Anthony Fauci Discusses Research Approaches to Sustaining HIV Remission Without Antiretroviral Therapy

Anthony Fauci, M.D.
Terri Wilder

In the nearly 40 years since the discovery of HIV, probably no researcher has been more central to U.S. research than Anthony S. Fauci, M.D. And, as the director of National Institute of Allergies and Infectious Disease (NIAID), the International AIDS Conference is often a busy time for Fauci. At this year's AIDS 2018 in Amsterdam, Fauci spoke to TheBodyPRO before his two conference sessions on approaches to HIV remission and durable control of HIV without antiretroviral therapy.

(In part 2 of this interview, Fauci discusses his remarks to the U=U Preconference, detailing the scientific research that led us on the path to the discovery that undetectable equals untransmittable.)

Terri Wilder: Well, let's discuss your talk for tomorrow, which is going to be focused on the induction of durable remission.

Anthony Fauci, M.D.: Right.

TW: There are two key pathways towards remission. There's the use of intermittent, non-[antiretroviral treatment] (ART) interventions, and then there's the induction of durable mediated controls of the virus. Can you talk about each of these and the progress in each area?

AF: Right. Well, the underlying premise of the talk is that, given the difficulty and risk of trying to truly eradicate the virus with very aggressive, draconian means, what we have been focusing on intensively of late, and what is the topic of my talk, is how to have an ART-free remission without even trying to eradicate the virus. And you mention correctly, there are two pathways. Can you intermittently intervene by any of a number of interventions that would be a much more user-friendly way to get the patient off having to take antiretroviral drugs every day? I'll give you a couple of examples of that in a second.

The next approach is to induce something on a one- or two-time basis that then does not require further intermittent intervention but actually induces in the body the kind of immunological control that natural infection was incapable of doing.

What we're proposing -- and this is a proposal in my talk; it hasn't been done yet -- is to use the same strategy that we used historically years ago when we were dealing with antiretroviral drugs -- namely, first we gave one drug, AZT [zidovudine, Retrovir]. It dropped the virus a bit, but not durably. Then, a few years later, we had two drugs. It dropped the virus even more but, again, not durably. And then, finally, when we had the combination -- a triple combination of drug -- for the first time ever, we dropped the virus to below a detectable level, and it lasted indefinitely as long as you were giving antiretroviral drugs.

So, what I'm proposing is that we do the same thing with broadly neutralizing antibodies that we know we have antibodies against each of the individual neutralizing epitopes on the HIV envelope, very similar to antiretroviral drugs against reverse transcriptase, against protease, non-nucleoside, and the others.

We have antibodies -- we have drugs against each of those. If you look at the sites of the envelope, we have the CD4 binding site; we have the membrane proximal external domain; we have the V1-V2; we have the apex; we have the others. We actually have the antibodies against each of these, so if we do the same scenario as we did with the drugs, instead of giving passive transfer of one antibody, we give two or three.

Now, we have the ability to engineer the antibody so instead of it lasting two months, you could last four to six months. So now, it is conceivable (and not only conceivable; we're going to do it) to passively transfer an antibody to an individual twice a year, in combination, and that would be their antiretroviral drug. So, they would not have to take daily antiretroviral drug. They just need passive transfer of triple combination, maybe twice a year.

Now, the next broad strategy is not having to give an antibody, even as infrequently as every six months, but to do something to the patient that would induce a response that would essentially have them be protected forever. In other words, induce an immune response that the virus itself is incapable of doing.

And there are two pathways to that. The first is this issue with alpha-4 beta-7. Alpha-4 beta-7 is an integrin, which is a molecule on the surface of the CD4+ T cell. To make a long story short, we've been studying this in my laboratory for the past 12 years, and we found it is an alternative receptor for HIV that is not necessary for infection, but it is an enhancer of infection. It identifies a subset of cells that are highly susceptible to HIV infection. So, when HIV infects cells, it selectively infects these cells.

We showed that individuals who have high levels of alpha-4 beta-7 cells are ones who have a higher likelihood of getting infected and a worse course. Based on that, we partnered a few years ago with collaborators who did monkey work, since we don't do monkey work in our lab. We showed that if you block alpha-4 beta-7 with an antibody, you can prevent infection in a macaque challenge model. And, if you deliberately infect the macaques and then give them antiretroviral therapy, and then give them antibody to alpha-4 beta-7, when you stop everything, the animals don't rebound -- which was a very striking finding, which we published in Science.

Based on that, we did two simultaneous studies. We repeated the animal study, the macaque study, in our own lab at [the National Institutes of Health] as opposed to the collaboration. And then, we started a human study with the drug vedolizumab, which is a drug that's approved by the [U.S. Food and Drug Administration] -- it's an antibody to alpha-4 beta-7 -- and it's used for the treatment of ulcerative colitis and Crohn's disease.

We found when we repeated the animal study, to our interesting surprise, we were not able to get as dramatic results as the original study that was done at Emory and Tulane. But what we did notice was that, in the animals, there was a great degree of variability -- which is a big, red flag about be careful with animal studies; they're so variable that a result in one lab may not be able to be completely replicated.

Having said that, the proof of the pudding is in the humans. So, we went ahead with the human study, in which we took individuals who were treated for two or more years, successfully suppressing their virus. We then gave them multiple passive transfer of vedolizumab, similar to the animal model. Then, we stopped everything to see if we had a delay in rebound.

To our disappointment, we did not see it. We saw some variability, but when you did a comparison of the historical control, it was an open-label study for safety and to see whether there was an effect. So, if we had seen a dramatic effect, an open-label, nonrandomized control would have showed us. But short of the dramatic effect, when you have an open-label, not-controlled study, it has to be very dramatic. We didn't see a dramatic effect.

Whether there's an effect buried in there, we don't know. But it certainly was not a dramatic effect.

I close my talk on showing data from a very interesting study that was done on macaques by Malcolm Martin, Michel Nussenzweig -- and I'm also on that paper -- in which they took monkeys, they infected them, and after acute infection, a few days, they passively gave them two monoclonal antibodies against human virus, because they were using a SHIV, not an SIV.

Interestingly, of the 13 monkeys that got that, six became elite controllers. When they stopped everything, the monkeys totally controlled the virus, which is strongly pointing to the second strategy that I was saying: that even after infection, you could induce in the infected animals a response that can durably control. So those animals now are without therapy and doing fine.

The theory that we think is accounting for that is that when you get a highly potent broadly neutralizing antibody early in infection to bind to the virus, it forms immune complexes, which get presented to the immune system, which induces a potent CD8+ T cell. How do we know that? Because in the animal study, when we depleted CD8+ T cells with an antibody to CD8-beta, the virus jumped right back. And, as soon as the CD8 cells came back, the virus came back down -- which I think is a very important finding, proving that the mechanism of elite control is a highly reactive CD8+ T cell.

That was the end of that study. And the conclusion is that one, therapeutic vaccines are unlikely going to be a route to have durable control. And two, despite the fact that there was some interesting suggestion that antibody to alpha-4 beta-7 could induce durable control, at least in the uncontrolled open-label study that we did, we didn't find any durable control. And finally, in an animal model of SHIV, by passively transferring broadly neutralizing antibodies, we were able to induce durable control. And finally, based on the passive transfer of antibody in infected people, we feel that would likely be an alternative to daily antiretroviral therapy.

TW: So, you covered a lot of information.

AF: I did -- in 26 and a half minutes.

TW: Yeah. Yeah, you did great. So, in terms of some of the things that seem promising ...

AF: Right.

TW: What's the next step?

AF: The next step is to get combinations of broadly neutralizing antibodies, and we're going to be passively transferring them in individuals who are antiretroviral therapy suppressed. And then, we're going to stop antiretroviral therapy and see if we can space it intermittently enough that we never have to put them back on antiretroviral therapy.

TW: For these individuals that are suppressed, does it matter when they acquired their HIV?

AF: No, it does not. Because we've shown already in chronically infected individuals that the antibodies work just as well.

TW: OK. So, I guess what I'm asking is whether they need to be newly diagnosed.

AF: No, they don't.

This interview has been lightly edited for clarity.