Antivirals and Immune Recovery: Interview with Michael S. Saag, M.D.

Michael S. Saag, M.D., is Associate Professor in the Division of Infectious Diseases at the University of Alabama at Birmingham. He is both a laboratory scientist and a physician who treats patients. "I combine my laboratory experience and my clinical experience, and participate as a member of a dynamic scientific group here at the University of Alabama at Birmingham, where I help translate what's happening in the lab and what's happening in the clinic, so the clinical and laboratory groups can communicate efficiently. We are trying to coordinate new developments in the laboratory with clinical care and apply the developments to patients as soon as possible."

Dr. Saag's group has done some of the earliest work with L-524, the Merck & Co. protease inhibitor, a new anti-HIV drug which is now in small-scale clinical trials. He is finding evidence that the immune system can recover on its own to a surprising degree, even from late-stage AIDS -- if the virus can be suppressed enough, as measured by experimental blood tests.

AIDS Treatment News: In a time of widespread pessimism, you have said that there are important grounds for hope.

Dr. Saag: Much of the pessimism among HIV patients and their physicians accelerated a little over a year ago, as the Concorde data [questioning early use of AZT] started to be discussed. I saw people get hung up on the details of a single study, and lose sight of the big picture.

In the big picture, this is a viral disease. The virus constantly wants to replicate, and the immune system tries to suppress replication. The use of antiretrovirals helps suppress the virus, but to date, the available agents are effective for a limited time. The good news is that we have learned a lot about HIV in less than a decade. We can target specific elements of its life cycle to attack with antiviral therapy. We can test new approaches [for example, protease inhibitors] in the laboratory and find that they work; now we are beginning to test them in people and find that they work quite well, at least so far. The fact that we can figure these things out is quite a testimony to basic sciences being applied to clinical medicine. This is unprecedented in the history of treating infectious diseases, where discoveries are usually made by lucky accident, rather than by a reasoned approach to drug development.

And now we have newer techniques to measure how well these drugs work; we can follow the viral burden, as measured by circulating RNA levels of the virus. We need more research to be sure that tests not only tell us that the drugs are active, but also translate into clinical benefit. Then we can use the tests to tell us which drugs are working best in a specific individual. If we can achieve that, it will be a real step forward, and we will be great shape in the next two or three years.

ATN: Is the P24 antigen test not very effective?

Saag: It could be OK if someone has a level that is measurable. But for the majority who are not p24 positive, or for those who are marginally positive and convert to negative, you cannot tell how well the drugs are working and how long their effect lasts. I think the HIV RNA is a much more direct marker in people; I believe that ultimately it will be proven to be useful in day-to-day practice.

ATN: What needs to be done to get more certainty, and to make these tests more widely available?

Saag: We need to show that the tests correlate to clinical outcome. That is easy to say and hard to do.

One approach would be a retrospective study -- a case- controlled evaluation of a known cohort. We could use stored specimens from previously conducted trials, and select specifically those individuals who had disease progression, and compare them with those who did well on the study. This might require several hundred patients over many time points. Then we will see if these tests can indeed predict who is going to do well. Other studies may prospectively evaluate the marker, allowing those with a poor prognosis to keep trying different treatments until the tests show that the prognosis has improved.

It has been shown that CD4 counts (T-helper counts) can show indirectly whether or not an antiviral drug is working. But they are not good at telling us precisely what is going to happen to an individual clinically. That has always been a problem in comparing drugs; first, because existing drugs were not powerful enough to make profound changes in T-helper counts and immune system restoration, and secondly, because there is much variability in the measurement of the counts. T-helper counts are still useful, but not as good as they need to be. These new blood tests (for levels of HIV RNA) may fine-tune the measurement so we can predict how someone is going to do clinically.

ATN: Has early experience with the protease inhibitor L-524 shown that, if you can suppress the viral load, there is evidence of immunological and clinical improvement?

Saag: AZT seems to do this as well, but to a smaller degree. If you reduce viral burden by ten fold, with AZT, you'll see an increase in T-helper counts, but not a profound increase. They may go up from 50 to 100, for example, or 100 to 200. But it seldom does much better than that, and the durability of the increase is about six to 12 months, if you start later in the course of the disease. If you start earlier, for example with baseline counts of 400 to 500, the durability may be 18 to 36 months.

With the protease inhibitor L-524, we have seen some examples of viral burden dropping around 100 fold. When we achieve that, we can see at least temporarily a substantial increase in T-helper counts -- in some cases from below ten, to around the 200 range, lasting for several weeks or months. This is important because it is proof of concept that the protease inhibitors indeed can work, and that they mirror what we hoped to see from the test-tube experience. Also, it shows that if you can suppress viral replication to a more profound degree than with AZT, then the immune system has some ability to reconstitute itself, at least as measured by the T-helper count. To a degree we knew that from AZT; now, with new agents which may be more potent than AZT, we are seeing more immune system reconstitution as a natural consequence of reduced viral replication. [So far the effect is temporary, however, since HIV develops resistance to the drug, and then the viral load increases.]

The recent experience with L-524 and other antiviral drugs has been telling us a lot about viral pathogenesis. The virus is becoming easier to understand in terms of what it is trying to do -- to survive and replicate. It is constantly trying to infect new cells, and to turn itself over [replace the current virus with new generations]. This suggests that ideally we should be treating patients as early as possible. Whether people believe that AZT is good enough to begin therapy when T-helper counts are above 500 remains to be seen in clinical trials. But if we could imagine drugs that are two, three, or ten times better than AZT, then I don't think there would be much of a debate. If they are relatively nontoxic, we will be using the drugs as early as possible.

ATN: We heard about one person who started L-524 with a T- helper count of 600, and had a large increase which continued even after the drug was stopped.

Saag: One patient was in our earlier studies, when we just treated patients for 12 days. He had a T-helper count in the 600 range, and it did go up to over 1000. He only had 12 days of therapy, and he stayed around 1000; he is still there at this time (12 months later). I can't say that this was due to the L-524, but it is a possibility. If we consider that this is a virus that is constantly trying to replicate, when you give antiviral therapy, you're adding air support for the infantry; perhaps using a drug with that much activity early in the course of disease may have allowed his immune system to get a better handle on the viral infection, and gain better control -- having effects even long after the drug was present.

The other possibility is that the T-helper increase would have happened anyway, due to variability in the expression of the disease. But usually one would expect that T-cell counts, left to their own devices, would continue to drift downward, not bounce up by 400 cells per cubic millimeter.

ATN: It would make a lot of sense to try this treatment with a few more patients, and see if it happened again.

Saag: I think so. We should do it in a controlled way, following people with that goal in mind. Remember that in the early studies with this compound, all we were really looking for was safety and the pharmacokinetic profile; what happened to the patients clinically was more of an incidental observation; we were certainly interested in it, but that was not an objective of the study.

We need more research on earlier infection. Indeed there are some trials in development right now, not so much with protease inhibitors just yet, but with some of the nucleoside analogs [e.g. AZT, ddI] and non-nucleoside reverse transcriptase inhibitors [e.g. nevirapine, or delavirdine], looking at earlier interventions to see what antiviral effects we get in people with higher T-helper counts.

ATN: The big interest now is combination treatments, for delaying or avoiding resistance; this has worked for other diseases. Is there anything you can add from your work? What combinations would make sense to test with L-524?

Saag: The results so far with the Merck drug (L-524) are all with monotherapy [the drug alone, not in a combination]. Roche is finding some success with their protease inhibitor in combination with AZT. So the first approach would be to use the protease inhibitors with AZT. The rationale is that you're attacking two different areas of the viral life cycle; that is sound reasoning.

Another approach would be to combine a protease inhibitor with other nucleoside drugs that have different toxicity profiles [for example, ddI], or perhaps with non-nucleoside drugs, such as nevirapine.

We may also want to try combining different protease inhibitors. The resistance patterns for the different protease inhibitors are different. After growing HIV in cell culture in the laboratory [with low doses of a protease inhibitor, to deliberately produce resistant viruses], resistant or less susceptible viruses can be identified, and they have certain changes in their genetic structure that confer the resistance. If you do this with another drug, you get a virus which is less susceptible to that drug, but often not less susceptible to the first drug; there may be little or no cross resistance. The reason is that different changes in the genetic makeup of the virus give resistance to the different drugs.

So if you have the two drugs together in the same patient, you might have broader coverage against the two types of genetic changes that could occur, just using protease drugs. That may work well in the future. But first we must prove that the new drugs are safe and well tolerated, and that the drugs being combined do not have harmful interactions. If they can be safely administered together, a combination test will be important.


***** Government AIDS Research: the ACTG. Interview with Michael S. Saag, M.D.
by John S. James

[Note: This interview was conducted as a continuation of "Antivirals and Immune Recovery," above.]

Michael S. Saag, M.D., a researcher at the University of Alabama at Birmingham, is a new member of the Executive Committee of the AIDS Clinical Trials Group (ACTG), the main U.S.-government program for clinical trials of AIDS drugs. Among other responsibilities, Dr. Saag directs the subcommittee in charge of the ACTG's interaction with outside groups, including the pharmaceutical industry, community organizations, and government medical-research teams outside the ACTG.

ATN: What is most important for our readers about the ACTG?

Saag: I think the ACTG is a very valuable resource for the country. It has taken a lot of criticism, some of it fairly applied. But the success of the ACTG is not widely understood.

If you think about what happens in clinical practice today in a doctor's office, and compare it to six or seven years ago, you will find that much -- perhaps 70 to 80 percent of the activities of a physician, in terms of new approaches to management of their HIV patients -- was initiated through the results of an ACTG trial. When you go to conferences, and hear discussions about patient care and trial results, you almost always hear references to ACTG studies. There have been many contributions, not only in the primary infection area [treating HIV itself], but especially in opportunistic infectious diseases. What types of prophylaxis work best, what is the best treatment for cryptococcal meningitis or pneumocystis pneumonia or CMV retinitis, these results have usually come from ACTG trials. People should recognize that the ACTG has made major contributions.

Also, much that goes on behind the scenes at the ACTG helps pharmaceutical-industry studies. Consider the quality assurance program that assures that virology labs working with the ACTG all operate in a standardized way, and that the measurements of T-cell counts are standardized. All these spill over into industry trials. The sites that are ACTG funded also can do industry studies, and whenever a company goes to those sites, it is using the same laboratories that are ACTG certified. That's been a very important plus for studies that are not ACTG affiliated.

In the future, due to a recent reorganization, the work of the ACTG will be based more directly on a foundation of its scientific agenda. In the past, an individual researcher at an ACTG site would come up with an idea, propose it as a concept sheet, and go through the system from the bottom up. While that is democratic and taps into the great resources of a number of very talented people, it can also be inefficient. In the future, all these researchers with their talents and ideas will first come together to brainstorm and come up with a scientific agenda, and then develop specific protocols to address that agenda. This should speed protocol development and the conduct of trials.

ATN: We have heard of many cases where concept sheets don't go anywhere, probably because the leadership does not agree with them. This change might help to avoid that problem.

Saag: I think it will. And I think the other problem is that some members of the ACTG, especially in leadership positions, get so inundated with paper, having to keep up, that it's hard to give every concept sheet a clean, fresh look every time. You never know when one is going to be coming in.

So we will back up and say, "What are the big-picture issues that we should be addressing as a group?" We will define those questions, establish that agenda, prioritize, and then ask, "What kind of trials will help us answer this question?" I believe that the proposal I have seen will reduce protocol development time -- going from an idea to a protocol open for enrollment -- from over 230 days down to about 85 days. This is about to be implemented.

Also, we also should keep in mind what would happen if the ACTG did not exist. Suppose it had never been created; what would people be saying about what ought to be done? They would say that we should get together some of the best scientists and clinical investigators in the country, and form a network where they could work together and study new approaches to treat the disease. Also, you would want to bring those people together, at least semi-annually, to talk about progress, discuss current issues, decide directions together, and then implement those new protocols and studies that they come up with. That is precisely what the ACTG is. The problem is that it has been inefficient in its ability to get things done quickly; it has been too bureaucratic in the past. I think that with the reorganization it will become more streamlined, and enable us to fulfill those desperately needed missions of having some of the better investigators in the country working together on joint projects.

I'm very encouraged by what is happening. There is a dynamic interaction within the ACTG leadership, and a lot of dedication to making the organization run efficiently, and be as productive as possible. I'm optimistic that we can do that. But it will take much work to get us to that next level.

ATN: Recently some leading scientists have called for a redirection of AIDS research. [See "AIDS Research Direction: New Scientific Approach?," AIDS Treatment News #199, May 20, 1994.] Some have interpreted Dr. Fields' recent paper as calling from a shift from clinical to basic research; but what he really seems to be saying is that there should be better communication between the laboratory studies and the clinical trials which test new drugs in patients. Where do you think we should focus now?

Saag: There has been a fairly strong emphasis on basic science from NIAID [the U.S. National Institute of Allergy and Infectious Diseases, which runs the ACTG] over the last five or six years. In 1987, there was a project called Programs for Excellence in Basic Research on AIDS (PEBRA); our team at the University of Alabama was one of the recipients of that award, and that is what started us coordinating the clinical side and the basic science side. That's the kind of initiative we should continue with.

NIAID has continued this approach, with the new SPIRATS grants, which encourage investigators to work together from the beginning of a new concept of treatment, to developing that approach in the laboratory, and carrying it through all
the way to the initial clinical trials. [Note: SPIRATS stands for Special Program for Innovative Research on AIDS Treatments. The first five to eight SPIRATS grants are expected to be awarded in August 1994.] This is an important initiative that encourages creativity among investigators within different disciplines, having them work together.

This kind of approach is where progress in HIV will come from. I think we do it at the University of Alabama about as well as it's done anywhere; it was the U.S. National Institutes of Health programs that got us going. The PEBRA, our Center for AIDS Research, and other government programs, have all been used to have scientists from different disciplines work together. This is the critical issue that should be emphasized, rather than segregating out more money for basic science, or for clinical science. We are trying to get money used in a way that maximizes the interaction between the different disciplines and different research groups.