| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Chicago IL
Dr. Gross is Professor Emeritus, Department of Medicine, Stritch School of Medicine, Loyola University of Chicago, and Staff Physician, Hines VA Hospital.
Correspondence to: Nicholas J. Gross, MD, PhD, PO Box 1485, Hines, IL 60141; e-mail: Nicholas.gross{at}med.va.gov
Many lessons can be drawn from the article by Tashkin and Kesten in this issue of CHEST (see page 1441), but I will limit myself to just two. The first is that, yet again, it has been shown that patients with COPD do not have "irreversible" airways obstruction although many medical textbooks and even some guidelines continue to propagate this myth many years after it was shown to be untrue, as I argued in an editorial of 1986.1 What is true is that the bronchodilation that patients with COPD are capable of is less than that of asthmatics, on average. But the overlap in bronchodilator responsiveness between the two groups is so considerable that it is not possible to differentiate between them on this basis.2 When, as in the two studies they report on, a bronchodilator as potent as tiotropium is used in a typical COPD population (subjects were not selected on the basis of their prior response to a bronchodilator), the mean improvement in FEV1 can be as high as 25% over baseline, or roughly 250 mL (derived from Table 1 of the article by Taskin and Kesten). Moreover, half the patients in these studies, the "responder" group, obtained a mean FEV1 improvement of 400 mL. These are not small amounts; they compare quite favorably with those obtained by lung volume reduction surgery.3
It cannot credibly be argued that, because of their substantial bronchodilator responses, half of these patients must have been occult asthmatics. In the first place, anticholinergic agents generally do not provide much bronchodilation in asthmatic patients. But more importantly, the patients in these trials met the current diagnostic criteria of COPD4 5 : a mean age of 65 years, a 60–pack-year smoking history, dyspnea as the dominant symptom, persistent airways obstruction, and the absence of any prior diagnosis or clinical features of asthma. Clearly, these are patients to whom any practicing physician would give an operational diagnosis of COPD, and at least half of them had a "significant" bronchodilator response the first time they were tested.
I qualify the word significant thus because the conventional requirements for a significant bronchodilator response—an improvement of at least 12% and at least 200 mL over baseline—are arbitrary. There are some but not many data that show a difference in any outcome between patients who fall on one or the other side of that arbitrary spirometric dividing line. My implication is that it is an open question whether a bronchodilator response should even be considered as a prognostic feature of the clinical value of a bronchodilator in COPD.
Does the absence of a bronchodilator response on a single challenge predict that the patient is not capable of bronchodilation? No. The proportion of patients with COPD who show a significant bronchodilator response increases cumulatively with the number of times the test is performed, as has also been shown many times.6 7 8 Approximately 80% of patients with COPD will exhibit a significant bronchodilator response over the course of three separate challenges, so a single bronchodilator challenge is not a reliable assay of responsiveness. Besides this, there may be benefits that are not measured by the FEV1 response itself, as in the present study and several previous studies.9 10 Patients who are "poor responders" in terms of their first FEV1 response to an agent may nevertheless experience significant symptomatic relief and improved quality of life from regular use of the same agent. I agree with the conclusion of Tashkin and Kesten that "the test does not accurately predict whether benefits may be achieved with maintenance ... therapy and therefore should not be used to guide decisions to prescribe a bronchodilator therapy."
The second point concerns how one should evaluate the benefits of a treatment for COPD. At a symposium of the American Thoracic Society Annual Meeting in 2001, Dr. Bartolome Celli remarked on our "schizophrenia" in defining COPD as a disease of irreversible airways obstruction and then using the improvement in FEV1 to assess the benefit of a treatment. I hope the above remarks reassure Dr. Celli that the profession is not in need of psychiatric help. It is the definition that is out of touch with reality, but he does have a point. By tradition as much as for any other reason, pulmonary function has been a benchmark of the therapeutic efficacy of drugs in COPD—FEV1 being the most reproducible and available single test for both acute and long-term outcome studies.
The main problem with using the FEV1 is that it correlates only poorly with clinically relevant outcomes. Jones et al11 have shown that it bears little relationship to quality of life. In acute studies, it has been quite difficult to show that an improvement in FEV1 correlates with an acute improvement in symptoms such as dyspnea or exercise tolerance12 that one certainly would expect. Why else would patients use (and sometimes overuse) bronchodilators when they are dyspneic? The explanation for this paradox may be that the FEV1 is a maneuver that does not correspond to any common breathing activity except perhaps sneezing or blowing out candles. It is a rather poor if convenient surrogate. One is reminded of the old joke about the man who lost his wallet in an alley. When asked why he was looking for it the next street, he replied "because the light is better in the street."
Would some other test of pulmonary function be a better surrogate for the clinical efficacy of a bronchodilator, or other agent, in COPD? Perhaps the reduction of hyperinflation, eg, inspiratory capacity?9 We simply do not know.
The regrettable fact is that we do not have a reliable, convenient surrogate marker for the clinical benefit of any agent in COPD. I do not advocate that we should reject the FEV1 response as an index of the benefit of a bronchodilator (it is certainly an important test in clinical practice); however, I do suggest that it should never be the only outcome in a clinical drug study. Studies of new therapies for COPD should include, as end points, at least one and preferably several clinically relevant outcomes. There are now several appropriate, specific, and well-validated clinical outcomes. Instruments such as measures of quality of life, dyspnea, and effort tolerance have the advantage that they are clinically relevant to COPD and can detect changes over quite short periods of time. Their drawback is that they are time consuming and subject to the skill of the experimenter and the cooperation of the subject. Their noise-to-signal ratio is fairly high, so their sensitivity to subtle changes is low.
Other important clinical outcomes include the use of health-care resources such as unscheduled visits to doctors or emergency departments, acute exacerbations of COPD, and mortality. These events are highly relevant to both health and health-care costs. They have the advantage that each can be captured with confidence although they remain susceptible to some subjectivity. How certain can one be that the event was related to COPD rather than to one of the comorbidities that are common in the COPD population? Other disadvantages are that these events, particularly mortality, are relatively uncommon, so a large and expensive trial over a fairly prolonged period of time may be required to detect a significant treatment effect with statistical confidence. One is mindful, too, that a treatment can, in theory, have benefit by improving quality of life, for example, without increasing survival, and vice versa.
For all of these reasons, a wide range of outcomes, including both physiologic and clinical end points, is needed before one can understand whether and how a novel therapeutic agent may be of clinical benefit to patients. Despite the above-mentioned difficulties, it is good that many current trials include clinical end points.
The report of Tashkin and Kesten addresses some of these concerns. Apart from an improvement in trough FEV1—the primary outcome—clinically relevant outcomes such as quality of life and dyspnea were significantly better in the treatment group, which increases one’s confidence that tiotropium was clinically beneficial.
One must draw attention to the impressive effects of tiotropium. It is probably the most potent bronchodilator yet for COPD. In part, its potency may be attributed to its unique selectivity for M1 and M3 muscarinic receptors, a feature that ipratropium and all other available antimuscarinic agents lack. With a nominal duration of action of 24 h, and probably much longer,13 tiotropium is by far the longest-acting bronchodilator of any class. Apart from the convenience of once-daily dosing for patients, the long duration of action ensures that significant bronchodilation is maintained around the clock. If patients use tiotropium consistently, their FEV1 would never fall to baseline. Out of interest, one can calculate the potential magnitude of this effect in relation to the natural decline in lung function from data provided in Table 3. The trough FEV1 increased by a mean of 115 mL in the group that received tiotropium for 1 year, while that of subjects who received placebo declined by 38 mL, giving an overall treatment effect of 153 mL. If the mean annual decline in FEV1 of this population is 38 mL, an amount which accords with previous reports,14 15 then an increase in trough FEV1 of 153 mL due to tiotropium use would correspond to the amount the FEV1 would otherwise be expected to decline in 153/38 years, approximately 4 years. It would be as if their FEV1 had been restored to its level 4 years previously. This calculation is quite speculative, of course, and it cannot be said that patients will survive 4 years longer if they use tiotropium; there are no data on the effects of tiotropium on survival at this time. Nor are there definitive data that use of tiotropium decreases the rate of decline of lung function or the frequency of important events such as acute exacerbations; such studies are ongoing. But the data do suggest that tiotropium will become an important addition to our therapy for this common and highly symptomatic disease.
Footnotes
Dr. Gross has received research support from Boehringer-Ingelheim, and is on their speaker’s bureau.
References
This article has been cited by other articles:
|
|
 |
|
  D. W. Mapel Treatment Implications on Morbidity and Mortality in COPD Chest, August 1, 2004; 126(2_suppl_1): 150S - 158S. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
