Project Inform's Response to Dr. Jay Levy's Letter to The Lancet

San Francisco, October 1, 1998

-- In a press release on September 15, Dr. Jay Levy of the University of California, San Francisco described a letter he placed in the British medical journal, The Lancet, on the subject of early vs. later treatment for HIV disease. Essentially, Dr. Levy appears to question the value of early treatment of HIV disease, when viral load is low, and theoretically proposes such treatment could be harmful to the development of a potent immune response against HIV. He makes a careful distinction in noting that he does support treatment immediately after a person becomes infected, if such an opportunity is available. He defines "early" treatment as any time more than a few months after initial infection but before a person nears a diagnosis of AIDS.

Although Dr. Levy has expressed this view previously, it has seldom received much attention from the media. His perspective has often been drowned out by the more common theme of AIDS researchers calling on physicians to "hit HIV hard and hit it early". Dr. Levy discussed his beliefs extensively at a recent meeting of the Immune Restoration Think Tank, an annual strategic planning meeting of top researchers sponsored by Project Inform. Many Think Tank members also expressed counter views. Similarly, contrary perspectives are frequently raised at University of California meetings when Dr. Levy describes his position.

From a scientific viewpoint, there is nothing wrong with debating either point of view. Science proceeds by discussion, dialogue and debate. The problem here, however, is that the debate is being raised in the form of a press release which in turn describes an editorial piece in a medical journal. Press releases can easily provoke the media -- which always love a good fight -- into overstating the controversy without really providing enough information to help people understand the issues involved.

Patients and physicians trying to make decisions about therapy may find it hard to know what to make of Dr. Levy’s views and why these views seem so contrary to what they’re hearing from others. Similarly, people who have already made the choice to initiate therapy may begin to question whether they’ve made the right choice and wonder whether there is a way to turn back the clock. Project Inform’s goal, as always, it to try to clarify the situation and make it easier for people to understand what is proven, what is suspected and what is a merely a matter of opinion. In this debate, there is some of each.

What Does Dr. Levy Actually Say?

A careful reading of Dr. Levy’s position suggests that his views are not dramatically different than those of most scientists, nor do they represent a fundamental challenge to the current Federal Guidelines on the treatment of HIV disease. For example, the Federal Guidelines suggest that treatment should normally be recommended for people whose CD4+ cell counts fall consistently below 500 and who have a viral load above 10,000. Dr. Levy suggests starting therapy when the CD4+ count is below 400 and the viral load is above 30,000. These are very small differences, and the hard fact is that data from clinical trials could be used to support either position. Looking across the breadth of clinical trial data available, either recommendation can be made, and it is unlikely that the patient outcome will differ much whichever recommendation is followed. They are, for all practical purposes, the same.

In most people who have not yet started on treatment, it is not uncommon to see CD4+ cell counts move around by a hundred points or more for no particular reason other than normal fluctuations in the white blood cell count. A patient with a count of 400 one week might easily get a result of 500, or even 300, a month later. Similarly, untreated viral load varies widely for a variety of reasons. A simple bout of the flu, or even receiving a flu shot (vaccination), is often associated with an increase in viral load. Perhaps more importantly, different versions of the test, or taking the test at different labs, can produce changes of 50% to 100% without reflecting any real or lasting change in the status of a person’s HIV infection. The more frequently people take such tests, the greater the chances that odd or extreme results will occasionally appear.

It is only when we look at people who fall above the general thresholds described by the Federal Guidelines or Dr. Levy that the scientific waters become muddy. For people with CD4+ counts above 500 and viral load below 10,000, the Federal Guidelines suggest two alternatives: treat or monitor. They correctly point out that scientific opinion is divided about whether to treat this class of patients or to simply monitor the patient’s condition. Like the Federal Guidelines and, for example, Project Inform’s patient support materials, Dr. Levy’s letter to The Lancet discusses some of the possible reasons why people who fall above the threshold limits might choose a more conservative approach to treatment. They can be easily summarized:

1. Roughly 50% of HIV-infected people can go for 10 years or longer without anti-HIV treatment before progressing to serious immune deficiency and an AIDS diagnosis. Therefore, many people will do just fine for several years after infection, with or without anti-HIV treatment. It might make sense to take advantage of this period of natural control of the virus and bring on the use of the drugs only when the risk of disease progression increases with the passing years. Charts from natural history studies can tell people their individual risk of progression to AIDS, without anti-HIV treatment, over a course of 3, 6 or 9 years. All that’s needed is knowledge of approximately when a person was infected, along with a few current CD4+ cell count and viral load tests. Such charts can help predict the risk of disease progress if a person chooses to delay treatment. Those favoring early treatment point out that it’s important to realize, though, that if 50% can go 10 years or longer without treatment, the same data mean the other 50% will progress to an AIDS diagnosis or death within 10 years or less after infection. Some people progress and die within 1 to 3 years.

2. Treatment comes with a price, other than the one measured in dollars. Some (but certainly not all) people suffer side effects from the drugs. There is nothing unusual about this -- it happens any time that medication must be used on a long-term chronic basis. The risk of side effects begins when treatment begins. While a necessary evil, it may be possible in some cases to keep that risk at bay by delaying treatment -- but only if the probability charts suggest the person can afford to wait. Some researchers, however, counter this concern by noting that people who begin treatment early when they are at their greatest levels of health are less likely to suffer side effects than those who use treatment only later, when the body and immune system have been seriously weakened. Moreover, they argue that HIV is quietly but relentlessly damaging the immune system, long before the patient is aware of it or before serious declines in lab work appear. Most scientists believe that if we had perfect drugs, without risk of side effects or resistance, there would be no debate about whether to start early or not.

3. HIV can develop resistance to drugs over time. While treatments do not technically cause the development of resistance, they do create the conditions that make life easier for resistant mutations of the virus to occur and therefore give them the opportunity to accumulate and become dominant. The bottom line is that some people can and do develop resistance to any or all of the drugs currently available. Therefore, a conservative view argues in favor of using the drugs only when you’re sure they’re necessary, thus preserving their potency for the most important time in the course of HIV infection. Those who favor early treatment argue that drugs generally last longer without developing resistance when used in healthier people at earlier stages of HIV infection and that a steady supply of new and better drugs is likely. Additionally, they point out that it is often easier to achieve maximum suppression of the virus to "undetectable" levels when treatment is begun early. Thus, they feel early treatment provides a net benefit compared to delayed treatment.

4. Using the drugs is no picnic. Although the newest drugs are becoming vastly easier to use (e.g. less frequent dosing requirements), some of the older therapies -- while highly effective -- make great demands on patients. Large numbers of pills must be taken, often three times daily, sometimes with or without food. In short, they can greatly complicate your life. Failure to use the drugs as required can greatly increase the risk of HIV drug resistance, so once a person gets on the "treatment train," it’s like choosing a way of life.

5. Finally, some scientists, like Dr. Levy, have noted that when people are successfully treated with combination therapies, over time some of the markers of the immune response against HIV itself decline. Levy, though not all others, interpret this as a bad sign and believes it signals that the body is losing its ability to fight the virus naturally. In support of his view, Levy cites anecdotes about individual patients he has seen who, after choosing to stop all therapy, saw their viral load climb to higher levels than before they ever used treatment. Thus, he concludes, their immune system has lost the ability to fight the disease. This is perhaps the most controversial of the positions Levy takes, as other scientists cite completely contrary interpretations of the same phenomenon.

Taken together, these five concerns might indeed raise caution about the possibility of starting treatment too early. But likewise, an examination of the counter arguments gives sound reasons for choosing early rather than delayed treatment. Like the Federal Guidelines, Project Inform has always taught that personal knowledge and consideration of these issues should help inform the decision about when to start therapy, not just slogans like "hit it hard and hit it early," or, for that matter, the opinions of any single researcher or physician.

An important point not addressed in either the press statement from the University of California or the coverage of the issue in local press is the issue of "early treatment" when CD4+ cell counts are high (e.g. above 400 or 500) and viral levels are high. Project Inform staff pressed Dr. Levy on this point after the release of the September 15 statement, and Dr. Levy noted that in patients with high CD4+ cell counts whose viral load was confirmed high (e.g. above 30,000 on at least two consecutive measures), he would recommend that someone consider the use of anti-HIV therapy. In other words, Dr. Levy’s position is precisely in line with the Federal Guidelines, noting that in some instances, even when CD4+ cell counts are high, intervening with anti-HIV therapy may be warranted. This does not contradict Levy’s viewpoint piece that intervening with anti-HIV therapy too early could dampen immune responses to HIV. In essence, it contextualizes his statement and puts it in a more reasonable light. Obviously, high viral activity, regardless of CD4+ cell count, could also damage and dampen the immune response. Levy’s cautionary statements encouring individuals to hold off on intervening with anti-HIV therapy are primarily directed at individuals who, without anti-HIV therapy, are maintaining high CD4+ cell counts and low viral levels.

Which Point of View is Correct?

Unfortunately, there simply have been no long-term clinical trials that prove which strategy -- for people who fall above the current guidelines for treatment -- results in the longest overall life for the greatest number of people. Therefore, the Guidelines say that both options are reasonable and the patient should be made aware of alternatives and, working along with the physician, make a personal choice.

This, however, is precisely where Dr. Levy parts company with the Federal Guidelines and many other researchers. While many willingly admit they aren’t sure what the best strategy is for the healthiest patients with the best lab work, Dr. Levy feels he is sure. He argues against treatment, period. In this sense, he is at the opposite end of the spectrum from those who insist on "hit it hard and hit it early" even in the healthier patients.

Since no meaningful clinical trials have been run to support either his view or the alternative, Levy bases his on a different concern, one seldom raised by other researchers. He argues that early treatment for the healthier patients runs the risk of blunting their anti-HIV immune response. In turn, he believes this makes them more dependent upon antiviral drugs and diminishes the capacity of their immune systems to fight the disease. As evidence of this, Levy cites data, available from many sources, which show that the total number of CD8+ cells declines after long periods of effective treatment. He argues that high levels of CD8+ cells, specifically those fighting HIV infection, are a common characteristic of long-term non-progressors -- people who retain strong immune status and low viral load for 15 years or more without treatment. Therefore, he concludes, if those cells decline as a result of other successful treatment, the patient is becoming more dependent upon drugs to control HIV and losing the natural response.

This sounds quite convincing, at least until one hears what other researchers say about it. One obvious point left out of Levy’s description is that the high level of CD8+ cells seen in HIV infected people is in fact an abnormality, a characteristic so striking that it can almost be used to diagnose HIV infection. For example, an inverted or reverse ratio of CD4+ to CD8+ cells (i.e. twice as many CD8+ cells as CD4+ cells) is characteristic of the disease. The abnormal "ratio" of CD4+ to CD8+ cells continues to get worse with HIV disease progression, as CD4+ cells die off. In the end, the CD8+ cells themselves often suffer a rather sudden and final collapse. Other researchers say that the reduction of CD8+ cells in response to therapy is in fact a sign of improvement in the immune system, a return to its more normal state and its normal ratio of CD4+ and CD8+ cells. Levy is concerned primarily with a specific subset of this population of CD8+ cells, called HIV specific CTLs. Since few if any studies measure these cells directly, he assumes that if the total CD8+ population is decreasing, then so too is this subset. However, other researchers argue that the HIV fighting CD8+ cells decline precisely because the drugs are so effective in suppressing HIV infection. In short, the body stops making the HIV specific CD8+ cells because it is no longer fighting HIV, or at least the fight is now at such a low level as to be imperceptible.

Still, Levy worries this may signal a complete loss of such cells, and that if people stop treatment, their immune system won’t know how to make them anymore. He cites as evidence a few cases in which patients went off therapy and saw their viral load rise to levels higher than when they began. Whether this is true -- and what it means -- is unclear. Others have seen people go off therapy and observed only a small or modest rise in viral load to a level that reached or remained below pretreatment baselines. Also, ten unusual cases are being followed in the US and Europe of patients who have seen no return of viral load at all after going off therapy -- sometimes for as long as a year and half after withdrawing treatment. In short, there is no proven pattern here as to what is going on, nor does Levy claim there is. No clinical trial has yet reported on these kinds of data in a controlled, scientific fashion.

Some researchers even question the significance of the cases that Levy cites. It may be that the patient’s initial virologic response, after stopping therapy, is an increase in viral load beyond original baselines. Over time, however, this new viral load may decline to baseline again on its own. HIV-specific CD8+ cells may still lurk in the immune system, but it may take time for them to be brought back up to "fighting" levels. A single viral load test would tell us nothing. Still others point out that the populations of HIV-specific CD8+ cells that Dr. Levy is concerned with are simply absent in most people anyway, as a direct consequence of HIV infection. A number of studies have reported that the populations of HIV-specific CD4+ and CD8+ cells are among the first casualties of HIV infection, and it is their early loss that explains the failure of the immune system to successfully combat the disease. If so, Levy’s analogy to the high levels of such cells in long-term non-progressors may be irrelevant because such people represent a tiny minority (estimated at 3% or less) of HIV-infected people who are able to retain those special cells despite HIV. If so, their experience cannot be fairly compared to what happens to people who lost such cells early on and were subsequently treated.

The Bottom Line

In short, there are two ways to look at Dr. Levy’s views. The first is to take a hard look at precisely what he is recommending rather than getting hung up trying to interpret what he is arguing scientifically. When this position is taken, it becomes clear that he is making a recommendation about starting treatment that is only slightly more conservative than the standard Federal Guidelines. Rather than reacting to the real standards recommended by expert opinion in the Guidelines, he is arguing against the slogan, "hit it hard and hit it early." He is disputing the notion that everyone should be treated immediately simply because they are HIV-positive. In this regard, he is squarely in the middle of medical opinion. Only a minority of clinicians or researchers argues in favor of immediate treatment for everyone without regard for CD4+ cell counts and viral load levels, risk-progression tables or concerns about side effects and drug resistance. These are reasonable and important issues that should be addressed each time a patient and physician make a decision about treatment.

A second way to look at Levy’s perspective is to get hung up in the scientific debate he raises in describing his rationale for delaying treatment. Choosing this path is much less likely to lead to a satisfactory conclusion, because neither he nor those who most strongly dispute his views have hard data to prove their points. Each side is reading the tea leaves of modern research somewhat differently. Neither side has the right to claim proof. Available clinical trial data simply do not answer the question of whether people live longer, shorter or for the same net length depending on whether they start treatment at 400, 500 or even 1,000 CD4+ cells. Unfortunately, we may never get a hard answer to these questions since no one seems motivated to running the complex clinical trials needed to get a firm answer. The pharmaceutical industry is perfectly content with the current confusion, since it allows them to sell early treatment to anyone who will listen. Government doesn’t want to do the trials because they will cost a fortune and may ultimately prove impossible to run effectively. Thus, for now and for the near future, there will be no hard recommendation about precisely when to start treatment as a guideline for groups. On an individual basis, however, most physicians find it relatively easy to work out this decision with most patients, the kind of people who don’t lie in the vague gray zone debated by Levy. For most, the picture is clear and the tools are available to help one decide.

Nothing in Dr. Levy’s letter would suggest that people who have already started treatment should stop. For those first contemplating a treatment decision, Levy’s points serve as an important reminder that treatment is a serious decision that should be entered carefully and only when the patient is fully informed.

(For more information on this subject, get Project Inform's Antiviral Strategies Discussion Paper or the Mellor's Chart Quick Sheet from the Project Inform HIV/AIDS Treatment Hotline at 1-800-822-7422. The "Mellors’ Table" can also be found on page 28 of the Federal Guidelines. A copy of the Federal Guidelines can be obtained by calling 1-800-448-0440.)