For September 27, 2003
DOES TRAVELING FOR TREATMENT INCREASE SURVIVAL? PART TWO
The patients in Dr Lamont's study all had head-and-neck cancers and were all enrolled in Phase II clinical trials at the University of Chicago Cancer Center. Dr. Lamont reported that patients who lived further away from the treatment center, and therefore had to travel further to receive treatment, fared better in terms of overall survival than those who lived closer and had only local traveling to do. In fact, if they traveled 15 miles or more they had only one-third the risk of dying as compared to those who lived closer to the cancer center. This finding was particularly surprising since the beneficial effect of travel remained significant even when patients' age, stage of disease, performance status and income level - all of which can influence survival - were taken into account. One is left with the inescapable conclusion that the distance traveled for care - which is rarely reported in the analysis of trials - is a potentially important variable, which probably reflects some other more fundamental but as-yet-unmeasured factor. The study has profound implications for the way that new cancer treatments are evaluated and approved. According to Duke University biostatistician Dr. Stephen George, writing in an accompanying JNCI editorial, "this conclusion is unsettling because it suggests that other unknown or unmeasured variables have important prognostic impact." This is an important reminder, he says, of the degree to which various known or even unknown selection factors can influence the outcomes of clinical trials. After all, if clinical trial enrollment is restricted to patients who come from a distance, then the results might turn out to be "impressively positive." On the other hand, if the trial is restricted just to local patients, then the results - using exactly the same protocol - might be "discouragingly negative." Here we have a prime example of what statisticians call a "selection bias" (i.e., a testing error introduced by the enrollment process itself), which, in Dr George's words, "can seriously damage" the validity of a trial's conclusions. The results from nonrandomized trials in general, and Phase II trials in particular, "depend heavily on the specific characteristics of patients studied, including the known and unknown variables." So, how does this insight affect our understanding of the value of clinical trials? Simply put, says Dr. George, patients enrolled in clinical trials "often bear little resemblance to the larger population of patients." That is because people who take part in clinical trials are not identical to the cross-section of people in the general population who will ultimately get the new treatment. They are different in many ways, both known and as yet unknown - and how far they have traveled to receive treatment may be one of the most important factors ever to be routinely overlooked in clinical trial design. In fact, it is often the case with well-publicized new treatments that highly motivated patients, who have heard about the exciting new "breakthrough" from friends, the media or the Internet, will move heaven and earth - including traveling long distances - to obtain the treatment. Perhaps due in no small part to the fact that they have fought to enroll in the trial and have traveled from afar to obtain the treatment, they do well on it. The results of Phase II trials involving such motivated patients end up looking so promising that the treatment gains FDA approval on the basis of this outcome, bypassing more stringent testing. Once the treatment is approved, however, and goes on to be prescribed to a wider and less selected population, it no longer works so well, if at all. But the world (or at least the USA) is now stuck with this new treatment. Rigorous randomized controlled trials (RCTs or Phase III trials), performed by many hospitals together, used to be required for new drug approvals, but more and more these days, favorable and smaller Phase II trials are being substituted, effectively circumventing this requirement. As the Lamont study so eloquently demonstrates, a Phase II trial may have a favorable outcome not because the drug in question is intrinsically better than previous treatments but because the patients enrolled in the trial are motivated 'distance travelers' whose will to live may be the source of their prolonged survival. Even RCTs are not immune from selection bias, says Dr. George, but at least "they have the strong advantage that the treatment comparisons are unbiased because of the randomization." Large multi-center trials have been very good at cutting exaggerated claims down to size...which is one reason that far fewer drugs were approved when RCTs were required. In nonrandomized Phase II trials, such as Dr. Lamont analyzed, a relatively small number of patients are given a new treatment but there is no mechanism for randomly assigning those patients to receive either the treatment in question or some competitive approach (such as either state-of-the-art therapy or the best supportive care currently available). The FDA now approves a growing number of cancer drugs in this way, using its so-called "accelerated approval" process. This mechanism was first used in 1995 and allowed approval based on what was called a "surrogate endpoint." A surrogate endpoint is a change in the patient's condition that may be of minimal clinical significance, but which is considered reasonably likely to predict ultimate patient benefit, such as increased survival (Hirschfeld 2002). For instance, the temporary shrinkage of a tumor by 50 percent or more, for a month or more, is a commonly used "surrogate" that has been accepted as a substitute for actual proof of life prolongation. In actuality, however, a temporary tumor shrinkage (called a "response") does not necessarily correlate with increased survival. Notice that accelerated approval requires no clear evidence that patients actually benefit from a new treatment. Approval is based on evidence that is by its very nature "less than sufficient to grant full approval," says Dr. George. Yet "[t]he majority of the more than 20 accelerated approvals granted to date have been based on the results of Phase II trials." The rationale for accelerated approval is "to provide rapid access to potentially effective agents in serious or life-threatening diseases when no other effective therapies exist." Although this altruistic-sounding rationale understandably pleases some patient advocates, and certainly pleases the drug's sponsors, the truth is that many of the treatments that are approved by the accelerated process are toxic and dangerous. When the FDA lowers its own standards in this way the risk of it approving "toxic placebos" is greatly increased. For all we know, patients' lives may actually be shortened, rather than lengthened, by taking these inadequately studied treatments. These developments in oncology are part of a trend at the FDA in recent years. A December, 2000, investigative series on this subject by the Los Angeles Times staff points to at least seven instances in which FDA laxity has led to harm or even to the death of patients. "Once a wary watchdog," said the L.A. Times report, "the Food and Drug Administration set out to become a 'partner' of the pharmaceutical industry." This report makes for very scary, but still timely, reading. To view the Los Angeles Times articles click or go to:
Technically, FDA's accelerated approval regulations require the
completion of more definitive studies that either confirm or deny
actual patient benefit, and don't just rely on such arbitrary, stand-in
endpoints. There is also the possibility of the FDA rescinding its
approval if the sponsor (usually a big pharmaceutical company) fails
to swiftly complete these confirmatory studies with 'due diligence',
or if the subsequent results do not live up to the initial promise.
However, as Dr. George notes, "to date, no accelerated
approval in oncology has been withdrawn...."
|
|||||||
|
| Dr. Lasagna |
I note with sadness the passing, last month, of Dr. Louis Lasagna, Dean Emeritus, Sackler School of Graduate Biomedical Sciences and founder, in 1976, of the Tufts Center for the Study of Drug Development. He died of lymphoma on August 6, 2003, at the age of 80.
His career in medicine spanned five decades and he was the author of nearly 250 PubMed-listed journal articles on a wide range of topics relating to new drugs. He was a pioneer in understanding the placebo effect and helped expose the ineffectiveness of many supposedly active drugs. In 1954, Lasagna published his classic paper, "A study of the placebo response," in the American Journal of Medicine.
In the late 1970s our paths crossed in a very unexpected way. I had recently been fired from Memorial Sloan-Kettering Cancer Center (MKSCC) for my public dissent over the hospital's handling of the testing of amygdalin (laetrile). To say that I was persona non grata in conventional medical circles would be an understatement. I was out of a job and working on a book that no one seemed to want to publish.
Out of the blue, I received an invitation from Dr. Lasagna, then chairman of Pharmacology and Toxicology at the University of Rochester School of Medicine and Dentistry, to contribute a chapter to a medical textbook that he was editing! The book was called "Controversies in Therapeutics," and its unique format was to present arguments for and against various disputed questions.
My contribution was called "In Defense of Laetrile." The book appeared in 1980 and was published by W.B. Saunders, a major medical publisher. This essay later found its way, in a modified form, into The Cancer Industry (which was published, as The Cancer Syndrome, by Grove Press in that same year).
This essay may have been the sole instance of a serious defense of amygdalin being published in the peer-reviewed literature during all those years of controversy. Although I was far from being a "laetrile advocate," I did feel that its use should be a matter of scientific discussion, and I was appalled by the crude mudslinging that often took the place of rational discussion on the subject. I also drew on the knowledge of my colleagues Robert G. Houston, Ernst T. Krebs, Jr., and Michael Schachter, MD, to put together a credible defense of the theory behind its use.
Most other academics had flat-footedly taken their stance against this "fraudulent" therapy. But Dr. Lasagna saw it as an open controversy. He stated his belief that "a perverse characteristic of the human race is its preference for spurious simplicity as opposed to truthful complexity." Nothing better described the politicized atmosphere in medicine at the time.
It was also a great experience for me personally to suddenly find myself in the company of more than 60 distinguished scholars in the field of medicine. It restored my faith in the possibility of a fair hearing for controversial alternative treatments. As I scanned the list of contributors I was stunned to realize that I was the only one without an academic, governmental or drug company affiliation. Most were medical school professors. I was listed simply as "Editor, Second Opinion," i.e., the underground cancer newsletter that I had edited in my Sloan-Kettering days!
Much has changed since 1980, but many of the questions that Dr. Lasagna posed in his book are still open. This includes the ultimate value (and mechanism of action) of amygdalin in cancer and, I would say, the more philosophical questions of how to decide whether any new treatment is really effective and worth taking.
Robert Temple, MD, an associate director of the Food and Drug Administration, has said that Dr. Lasagna was a leader among those clinical pharmacologists who, during the late 1950s and early 1960s, "started to point out that our data to support the efficacy of drugs was utter [rubbish]."
Although he didn't use the term, Lasagna was in fact an early advocate
of the "friendly skepticism" that I think is
the proper attitude to take towards any new or controversial treatment.
A person of such intelligence and integrity as Lou Lasagna comes
along only rarely. His probing and courageous examination of contemporary
medical practice will be sorely missed.
--Ralph W. Moss, Ph.D.
References:
Lasagna obituary: http://www.biomedcentral.com/news/20030908/02
FDA: U.S. Food and Drug Administration. Oncologic Drugs Advisory Committee Briefing Document. September 24, 2002. At: http://www.fda.gov/ohrms/dockets/ac/02/briefing/3894b1.htm
George, S. L. (2003). Selection Bias, Phase II Trials, and the FDA Accelerated Approval Process. J Natl Cancer Inst 95:1351-1352. http://jncicancerspectrum.oupjournals.org/cgi/content/full/jnci;95/18/1351
Hirschfeld S, Pazdur R. Oncology drug development: United States Food and Drug Administration perspective. Crit Rev Oncol Hematol 2002;42:137-43.
Lamont EB, Hayreh D, Pickett KE, Dignam JJ, List MA, Stenson KM, et al. Is patient travel distance associated with survival on Phase II clinical trials in oncology? J Natl Cancer Inst 2003;95:1370-5
**NOTE** To view this page in a more printable format, please CLICK HERE.
IMPORTANT DISCLAIMER
The news and other items in this newsletter are intended for informational purposes only. Nothing in this newsletter is intended to be a substitute for professional medical advice.
|
||||
