Sunday, April 24, 2011

The Safety and Effectiveness of Revlimid

A few weeks ago, the FDA issued a safety announcement concerning Revlimid, indicating its concern that longer-term exposure to the drug increases the risk that the patient will develop additional hematological malignancies (translation: leukemias and lymphomas). This isn't exactly shocking new news, but it does betoken a heightened level of concern and awareness of the problem. Of course, in my case, "longer-term exposure" is exactly what we are hoping for, since the idea is that I will continue to take Revlimid as long as possible, until it stops working.

But we already knew that the answer to the "how safe" question is "not very". Lenalidomide (Revlimid's chemical name) is a derivative of thalidomide, which was originally introduced in the 1950's as a sedative. Thalidomide was prescribed to large numbers of patients, among them pregnant women, who proceeded to produce tens of thousands of babies with severe birth defects, until the drug was withdrawn in the early 1960's. It was one of the most spectacular failures in the history of the pharmaceutical industry. But in the 1990's, thalidomide was resurrected as a chemotherapeutic agent effective against none other than our old friend, multiple myeloma. Subsequently a derivative, lenalidomide, was introduced for the same purpose, but with better responses and fewer side effects for many patients. Today, both thalidomide and lenalidomide are being prescribed for myeloma, but they are distributed in a tightly controlled fashion. Every 28 days my oncologist generates, and the insurance company approves, a prescription for Revlimid capsules, quantity 21, in 25mg strength. Eventually I am contacted by the specialty pharmacy, and I receive the same "counseling" from a pharmacist every time: I am warned against donating blood or sperm, and against engaging in sex with "a woman who is pregnant, or who could become pregnant", unless I use a condom. The pharmacist then transfers me to a patient service line at Celgene, the manufacturer of Revlimid, and I take a "survey", which is the same every time: I answer questions about whether or not I have donated blood or sperm, and whether or not I have engaged in unprotected sex with "a woman who is pregnant, or who could become pregnant". Only after I have successfully jumped through all of these flaming hoops can I be shipped the goods.

Beyond this, Revlimid carries warnings regarding, among other side effects, deep vein thrombosis (blood clots) and tumor lysis syndrome (TLS), which covers a grab bag of metabolic disturbances, the most dangerous-sounding of which is acute renal failure.

So Revlimid is a dangerous drug, maybe more dangerous even than we were thinking. Given the risks, definite and possible, should I be "allowed" to take it? In the binary world of mass media health and science reportage, drugs are either "safe" or "unsafe". No gray areas, please. No nuances allowed. And the FDA is the God of this Manichean religion, sitting in judgement over which may enter the Heaven of Safe Drugs and which are consigned to the Hell of Unsafe Drugs. For many drugs, the end of the line comes in the Phase I clinical trial. The Phase I clinical trial has nothing to do with assessing the efficacy of the drug under test. The question "Does Revlimid do anything good for multiple myeloma patients?" was not asked, let alone answered, in the Phase I clinical trial. The only objective of the Phase I clinical trial is to determine whether or not the drug is going to hurt or even kill people. Those laundry lists of "possible side effects" that come with prescription medications are generated by the Phase I clinical trials. If a patient in the non-control arm of the trial (the group of patients actually getting the drug under test, rather than the placebo) gets a nose bleed during the trial, then "nose bleed" is going to appear in the list of "possible side effects" for this drug -- whether or not the drug actually had anything to do with the nose bleed. Because, in fact, there's no way to know for sure. And if patients in the non-control arm die during the trial? Then the drug is pretty much done for. And in fact quite a few promising drugs have not survived their Phase I trials for precisely this reason. Pharmaceutical industry professionals of my acquaintance assert, with the utmost seriousness, that should it have been subjected to the FDA's approval process, aspirin would not be available in the United States today. It's simply not "safe".

And given its litany of problems, no doubt many people would be surprised to learn that the FDA has approved the use of a drug such as Revlimid for any purpose whatsoever. Probably, it is flying under the radar: few are the patients who want or need to use it, and these few are sufficiently desperate to resort to it in spite of the dangers. In all probability, they are more likely to die without it than with it. So the all-powerful FDA has allowed this, calculating that repercussions will not return to haunt it. But now it's getting nervous about it, as the safety announcement shows.

There is no way that a mass-market drug like Lipitor or Viagra would be permitted to reach the market with a rap sheet as long as Revlimid's. That sounds right, except that there are bound to be gray areas in which certain drugs could benefit a relatively large number of patients, but the FDA has ruled that out, because it doesn't want to be raked over the coals by the media and by Congressional sub-committees later, if it guesses wrong. Better to keep this stuff away from the unwashed, who don't know what they're missing anyway. The FDA is rarely made to suffer for keeping dangerous drugs out of the hands of people who might be helped by them.

Which brings us to the topic of "effectiveness", which is the subject of the later-phase trials. This is where the gray areas become very murky indeed. Technically, Revlimid is approved, in combination with dexamethasone, only for myeloma patients who have failed at least one prior therapy. I have now been prescribed the drug not once but twice in circumstances that do not fit this description: once with dexamethasone as induction therapy for a stem cell transplant; and once without dexamethasone as post-transplant maintenance therapy. When this inconvenient truth is brought to their attention, oncologists merely shrug. This kind of thing is routine, universal practice in American medicine. Once a drug has been approved for any purpose whatsoever, it will never again be approved for any other, irrespective of any subsequent research that may support such usages. Formal clinical trials done in support of an application for FDA approval are so expensive and so drawn out that no one ever does more than one set of them for a given drug, if they can avoid doing so. Physicians feel free to dispense a drug for unapproved uses if current research supports them. But they, not the FDA, make these decisions. Hence my experiences with Revlimid to date. As one of the aforementioned pharmaceutical industry professionals commented to me: "You're in the real clinical trial, right now."

The thalidomide disaster gave birth to the FDA's existing approval process for drugs. But the problem with thalidomide wasn't the absence of this approval process, the problem was a lack of adequate testing. There can be no doubt that drugs like Revlimid need to be tested and retested, early and often. What I question is the FDA's power to arbitrate what drugs can or cannot be made available as a result of these tests. Why aren't the results of drug tests simply made available to the physician community, so that they can decide whether and how to use the drugs? They are doing this job anyway, for drugs that have managed to successfully negotiate the FDA's highly politicized approval process, as my experience with Revlimid shows. To me, the FDA's control of this process appears to be less about assuring safety and efficacy than it is about control.