Alzheimer's Association Research Roundtable Highlights Scales as Outcomes Measures for Alzheimer Clinical Trials
Riveting discussions of the pros and cons of various scales as outcomes measures of Alzheimer's disease clinical trials provided just a taste of the activities at the Alzheimer's Association Research Roundtable in Washington, D.C., April 29–30. Attendees also heard the viewpoints of regulatory officials from the U.S. Food and Drug Administration (FDA) and its European equivalent, the European Medicines Agency, and could not miss the plea from William Potter, MD, PhD, of Merck Research Labs., who, speaking on private-public partnerships in biomedical research, urged attendees, “Stop competing across your scales! Come to closure. The need is great,” suggesting that continued debate regarding scales can slow advances in science. Dr Potter specifically pointed to the need to reach consensus on cognitive scales and functional outcome measures for use in the Alzheimer's Disease Neuroimaging Initiative, which is facing possible renewal.
Begun in 2003 with four members, the Research Roundtable consortium now includes more than 20 corporate members from the pharmaceutical, biotech, imaging, and cognitive testing industries, each of whom sends several senior scientists to the Roundtable to benefit from the collegial interactions and networking opportunities available. Additional guest attendees include those from academia, regulatory agencies, and the National Institutes of Health.
The mission of the Roundtable is to facilitate the development and implementation of new treatments for Alzheimer's disease by collectively addressing obstacles to research development, clinical care and public health education. This spring's Roundtable was co-chaired by Ronald Black, MD, senior director, clinical research, Wyeth Research, and Yaakov Stern, PhD, professor of clinical neuropsychology, Cognitive Neurosciences Division, Taub Institute, Columbia University.
Working with a 12-member subcommittee, Drs Black and Stern prepared a two-day program addressing current and future scales as outcomes measures for Alzheimer's disease, including cognitive, functional, and global assessments. As disease-modifying medications become available for Alzheimer's, scales will play a key role in identifying patients who might benefit from medications as well as assessing treatment efficacy. Topics included methods of analysis for existing scales and scale development; measures sensitive to change in early Alzheimer's and “clinical meaningfulness”; computerized measures; measures useful for early phase trials; patient- and caregiver-reported outcomes; and data-sharing in existing data sets.
William H. Thies, PhD, Alzheimer's Association vice president for medical and scientific relations, welcomed the group, remarked on the Research Roundtable as an undeniable success story, and quickly launched into the meat of the meeting, asking, “Without scales, how do we tell if [drugs] are doing any good? And how do we know what the outcome measures of scales mean in patients' everyday lives?” Roundtable co-chair Dr Black added, “Measurement is at the core of clinical trials. If you care about clinical trials, you care about scales.”
Lon S. Schneider, MD, professor of psychiatry, neurology, and gerontology at the University of Southern California Keck School of Medicine, provided the keynote address on challenges in assessing cognition, function, and neuropsychiatric status in clinical trials for Alzheimer's. The address highlighted the “fractured” history of Alzheimer clinical trials, whose durations have slowly crept up from 3 months to 6 months to 12 months, and finally to 18 months, with numerous 18-month trials under way. Over and over, the same measures have been used: Staging measures such as the Mini-Mental State Exam (MMSE); cognitive measures such as the ADAS-cog (Alzheimer's Disease Assessment Scale-cognitive subscale); activities of daily living (ADLs) measures such as the DAD (Disability Assessment in Dementia); global measures such as the CDR+sb (Clinical Dementia Rating Scale+Sum of Boxes); and behavioral measures such as the neuropsychiatric inventory (NPI). “What is a meaningful difference on an ADAS-cog, ADL, or NPI?,” he asked. “Why are so many drugs failing [based on data gathered using such scales]?
A useful measure, said Dr Schneider, should be appropriate to the task, valid, efficient, easy to use, sensitive to change in the underlying condition, relatively insensitive to symptom fluctuation, and consistent over time. It must also be reliably reproducible and accurately measure what it purports to measure. Comparing and contrasting common scales used as outcomes measures, he found strengths and weaknesses in each and suggested that what may be needed are better models of illness change; new scales to describe multidimensional stages of illness to clearly assess whether an individual improved; and measures that can easily be interpreted to determine whether a clinically meaningful change has occurred.
To better understand the range of knowledge regarding scales, Roundtable organizers developed specific sections of the meeting to address methodological issues in cognitive scales used in clinical trials; cognitive scales; functional scales; and neuropsychiatric scales and global ratings, followed by sections on cooperative analyses of data and regulatory issues.
Steven H. Ferris, PhD, of New York University School of Medicine provided an overview of cognitive scales and related methodological concerns. While the ADAS-cog is the gold standard cognitive scale, it does have flaws, he said. Cognitive batteries are essential to clinical trials outcomes measures in Alzheimer's, but batteries with improved psychometric properties are needed, as are tools to assess cognition in milder and older unimpaired individuals. Computer-based assessment will likely grow, and home-based assessment may be needed as the field moves to primary prevention, he remarked. Mary Sano, PhD, of the Mount Sinai School of Medicine described her study of cognitive assessment in the home versus through the telephone and in clinical settings, while psychiatrist and epidemiologist Mary Ganguli, MD, MPH, of the University of Pittsburgh discussed cross-cultural challenges to the implementation of cognitive scales, focusing on results of her cross-national dementia epidemiology study comparing populations in the United States with populations in India.
“Sixty percent of those with Alzheimer's live in developing countries, and of them we know much less,” observed Dr Ganguli. Cross-cultural studies are needed, she said, to compare the burden of Alzheimer's as well as risk factors and outcomes of clinical trials and interventions. Assessment tools used in low- and middle income countries must be socially acceptable, cheap, and quick and easy to use. To overcome cultural differences, Dr Ganguli developed a Hindi version of the MMSE that was suitable for illiterate, rural elderly individuals, a Hindi version of the brief cognitive screening test battery created by the Consortium to Establish a Registry for Alzheimer's Disease (CERAD), and a Hindi version of the Geriatric Depression Scale, as well as a culturally adjustable scale for activities of daily living. While it's virtually impossible to eliminate the effects of culture on performance of cognitive tests, with sufficient resources and adequate adaptation, one can come close, she concluded.
Roundtable presenters analyzed cognitive scales from several vantage points. Ron Petersen, MD, PhD, of the Mayo Clinic provided an overview of assessment in cognitive change in mild cognitive impairment (MCI), while John Harrison, PhD, principal scientist at CogState Ltd., described composite measures of cognitive change. “You can't measure cognition as a single entity,” said Dr Harrison. Single composite scores such as ADAS-cog tend to oversimplify, while grouped composites scores may require justification, and noncomposite scores such as individual tests scores can present statistical challenges. Christopher Randolph, PhD, of Loyola University Medical Center, Chicago, took up the topic of RBANS (Repeatable Battery for the Assessment of Neuropsychological Status) in MCI and mild Alzheimer's. RBANS consists of 12 tests in five categories, are easy to administer, and are designed to minimize practice effects but probably are not as sensitive to the full range of disease progression.
Computerized scales offer yet another option, as Jeffrey Kaye, MD, of the Oregon Center for Aging and Technology reported. Computerized testing facilitates data capture, improves standardization, increases reliability, is more precise, and captures data not easily captured by a person, but like the other methods described at the Roundtable, has drawbacks, including barriers to acceptance.
Examining functional scales, Howard Feldman, MD, of the University of British Columbia noted that the diagnostic criteria for Alzheimer's includes “impaired social function,” whereas many of the functional scales available focus on ADLs and other domestic activities. “If we use ADLs as a surrogate for social function, we need to broaden ADLs to include social function,” he said. Sarah Farias, PhD, of the University of California at Davis provided insights into functional measures in MCI and introduced ECog scales, which measure everyday manifestations of cognitive impairment in different domains, such as memory, language, planning, and organization, that may be more sensitive in MCI than other scales.
The University of California at San Diego's Douglas Galasko, MD, provided an in-depth description of the ADCS (Alzheimer's Disease Cooperative Study)-ADL scale, which covers 23 functional activities and which he believes tracks functional change adequately for clinical trials involving individuals with mild to moderate Alzheimer's. Deborah Cahn-Weiner, PhD, of the University of California–San Francisco shared her research showing that level of executive function more accurately predicts instrumental activities of daily living (IADLs) than other cognitive measures and than demographic variables such as age, education, and health status.
During the opening presentation at day two of the Roundtable, Pierre Tariot, MD, of the Banner Alzheimer's Institute, remarked, “We talk about behavior as if we know what it means, but behavior is all that we do
…[and it is] a complex phenomenon.” That complexity was mirrored in his overview of scales for neuropsychiatric assessment in Alzheimer's, including focused and general scales and global ratings. While the NPI is still the gold standard, the addition of a global rating is “extremely important,” he stated. Colleague Constantine Lyketsos, MD, MHS, of Johns Hopkins Bayview Medical Center provided an examination of neuropsychiatric scales in community settings by focusing on the population of Cache County, Utah. Dr Lyketsos described the challenges of identifying neuropsychiatric symptoms in the early stages of dementia and a new project to add a clinical component to the NPI, creating the NPI-C.
Complexity also applies to the numerous global scales available, as John C. Morris, MD, of Washington University School of Medicine in St. Louis made clear in his presentation of their advantages and disadvantages. Following on his heels, Kenneth Rockwood, MD, of Dalhousie University narrowed the focus to clinical meaningfulness in global ratings. After touching upon standardized clinical global measures, Dr Rockwood turned his attention to individualized outcome measures (for example, the Clinician's Interview-Based Impression of Change + Caregiver Input, CIBIC+), which have the advantages of being clinical meaningful, being generalizable, and providing data that has face value. Individualized outcomes measures can complement standard measures, Dr Rockwood notes, to allow understanding of patient and caregiver preferences and clinically recognizable treatment effects.
Among other topics, Laurie B. Burke, RPh, MPH, director of study endpoints and labeling for the FDA's Office of New Drugs, discussed patient reported outcomes (PROs) and treatment benefits. Of special importance to clinical investigators is the FDA's 2006 document, “Guidance for Industry: Patient-Reported Outcomes Measures: Use in Medical Product Development to Support Labeling Claims,” and a companion document on target product profiles that Burke drew upon in explaining the FDA's stances in various aspects of labeling.
The Roundtable addressed cooperative analyses of data by opening the floor to Laurel Beckett, PhD, of the University of California at Davis to discuss the challenges of analyzing data from the Alzheimer's Disease Neuroimaging Initiative. Marilyn Albert, PhD, shared her own challenges in gathering and analyzing placebo data from Alzheimer's and MCI clinical trials and welcomed numerous suggestions from attendees to more narrowly define the type and format of placebo data she required so attendees and their companies could more easily provide the data to her.
During a general business meeting preceding the Roundtable, members heard an update of plans for the fall Roundtable meeting on optimal trial design and selected topics for the two subsequent meetings: diagnostic classification of Alzheimer's (spring 2009) and development of global clinical trials (fall 2009). Maria Carrillo, PhD, director in the Alzheimer's Association's medical and scientific relations division, provided an update on a grant awarded by Roundtable members to Greg A. Sachs, MD, to explore Alzheimer's disease and access to palliative care. Paula Moore, director of the Association's Clinical Studies Initiative, shared results of the initiative to date and information on a potential expansion of the initiative, while Stephen McConnell, the Association's senior vice president for advocacy and public policy, described the mission and progress of the Alzheimer's Study Group and Mega Community for Alzheimer's, initiatives to align Alzheimer's efforts nationwide.
The next Research Roundtable will take place October 20–21, 2008, in Washington. To join the Research Roundtable, contact Jay Thompson, associate director, corporate initiatives, Alzheimer's Association, at 312-335-5192 or jay.thompson@alz.org.
Pilot Program Raises Awareness of Enrollment in Clinical Studies
When the Pilot Program of the Alzheimer's Association Clinical Studies Initiative was launched in April 2007, it faced daunting statistics. Of the 9,380 interventional studies under way in the United States, 80 percent are delayed because of enrollment shortfalls. The challenge of recruiting and retaining study participants has become a significant impediment to developing next-generation drugs.
The Clinical Studies Initiative was formed to find effective ways to mobilize and motivate study participants in an effort to accelerate clinical research. Recruitment strategies were tested with the assistance of Association chapters headquartered in five pilot cities: Atlanta, GA; Indianapolis, IN; Providence, RI; San Francisco, CA; and Tulsa, OK.
The pilot program consisted of several components: market research; creation of a Clinical Studies Initiative Advisory Council; physician education; outreach to patients, caregivers, and the public; use of a healthcare marketing agency specializing in patient recruitment; and evaluation of metrics and feedback.
As physicians' attitudes strongly influence patients' decisions regarding study enrollment, market research focused on physicians' reluctance to encourage their patients to participate in clinical research and examined the reasons for their resistance. Results of interviews with 100 primary care physicians in each market showed that lack of awareness and information was the largest deterrent to referrals to clinical studies. Most interviewees did not know how to find out about clinical studies under way near their practices; 79 percent could not name a single clinical studies database. Physicians strongly preferred learning about clinical studies from other physicians, either one-to-one or through a medical conference. While nearly 75 percent had referred patients to clinical studies, only 25 percent had referred patients to Alzheimer studies. Almost 30 percent professed fear of the side effects and risks of clinical studies and of the responsibility inherent in referring patients to them.
These findings shaped physician outreach efforts to foster positive attitudes toward and thereby increase referrals to clinical studies. More than 150 primary care physicians attended continuing medical education (CME) presentations on recent developments in Alzheimer's disease research, receiving 1.5 CME units. Physicians also received toolkits containing clinical studies–related brochures, flyers, and posters and a USB flash drive with the CME presentation and information on how to find local clinical studies. A survey distributed after the pilot program found that referrals had increased 23 percent at pilot sites.
A Center Watch poll of the general public showed that 75 percent did not know how clinical studies worked, and 98 percent had no idea how to identify open research studies. To educate patients, caregivers, and the public, prominent local researchers were identified in each pilot city. Researchers were featured on brochures, served as spokespeople with the media, and attended CME dinners to establish relationships with referring physicians. Efforts to raise awareness included a focus on minority groups and a three-month grassroots outreach campaign. The grassroots campaign led to the distribution of 40,000 pieces of collateral material, including lists of clinical studies under way; attendance at more than 90 community events including health fairs and events geared toward senior citizens; and the establishment of partnership agreements with more than 60 local organizations and businesses to foster community awareness in a variety of ways.
Communication efforts also included the addition of a user-friendly version of clinicaltrials.gov to the Alzheimer's Association Web site, Internet advertising at sites such as WebMD and caregiver.com, and the implementation of a centralized toll-free Clinical Studies Hotline to provide information on clinical studies. The hotline fielded 510 inquiries over five months, testing the effectiveness of the pilot program's print, radio, and television advertising campaigns.
After completion of the pilot program, nineteen percent of survey respondents at pilot sites said enrolling participants into Alzheimer research studies was “not at all difficult,” while respondents at non-pilot sites said it was “very difficult” (50 percent) or “somewhat difficult” (50 percent). Asked how enrollment compared with the same time last year, 15 percent of respondents at pilot sites said it was less difficult, while respondents at non-pilot sites said it was about the same (57 percent) or more difficult (43 percent). After the pilot program, 38 percent of respondents at pilot sites reported that calls about participation in Alzheimer clinical studies had increased, while respondents at non-pilot sites said it remained the same (71 percent) or decreased (21 percent). Last, but not least, 54 percent of respondents at pilot sites said the number of individuals screened for Alzheimer clinical studies had increased, while 57 percent of respondents at non-pilot sites said it had decreased.
The Alzheimer's Association plans to expand the Clinical Studies Initiative to include additional chapters later this year.
FULL TEXT