| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Division of Pulmonary Medicine, Department of Pediatrics, University of Washington School of Medicine, Seattle, Washington
Correspondence and requests for reprints should be addressed to Margaret Rosenfeld, M.D., M.P.H., Children's Hospital and Regional Medical Center, 4800 Sandpoint Way NE, Seattle, WA 98105. E-mail: margaret.rosenfeld{at}seattlechildrens.org
ABSTRACT
Clinical trials in cystic fibrosis (CF) will continue to investigate a widening array of new therapeutics, ranging from gene therapy to antiinflammatory drugs and antiinfectives. A wide range of clinical trial endpoints is needed to adequately evaluate these agents. The existing "toolbox" of CF clinical trial endpoints includes pulmonary exacerbation rates, quality of life measures, growth, lung function, respiratory cultures, inflammatory markers, nasal potential difference, chest radiographs, chest computed tomography (CT), and newer imaging modalities. Clinical endpoints will be distinguished from surrogate endpoints; imaging is a surrogate endpoint. The qualities of an ideal outcome measure will be reviewed. The advantages and limitations of current endpoints will be identified, and the potential role for chest CT and newer imaging modalities discussed in this context. Finally, choosing appropriate endpoints for specific indications will be discussed.
Key Words: cystic fibrosis • endpoints • imaging
Clinical trials in cystic fibrosis (CF) range from small phase 1 studies of safety and pharmacokinetics to large phase 3 trials required for registration of new therapeutics. Similarly, they investigate a broad array of therapies, from gene therapy to inhaled antibiotics and antiinflammatory agents. Clearly, a wide range of outcome measures is needed to adequately evaluate these agents. Thus, the clinical trial endpoints in our current "toolbox" range from well-established measures, such as lung function, to emerging endpoints, such as chest computed tomography (CT) and newer imaging modalities (Figure 1). These endpoints must be viewed in the context of the important distinction between clinical endpoints and surrogate endpoints (1). Clinical endpoints are defined as direct measures of how a patient feels, functions, or survives, and include resolution or prevention of symptoms. Currently, there are only two clinical endpoints in the endpoint toolbox: pulmonary exacerbations and quality of life. Surrogate endpoints, in contrast, are defined as lab measurements or physical signs that are used as a substitute for clinical endpoints, and include the remainder of our armamentarium, including lung function, culture results, and all imaging modalities. Surrogate endpoints do not confer a direct clinical benefit to the patient, and may not correlate with clinical endpoints. The association between surrogate endpoints and clinical endpoints must be defined in order for them to be meaningful as outcome measures.
|
The ideal endpoint needs to be accurate (to measure what it's purporting to measure), precise (to measure the underlying quality with minimal error and minimal variability), and reliable (a measurement made twice should produce the same results). Surrogate endpoints must be biologically plausible, reflect clinical severity, and correlate with true outcomes (2). They should also be responsive, such that the measurement changes in the anticipated direction either with disease progression or with an intervention. There should be standardized equipment and techniques available with which to feasibly perform the measurement, as well as standardized interpretative strategies. Ideally, the endpoint would also be measured with minimal risk and be inexpensive and easy to perform. In addition, an endpoint that can be measured across all ages, such as through CT, offers distinct advantages.
CF clinical trial endpoints are compared in Table 1. Limitations of our existing surrogate endpoints include the fact that they are relatively insensitive to early disease and have a very limited ability to detect regional heterogeneity of disease. Many of these measurements cannot be made in uncooperative young children, and they assess function rather than structure. In children under 6 years of age, the endpoint toolbox is particularly limited. Young children cannot cooperate with spirometry, and their lung disease is generally mild, so that lung function and chest X-rays are relatively insensitive. Young children typically cannot expectorate sputum or cooperate with sputum induction, so respiratory cultures either have to be obtained by oropharyngeal swab, which has relatively poor diagnostic accuracy compared with lower airway cultures (3), or by bronchoalveolar lavage, which is invasive. The only CF-specific quality of life tool, the Cystic Fibrosis Quality-of-Life Questionnaire (CFQ) (4), is not available for patients under 6 years of age. Thus, very young children are precisely the population that may gain the most from the development of CT or other imaging modalities as surrogate outcome measures.
|
CHOOSING APPROPRIATE ENDPOINTS FOR A CLINICAL TRIAL
Clinical trial endpoints must be appropriate to the phase of the trial, the target population (in terms of disease severity and age), the duration of the study, the sample size, the number and expertise of participating sites, and, perhaps most importantly, the therapeutic aim. In phase 1 and 2 trials, surrogate endpoints are an acceptable primary endpoint, but in phase 3 trials, the U.S. Food and Drug Administration is increasingly requiring that the primary endpoint be a clinical endpoint (i.e., a measurement that demonstrates tangible benefit to the patient).
The target population is a very important factor in terms of deciding the most appropriate endpoint. First, in terms of disease severity, which measurements are expected to detect abnormalities in this stage of disease, and what is the anticipated effect size, particularly if the agent is to be evaluated in patients with mild disease? Second, are the target subjects old enough to cooperate with the measurement? Study duration also influences the choice of endpoint. Over what time period is the therapeutic effect anticipated, and how rapidly is the proposed endpoint able to detect change? For example, spirometry can detect a change in minutes after a bronchodilator, whereas the CF-specific quality-of-life tool, the CFQ, only assesses symptoms over the previous 2 weeks. Sample size is generally constrained, so choosing an endpoint that provides adequate power to detect the anticipated effect is critical. The feasibility of the endpoint must be assessed based on the number of participating sites, costs, and required expertise. For example, infant lung function testing would not be appropriate to a large multicenter trial because of the considerable expertise and training involved at each site. Finally, the therapeutic aim, to a great degree, dictates the most appropriate endpoints. A therapy aimed at primary prevention, such as gene therapy or protein-assist therapy, would use a different endpoint than one directed at secondary prevention, such as an antimicrobial or an antiinflammatory agent. This paradigm of different endpoints for different therapeutic aims is outlined in Table 2 for antiinfective therapies. For each therapeutic aim, the target populations, endpoints, and timeframe for observing an effect is summarized.
|
Chest CT and newer imaging modalities have a potentially important niche in the endpoint toolbox for clinical trials in CF. They are likely to play a critical role in terms of the assessment of early and mild CF lung disease, and are the only endpoints capable of detecting structural damage and regional heterogeneity of lung pathology. Importantly, CT can be performed across all ages, although it does require sedation in the youngest children. A number of challenges remain, all of which should be surmountable, including: standardization of the procedure and of scoring; better assessment of the accuracy, reliability, and responsiveness of the measures; better understanding of the association of imaging endpoints with clinical outcomes; and further assessment of feasibility in a multicenter setting. Ongoing issues of cost and radiation risk will likely mean that CT is chosen for trials in which it is the endpoint that is clearly most likely to demonstrate a response to an intervention.
FOOTNOTES
Conflict of Interest Statement: M.R. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form November 29, 2006; accepted in final form February 26, 2007)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
