traditional R21/RO1 series. Rather, we and others have to either work with the established clinical trials networks or apply for funding via a complex three step process that involves a letter of intent, an R34 planning grant and eventually a U01 supported clinical trial. Each of these steps has its own payline and each takes a long-time. I figure that if everything goes well, and if we are approved at each of the three steps, it will take at least three years and probably longer to go from an initial submission to when we can begin our study.
Some have suggested that we should simply do our studies in the established networks. This would certainly make DIADS and the NIAID happier since there is an established mechanism for regulatory oversight. But from my perspective the networks are not really well designed to do this kind of really focused pathogenesis work, particularly as it pertains to studies regarding the immune system. Such studies require access to measurements not readily available in the core laboratories supported by the network. Also, the networks tend to very conservative in the types of work they are doing. And finally, the networks now have the own funding limitations, and seem to be focusing more on the larger multi-center studies, which makes sense.
So we're getting squeezed. There's a pathway that you can use for doing large clinical trials and networks, and there certainly is a pathway you can pursue for doing laboratory-based work. But that in between studies -- small, focused, intensive, very expensive translational studies -- are getting harder and harder to fund.
Would it be useful, do you think, for the Division AIDS to think about trying to deal with that by having somebody to work on those kind of regulatory issues for investigator-initiated projects?
No, I think that if NIAID feels that they have to do all the regulatory stuff for all of the clinical trials, then they likely don't have the resources. I think they're going to have to go back to a certain degree to the way they used to, where some of the regulatory stuff can be taken care of by the funded investigators. Still, And I don't really understand the issues here. I do not understand them. I appreciate why the NIAID is doing what they are doing, however. In the past, there have been non-clinical investigators who obtained funding to do clinical research, but who clearly lacked the experience and infrastructure to perform such work. These studies often failed to enroll or underperformed for a variety of reasons. I do believe that investigators new to clinical research should be required to first establish an infrastructure the R34 mechanism. My hope is that for those groups who have a track record, this time consuming step can be avoided.
Are you being hit here also with the issue of the diminishing pay line of investigator-initiated grants?
I believe the diminishing pay-line is an even greater problem for those of us doing clinical research, since we have to compete for funding at both the R34 and then eventually at UO1 levels. So we are hit at least twice.
So we're looking elsewhere, and we've actually been successful with amfAR, which I think has actually stepped up to fill some of this gap; and hopefully other foundations might be able to fill this gap as well. The State of California has also been very supportive of the type of research we do.
Do you have other recommendations, in terms of policy, as to how this problem might be addressed?
I think that -- in addition to us coming up with a way to actually fund this type of research -- we need peer reviewers who can actually adequately address this. Most of the review panels who have reviewed our work are either staffed by epidemiologists and biostatisticians, who know little about pathogenesis, or by laboratory-based investigators, who don't really understand clinical research. I think we need more clinicians and clinical researchers are needed for theses review panels.
Is there a dearth of clinician researchers in the field generally?
There is a huge dearth. There is editorial after editorial written about the lack of clinically-oriented translational research, it's talked about all the time. You know, and the problems go well beyond funding; it gets back to some of the institutional stuff. Most clinicians essentially get overwhelmed with clinical responsibilities and really can't devote the tremendous amount of time that's required to understand the pathogenesis stuff. The NIH K-awards have been very successful in addressing this, but they typically run their course in an abrupt manner after four to fine years, and it's hard for a clinician to fill in the gap without resorting back to doing a significant amount of clinical work.
Another issue is that generally you succeed academically based on first author and senior author papers, but clinical research is often, by its nature, very collaborative. The best clinical research, the best translational research, is very collaborative and requires multiple investigators contributing equally; Some of the best contributions that I think I've made to science have not been as a first or senior author, but such work does not really help much in getting promoted at most universities.
And you're involved in the Cleveland immunopathogenesis discussion? Has that been as productive?
Yeah, but that's very illustrative. We were involved in this Cleveland-based approach that's really focused in on immunopathogenesis and translating to the clinic. We participated in a recent program project in which we prospectively tired to link laboratory research with the type of focused interventional studies which I describe earlier. However, we were asked to to remove our clinical trials from the POI, and to submit it separately via the R34 mechanism. This has made it very challenging for us to link the lab with the clinic in what we had hoped would have been a highly synergistic relationship.
And are there research questions that have come up, do you think, because of your direct involvement with the pressing issues that people are facing clinically?
Oh, yeah. And so it's a two-way street, right? So I think that the lab informs the clinic, and the clinic needs to inform the lab. And one of the things that we're now really, really becoming very interested in is these individuals who have been on long-term therapy; viral load undetectable for many years; but their T cells essentially remain well below normal, and there's a sense that they are suffering consequences -- either accelerated atherosclerosis or other types of complications. We haven't confirmed that, but others have the same sense.
But as a consequence, we are now using our cohort to support two lines of investigations: one, the impact of aging on the immune system; and two, the impact of abnormal persistent inflammatory responses in the face of suppressive ART on cardiovascular function and so forth.
So it's these types of observations that feed some of the pathogenesis work -- I think, actually, it's one of our strengths.
Are there additional things could be done to help foster collaboration in those kind of projects with the basic science people? A more proactive approach by NIH or the Division of AIDS?
Well, I think the Division of AIDS needs to, again, have a peer review system that's made of up peers who do this type of work -- clinical translational work. I think the institutions need to acknowledge that clinicians require some set aside time to get this type of work ongoing. I think that academics, in general, needs to understand the issues regarding authorship, and the importance of middle authorship in some of this type of work. And I think people, in general, need to understand that human-based research is very, very, very expensive. Lots of regulatory and administrative stuff goes along with this, and at the end of the day, it just costs a lot of money to follow an individual over time.
And I think this is why that what we've done here at UCSF has been hard to do elsewhere, because here at UCSF, we've had tremendous institutional support, primarily through the CFAR system -- which I think has been a great success -- that allows us to do this type of research. But at other institutions, I don't think there's that same type of support for this type of clinical translational research; and there, it's practically impossible for a clinician such as myself to become an independent, successful investigator.
Do you think more consideration should be given to replicating that model?
We would love to see our model basically replicated elsewhere; and our model is essentially, with institutional funds and some NIH funds, to develop a prospective, biologically-oriented cohort of various individuals, led by epidemiologists, clinical investigators, who appreciate the pathogenesis work; understand it; and who also can inform laboratory-based workers of what the next questions might be, and access the samples for answering those questions. I think that that is a model that should be set up as a core at all major institutions; focus it on this type of research.
Are there any other successes that you've had here that you think could possibly be replicated in other institutions?
Well, at the end of the day, it comes down to the people. You know, we have this institutional support. We have a tremendous support by the community -- groups like Project Inform and so forth. We have a patient population that's highly informed and motivated and we have referring clinicians who are very supportive of the research we do.
But you might find that in most cities. What I think is key here is that we have a series of laboratory-based investigators who are engaged by this type of work -- people like Mike McCune, Doug Nixon, Jay Levy -- who are all immunologists willing to work with human samples. And without those individuals, we'd go nowhere.
And there are a fair amount of immunologists who are interested in human-based research, but it's very, very challenging to do really rigorous studies in humans with regard to Immunology for many, many reasons.
So it's hard to get them away from the mice?
You know, it's much easier to publish in Nature if you do knockout experiments in mice, and I'm sure that that stuff is important. But at the end of the day, it has to be confirmed in humans.
In terms of immunotherapeutic approaches, are there any particular approaches that you're intrigued by or excited about right now?
Yeah, we're very interested right now in a lot of work that's being done regarding the critical role -- the absolute key role -- that CCR5 plays in pathogenesis. The non-pathogenic monkey experiments by Guido Silvestri and so forth, really suggesting CCR5 is key. The genetics work by Sunil Ahuja suggesting that CCR5 is key.
And then there're some recent clinical trials data with maraviroc, an R5 inhibitor, suggesting that these drugs have an effect on the immune system independent of its effect on the virus. So we're very interested in CCR5 inhibition as an immune-based therapeutic.
And in large part based on the work of Danny Douek and our colleagues on the gut, regarding the loss of mucosal integrity and its potential impact on long-term outcomes, we are very interested in a series of TLR antagonists.
And I keep my eye very closely on what's happening in rheumatoid arthritis and the various different autoimmune disorders, because there's a tremendous amount of very focused biologics which can be used to manipulate the immune system in presumably a safe way; and I actually think that the people doing transplant biology and autoimmunity work might actually come up with the next great thing that can really push the field forward in HIV.
That's interesting. So you think there might be room to have some kind of collaborative discussion between researchers in those two fields?
I would love to see some kind of connection between the very successful, NIH-funded, Immune Tolerance Network and pathogenesis-oriented clinical investigators in HIV. Actually, I think that the Immune Tolerance Network should be a model for the future for individuals trying to figure out immunopathogenesis issues.
And I just saw you're involved in something that I actually hadn't heard of until a couple of days ago -- the NA-ACCORD?
Yes. To stay on the focus of pathogenesis, one of the best ways to move the pathogenesis field forward is to access information regarding what's happening in the clinics worldwide, or at least in the United States and so forth. And NIH has done, I think, a very nice thing by funding this NA-ACCORD multi-cohort collaborative approach, which is now getting up and going; which essentially is going to try to link all the cohorts in the states. And then I think that kind of database can lead to trends, like early mortality or new cardiovascular events, which is absolutely key for clinical management, for public policy. But I think also key for pathogenesis.
Was that something that you had a sense was happening in your clinic before you saw the SMART results?
You know, we've had a sense -- a growing sense -- before the SMART data came out that HIV itself is causing lots of the complications that we used to blame on treatment.
I've had this growing sense that HIV itself is causing progressive neurologic damage; loss of mental acuity; perhaps cardiovascular stuff; perhaps renal. And in a large part, based on that, I am becoming a bigger and bigger fan of early therapy, and think, really -- essentially -- everyone needs to be on meds unless there's a reason not to. That's been my sense, because I do spend a fair amount of my time in the clinic.
Are there any other sort of dream things that you'd like to see realized that would make your life and your work better and easier? I know that's kind of open-ended!
You know, I love my job. I love the institutional support we have here, and I love this group of collaborators we've developed. To a certain degree, I have an ideal job. The only issue is money; and so at the end of the day, more money, of course, would make my job even better. But I have this incredible mix right now of a nice proportion of my time in the clinic, and the rest of my time is collaborative work.
And in terms of advocacy groups like Project Inform and TAG, are there specific things you think we can be doing that would be helpful?
I actually was at this Project Inform town hall meeting last week, and I went there to talk about the Merck integrase inhibitor, which I think is essentially almost a miracle drug. Incredibly effective; fairly potent; it's going have a huge impact on my patients with drug resistance. So I figured we would talk for a couple of hours about integrase inhibitors.
But I wish people had been at that meeting, because it was just amazing. It was packed, and it was primarily people that had been on long-term therapy; and all people wanted to talk about were these issues which, at the end of the day, it seemed to me always came back to chronic inflammation -- joint pain, sort of weight loss, persistently low CD4 T cell counts, various complications despite having an undetectable viral load.
It really was a discussion not so much about drug toxicity. It certainly wasn't a discussion about getting a viral load undetectable. It was really about quality of life issues after being on therapy for 10, 15 years. And it really struck me that, at the end of the day there, we were talking mainly about what sounded like inflammation-related, autoimmune-type symptoms.
I actually thought that was a great success because it linked, via Project Inform, investigators with this broad cross-section of individuals struggling with the disease, and my only regret was that people who fund this type of work were not there. You could see where the next series of questions is going to be.
So as a means to generate research agendas responsive to the real concerns people have, you think that's the kind of thing maybe people should be trying to -- on a national level?
Yeah. On the national level, we should be bringing together individuals who have the disease, who are on the vanguard, with those who are trying to figure out how to fund things in the future. I don't know if that's done frequently, but that could be done, and groups like TAG and Project Inform would be ideally suited to do that.
You know, and I think that that continued informing of the public about the fact that, despite the miraculous changes with therapy, there are many, many issues to be resolved in the treated people. And so we can't forget about that as we focus more and more on prevention.
And I guess, one question often comes up, and it is just whether you're optimistic about the possibility of getting beyond antiretroviral therapy at some point; whether the body might be able to be persuaded to do a better job of controlling HIV?
I'm actually very optimistic but -- I gotta tell ya -- I'm a bit dubious about whether we're going to come up with an immune-based therapeutic that prevents disease progression such that people never need therapy. We can certainly delay it, and we can probably improve the immune system in people on therapy.
But I actually think that immunopathogenesis-oriented work and immune-based therapeutics will lead to a cure, and actually, we're really focused on these elite controllers as potentially a first step in that direction.
And you know, I'm naive. I don't understand the viral latency stuff as much as I probably should, but I'm somewhat optimistic that a cure might be feasible; and I'm happy that people are now willing to at least talk about it again.
Did I read something where you said there're some elite suppressors where you can't detect T cell responses to the virus?
Yeah. We have a subset of our people who are controlling the virus who have no evidence of an HIV-specific T cell response. None at all. They have an antibody that's positive, but that's it. So who knows? The assumption is the virus is there. We're going to look for it as carefully as we can, but proving the absence of something is not that easy.
Those CD4 T cell responses seemed pretty good correlates in the PCATs.
Yes, there's no question about that in our PCATs -- those, you know, partially controlled by antiretroviral therapy -- and in our elite controllers, they are -- in general -- they have very high (HIV-specific CD4 T cell) responses, in general. And if you just look at the average levels, they're actually highest in the elite controllers compared to anyone else; but if you look at the individual data, it breaks down into maybe a half, a little bit more, have high responses, and then have some middling responses. But upwards of a third basically have nothing, and we're not sure what to make of that. We find that third actually the most interesting.
The cause and effect debate seems to continue.
It's a cause and effect debate. I mean, to me, you just need the right control, and in our cohort, we look at people who are no virus due to an immune response versus no virus due to treatment. We remove the virus from that argument; and there, you can get very precise insights into cause and effect. And we think, clearly -- at least in a subset of individuals -- T cells are causally related to low virus level, and I don't think anyone really questions or that, or hopefully, no one really questions that. But perhaps some do.
What's the longest partial controller follow up that you have?
Oh, we have individuals who -- in terms of drug resistance -- we have a large number of people who, like, in 1991, 1993, were essentially on their deathbed, who went on Crixivan monotherapy back in the day; and got this robust increase in T cells, and essentially have been in this situation since that point in time. A lot of them -- you know, with low levels of virus just replicating along, moderate levels of virus, good T cells, maintaining reasonable health -- a lot of these people, however, we are going to be switching very quickly to the integrase inhibitors because even though these partial controllers do great for a while, I doubt they were going to do great forever. And no virus is better than a little bit of bad virus.
So we're not sure in terms of these questions, how everything's going to be impacted by integrase inhibitors. My sense is that a lot of people are going to do well, but a significant minority will not, and we'll need to continue this kind of work for those patients. Because if you look at the management of drug resistance right now, there's going to be a renaissance, right? It's going to be like 1996 over again, with this whole new paradigm. And it's going to be dramatic.
But just like in 1996 -- when HAART first became available, but it didn't work for everyone -- these next two drugs are not going to work for everyone. And then, we're in trouble, because the pipeline after this current generation of drugs is not looking particularly good.
So although in the short run we're going to do well with managing drug resistance, in the long run, we're going to be in trouble, for two reasons. First, there's not a lot of drugs being developed; and second, although we can manage the virus better, people are going to be living for decades. And although the rates of significant virological failure are going to be lower, we're going to be treating people for such a long period of time that we're going to eventually have large epidemic of multi-drug resistant HIV I think- in the distant future; and we're not going to know what to do, because I think drug development is slowing down now, and it may not pick up.
Do you think it's because of the more crowded market now?
The crowded market, and it's hard to make money on these drugs -- particularly when you're put into the category as a salvage drug. It's just that people don't go on these drugs and stay on them for as long as others.
So I don't think anyone's made money in that market, so I think that that is discouraging people from going into it. Plus all the money is made in the first-line regimens, and the first-line regimens are just getting better and better and better, and it's going to be hard to improve on that.
And I guess that problem is exacerbated for anybody developing an immune-based therapy?
No, the problem is I think people developing immune-based therapy need to look at successful ART as an opportunity. Right? Because it's keeping people with a damaged immune system alive much longer, where the subtle immunologic defects are going to become very evident.
Without question, most people going on long-term therapy -- their immune system basically goes back to normal, and I don't think anyone's going to see long-term consequences. But the subset -- particularly those who start therapy late, the low CD4s, or people coinfected with hepatitis C, or people who are older -- I think that these individuals are going to have significant immunologic defects that are going to need a safe immune-based therapeutic.
But you need to focus -- people doing immune-based therapeutic work need to understand what's going on in the clinic so that they can focus their approaches to what the need is. And if anybody had been at this Project Inform meeting last week, they would've gone away with the message that it's the immune system now that needs to be treated, not the virus.
Do you see a lot of autoimmune disease in the clinic?
We don't see autoimmune disease particular, but we see symptoms consistent with autoimmunity; particularly these chronic pain syndromes, these chronic myalgias, chronic fatigue, malaise, arthralgias, low-grade fevers -- you know, this type of stuff. When you talk to somebody, you go boy -- that sounds like autoimmunity; but it's not like lupus, or rheumatoid arthritis. It's just this sense.
And you see it in people who've been on what we consider reasonably safe drugs. No one's on d4T anymore. No one's on these other drugs. And their viral load has been undetectable forever. So we see a lot of it. But it's a Gestalt, right? I could not go out and do a study of this, because all these symptoms are vague. And it would be very hard to prove that they're not related to the drugs, or that they're actually not more common than you see in HIV negative people. So it's a tough situation. It's very descriptive. And I think if you ask community-based docs who do nothing but full-time HIV care, they would agree that this is an increasing problem. But there's no way to name the syndrome.
It's like lipodystrophy was back, I guess, in the late 90s. It was a vague sense that it's bad, and no one really had a good way of measuring it, defining it, and so therefore, research was basically slowed down as a consequence.
There's some literature on the association between lymphopenia and autoimmunity, Crystal Mackall just wrote a review, and then there was one a few years ago from Scripps, just in the small animal models that they have -- I think if you do something to make the animal lymphopenic, it can provoke autoimmune disease.
Well, it provokes T cell proliferation and T cell activation, which, in our work, is the best surrogate for these inflammatory outcomes. So that sort of makes sense.
But that -- you know, if you have access to people who are doing that kind of work outside of HIV -- a think tank, or a meeting or something -- bringing those individuals in with HIV people would be very worthwhile. Because even though I love reading the literature -- it's actually the best part of my job, I really enjoy it -- there's just no way I could keep up with all that and the clinical stuff.
All right. The key thing -- from my perspective -- to move these things forward is to avoid the anger. You'll talk to other people and they'll get angry, and they'll start attacking CHAVI and all this other stuff, and I don't think that's productive now. I think that NIH has legitimate concerns, and DAIDS has concerns. So they have multiple different individuals; the networks; the laboratorians; and so forth. I'm just saying this because I think I'm trying not to be inflammatory in my responses. Hopefully, what comes out of this will be more forward thinking.
Sure, that question has come up about what people think about the growth of big science approaches.
Well, investigator-initiated research, it's the American model. It's driven a lot of us, too. You know, we're generally individualistic, and so you can't give it up. Although I'm a big fan of big collaborative work.
And the general feeling is that it would be best to do both, I think, is the sense that I'm getting.
Okay, I really appreciate it.
TLR: Toll-Like Receptors. TLRs are specialized receptors that can recognize pathogens non-specifically based on certain structures that are common to many different pathogens. A number of different TLRs have been identified and it has been hypothesized that TLR recognition of components of HIV and/or other organisms (such as other pathogens or commensal bacteria from the gut) may contribute to the persistent immune activation that is seen in HIV-infected individuals.
Cleveland Immunopathogenesis Consortium (CIC): Collaborative project facilitated by Mike Lederman from Case Western Reserve University involving a diverse group of researchers interested in solving outstanding questions regarding the pathogenesis of HIV infection.
NIH: National Institutes of Health, www.nih.gov
NIAID: National Institute of Allergy & Infectious Diseases, www3.niaid.nih.gov
DIADS: Division of AIDS, www3.niaid.nih.gov/about/organization/daids/
This article was provided by Treatment Action Group.