Press room ... 1997 archive

Treatment Failure in the “Real World”

September 30, 1997

San Francisco, CA—Media reports from the recent ICAAC conference in Toronto are beginning to focus on the evidence of “drug failure” being reported at San Francisco General and a few other sites. Just as earlier media reports made the protease inhibitors sound like a cure and invited excess optimism, the new reports tend to make the drugs sound like a complete failure and invite undue pessimism. The truth, as always, is not to be found in either extreme.

Steven Deeks at San Francisco General Hospital (SFGH) is reporting that some 53% of his patients (in a group of 136) have now “failed” on protease inhibitors. This number apparently includes some that never responded in the first place, as well as a majority who responded initially but has now seen a return of detectable virus. In contrast, some other researchers at the same meeting in Toronto reported that they were seeing surprisingly high rates of continued success, often beyond their expectations.

Even at its worst, this data is not unexpected. Those who have paid attention have known all along that the current drugs are not a cure, even in the best of circumstances. We’ve known all along that they work much better in people who are just beginning treatment as compared to those who have been using therapy for years. It is obvious that this should be the case since people who have extensive prior treatment experience are almost certainly resistant to at least some of the medications they are using. We’ve also known that the drugs presented difficult challenges of adherence and any significant deviation from the prescribed program could contribute to hastened drug resistance. And it has been clear from the beginning that toxicity could interfere with peoples’ ability to use the drugs properly.

Considering all these things, it’s actually surprising that the drugs have help up as well as they have in so many people. It remains an undisputed fact that they have reduced the death and disease rates dramatically, as virtually any AIDS hospital or medical practice can confirm. By any standard, they remain far superior to anything that has come before.

What is Meant by “Treatment Failure?”
Before everyone gets unduly depressed about imminent drug “failure,” it’s important to take a careful look at how “failure” is defined in these studies. In the Deeks/SFGH study, “failure” simply means that a person using the drugs has either continuing or reappearing measurable viral replication. In other words, 53% either failed to become “undetectable” in the first place, or they initially became “undetectable” but have since seen some return of measurable viral load. This is a very narrow and perhaps misleading definition of “failure,” especially when it is accompanied by record low rates of new opportunitistic infections and death. “Failure” here makes no distinction between someone whose viral reached 800 and another whose viral load soared to 8 million. It also doesn’t tell us whether the patient was later able to go on to some different or more aggressive therapy that brought viral replication back under control. And it tells us nothing about the patients’ actual clinical condition.

It’s very unclear just what the clinical consequence is of seeing a return to measure viral load. A good deal of evidence already suggests that it is unlikely to be accompanied by an immediate return of clinical symptoms or a loss of the improvement experienced in measures of immunity. It just means that the viral load has moved above a certain arbitrary threshold.

One might also ask whether there is anything different or unique about the SFGH cohort. For example, it would be important to know what percentage of the people in the study began their first use of a protease inhibitor with saquinavir back in 1995. A separate study presented at ICAAC reports that prior use of saquinavir is the clearest predictor of failure for people using indinavir. We also know that most patients in most of the SFGH clinics were at least briefly put on full dose ritonavir back in March of 1996. Almost all switched to something else within the next two months because of toxicity, which was also contributing to serious levels of non-adherence. Also unclear is whether the population at SFGH is different from the average HIV positive population—were those in Dr. Deeks’ cohort more or less likely to be treatment naïve? Is there any reason to believe they were more or less likely to adhere to treatment requirements? Were they more or less likely to get all the care they need and able to see their medical support team often enough to support full adherence? We don’t know the answers to these questions, but they could all affect the “failure” rate. We do know, however, that Dr. Deeks has acknowledged that many of the people experiencing drug “failure” had had difficulties with adherence.

We also know the answer to some general questions about treatment failure and resistance from data reported at the Resistance meeting in the summer of ’97 in Florida. Essentially, it showed that if people achieved “undetectable” levels of virus on the new super-sensitive assays (measuring down to 50 or 20 copies), then they would be very likely to have a long and successful run on their treatment regimen, probably measured in years. In contrast, those who never reached “undetectable” in the first place or, surprisingly, even those who only reached it on the standard viral load tests at the limit of 500 copies, were very likely to experience rising viral load on their regimen with 6 to 12 months. The ability to reach “undetectable” status on either version of the test was greatly affected by the patient’s baseline viral load. If it was low to begin with, it was easy to keep it undetectable for long periods. The higher it was at the start, the tougher it was to reach “undetectable” status.

What does all this mean to people who are counting on these drugs to give them hope of a future? A few questions many people find themselves facing, and a few possible answers:

I’m doing OK for now, but does this mean that I am doomed to see my drugs fail? How soon?
No treatment yet known for HIV disease is likely to last for a normal lifetime. Until we can completely suppress viral replication, drug resistance is still likely to contribute to treatment failure over time. How long it can be suppressed is anybody’s guess. When people achieve undetectable levels on the new super-sensitive assays, the durability of treatment is greatly enhanced. People concerned about their own situations might check around to see if they can get access to the super-sensitive test. But for some, this may not be necessary. Recent research suggests that if people go for a year or longer at the “undetectable” level on the standard test, it’s very likely they are also “undetectable” on the super-sensitive test. Otherwise, they are likely to have failed within the first year.

What’s going to happen when the drugs fail?
We really don’t know. We don’t know what it means when measurable viral load returns or how quickly the immune system might be compromised again. Even if we assume that all possible options have been tried without success, and no new drugs bring viral load under control, it’s not at all clear that people will simply slide downhill clinically. As a general rule, we already know that this is not happening, since the levels of return of clinical illness and new OI’s are nowhere near as high as the reported incidence of “treatment failure”. There is plenty of evidence that even temporary success in suppressing viral load can lead to serious levels of immune restoration. In fact, many scientists report being pleasantly surprised by the degree of renewed immune response being seen. The longer suppression is maintained, the larger this response seems to be. Whatever the gains, they seem unlikely to disappear overnight, perhaps thus providing a cushion of time while awaiting new and better therapies.

Some virologists also believe that the highly mutated form of virus that evolves to escape the drugs is in many ways not as “fit” or destructive as the original, “wild type” virus that most people started off with. In other words, the virus has to pay a price, perhaps a high price, for accumulating all those mutations along the way. It may still be capable of replicating, but it may be diminished in other ways, such as its ability to damage the immune system. At worst, things would only be back to where they were before the advent of the protease inhibitors, with new treatments on the way.

When these drugs fail, will anything else help?
Yes, there is reasonable hope that other things may help. First, we needn’t always think of resistance as a simple, on-off thing. It comes in degrees ranging from high level, which will completely cripple a drug, to low level, which might only require changes in dosage or companion drugs. Some degrees of resistance might be modest enough to be overcome by a quick change to more aggressive therapy, such as dual protease inhibitors or combining protease inhibitors with newer non-nucleoside RT inhibitors (nevirapine, delavirdine, Sustiva). Sometimes resistance can be overcome simply by finding any two or more new, highly potent drugs that aren’t cross-resistant. And some times, viral load only reappears temporarily, then becomes undetectable again probably in response to some secondary infection.

Beyond these points, there are a number of new drugs coming, some of which might be useful despite current resistance. For example, one new protease inhibitor in early clinical trials from is very different from all current protease inhibitors and offers the potential of being active despite resistance to the earlier drugs. Also, several new classes of drugs are now in clinical trials. These include Sustiva (a powerful non-nucleoside RT inhibitor), T-20 (a fusion inhibitor), integrase inhibitors, zinc-finger inhibitors, etc. Some new research is also re-emphasizing the potential of hydroxyurea. Of course, there’s also the whole field of immune restoration to look to.

One important ray of hope came from another study presented at the ICAAC meeting. It showed that, although the longest duration of success came with “undetectable” viral levels, this goal was not essential for preventing short-term progression to AIDS. Instead, this study suggested that the ability to prevent short-term progression simply depended on lowering the viral low and keeping it below baseline.

Is there a bottom line here?
If so, it’s to re-emphasize the critical importance of adherence and using therapy aggressively enough to fully suppress viral replication. This is the only proven way to get the greatest possible benefit from the available therapies. Another bottom line is to re-state the importance of the principles of prevention, which have sometimes been left on the sidelines in the wake of talk about “undetectable virus.” Perhaps this new public discussion will encourage people and the media to take a more realistic look at where we are in the fight against AIDS and to adjust behaviors accordingly. Euphoria over the success of treatment has had an unfortunate side effect in diminishing the sense of threat people felt from HIV infection. This has lead to an increase in risky behavior among the uninfected, and to the irresponsible trend toward “barebacking” among a small faction of the HIV positive. We must all do what we can to help the public and the media achieve a realistic perspective on where we are and what we need to do to make the best of a still imperfect situation. Finally, we must get on with the business of AIDS activism to make sure that better, safer, and easier to use therapies become available as quickly as possible.