Adjunct Meeting to the National Neurotrauma Society Meeting, Las Vegas
Sponsored by the European Commission and NIH/NINDS
June 16, 2010 1:00 – 5:00
The European Commission and NIH jointly sponsored a meeting to explore opportunities and challenges for promoting effective research on traumatic brain injury (TBI). TBI is a major global health problem, with no proven therapies. Therapy development for TBI is challenging because of the patient heterogeneity, as well as the large number of diagnostic and treatment options, which are in highly variable use across centers and geographic regions. This variability makes it very challenging for clinical trials to detect a treatment effect. Speakers were invited to discuss their experiences and/or vision for using comparative effectiveness research to determine which TBI treatments work best for which patients in real world settings. Major points are embedded in the agenda below. A manuscript summarizing the speakers’ presentations is in preparation, with plans to submit it to the Journal of Neurotrauma in Summer 2010.
Meeting Co-Chairs : Patrizia Tosetti, PhD, European Commission and Ramona Hicks, PhD, NIH/NINDS
Presentations (20 min + 5 Q&A)
TBI Research: State of Play and Unmet Needs
David Menon, MD, University of Cambridge, United Kingdom
- Aggressive critical care improves outcomes, but the “guidelines” lack hard data
- Preclinical research has generally failed to translate into significant clinical data
- Patient heterogeneity requires stratification based on dominant pathophysiology, which will require sequential neuroimaging
TBI: Phenotypes, Treatments and Outcomes
Geoffrey Manley, MD, PhD, University of California, San Francisco
- TBI research is challenging because there are so many different patient-specific factors and treatment-specific factors
- A systems biology approach makes sense, but we are not applying this approach for TBI clinical trials
- Pathoanatomical classification will require information science, standardization, infrastructure, and data
- Observational clinical data followed by high powered data analysis is needed to advance the field
Studying TBI Using a Natural Experiment Approach
M. Elizabeth Sandel, MD, Kaiser Foundation Rehabilitation Center and Hospital, Vallejo
- Kaiser is currently studying a cohort of 60,000 stroke patients
- Important to consider which items to collect in the data set
- Developed an efficient outcome measurement for stroke research
- Restricted cohorts used for further analysis
Comparative Effectiveness Research: What Can it Bring to the Field of TBI?
Hester Lingsma, MSc, Erasmus MC, Rotterdam, the Netherlands
- Retrospective analysis of clinical trials for TBI demonstrated 3.3 fold difference between centers on outcomes (adjusted for co-variables such as patient severity, age, etc.)
- Reasons for the center differences remain unanswered because of a lack of data, but may be caused by differences in ICP management, type and timing of surgery, trauma organization/workflow, post-acute care, etc.
- A prospective observational study is needed, but how do we prioritize the questions?
Pediatric Considerations for TBI Research
Jose Pineda, MD, MSc, Washington University, St. Louis
- Center variability is even greater for pediatric TBI because some cases go to adult trauma centers
- ICP monitoring ranges from 19 – 90% of children with severe TBI
- RCTs could be improved in terms of efficiency and relevance to a broader spectrum of patients, see “Rethinking randomized clinical trials for comparative effectiveness research: the need for transformational change”, Ann Intern Med. 2009, 151:206-209.
- There are several pediatric research networks available for collecting data
A Vision for the Future: Benefits of International Collaboration in TBI
Andrew Maas, MD, PhD, Univ. Hospital Antwerp, Belgium
- 33 phase 3 RCTs in TBI have had a low impact on clinical guidelines
- Only 3% of the recommendations in the TBI guidelines are based on Class I evidence
- Take advantage of the existing treatment heterogeneity to determine which treatments are best and for which patients
- Additional data, such as sequential neuroimaging may be required – it is unlikely data routinely being collected for TBI patients will be sufficient
- Need to identify and prioritize research needs to ensure sufficient data is collected to meet these priorities
- How do we get there? Opening question for panel discussion.
Strategies for an International Comparative Effectiveness Study for TBI
Moderator: Andrew Maas
Panel: Kenneth Curley, MD, Army Medical Research and Materiel Command, Alan Faden, MD, U. Maryland, Nicole von Steinbeuchel, PhD, University Medical Center Gottingen and above speakers: Hester Lingsma, Andrew Maas, Geoffrey Manley, David Menon, Jose Pineda, M. Elizabeth Sandel)
How do we get there?
- Develop better quality of life outcome measurements
- Translational research needs to be bi-directional
- Bring in health economists
- Use open source materials and data sharing – models are in place
- Globalize the common data element effort for TBI
- Need to ensure that we collect high quality data (avoid the garbage in –garbage out problem)
- Need to prioritize research questions
- Need biomarkers in addition to longer term clinical outcomes and quality of life measures
- Need to compare different systems of care, too, not just specific treatments, e.g. the continuum of care which is typically very fragmented
- NIDRR TBI model systems approach may be relevant