Meeting on Skeletal Muscle Magnetic Resonance Outcome Measures for Muscular Dystrophy Clinical Trials
NIAMS, NIH, Bethesda , Maryland .
This meeting provided an opportunity for investigators using magnetic resonance (MR) modalities in studies of muscular dystrophy (MD) patients to share data and to discuss strategies for coordinating efforts to further develop and validate MR outcome measures for use in clinical trials. Researchers from the United States and the United Kingdom studying boys with Duchenne muscular dystrophy (DMD) presented data collected through observational studies utilizing a wide range of MR modalities including T1 (T1w) and T2 (T2w) weighted imaging (MRI), diffusion tensor (DTI) and three-point Dixon (3PD) imaging, and proton ( 1 H) and phosphorous ( 31 P) MR spectroscopy (MRS). Participants concluded that MR approaches have the potential to noninvasively assess skeletal muscle volume, lipid and water composition, and sarcolemma integrity and may yield early surrogate measures predictive of subsequent changes in muscle function. An early surrogate of important changes in muscle could be very valuable in the assessment of interventional drugs. Participants were also enthusiastic about the expanding potential of MR approaches in advancing an understanding of the pathogenesis of muscle diseases.
Several meetings in recent years have addressed outcome measures and the design of MD clinical trials. The Muscular Dystrophy Association sponsored a meeting in 2005 titled "Challenges in Drug Development for Muscle Diseases: A Stakeholders' Meeting." At that meeting, academic, private sector, and government investigators and administrators discussed outcome measures-including imaging approaches, muscle function assessments, and quality-of-life measures-for muscle disease clinical trials (see Mendell et al., Muscle Nerve 35:8-16, 2007). Also in 2005, a working group of scientific experts recommended the high-priority research objectives in the Muscular Dystrophy Coordinating Committee's Action Plan. One recommendation articulated the need for more advanced imaging approaches for diagnosis and for use in clinical trials.
Relatively few clinical trials are underway for MD. Most are early-stage studies, testing for safety and using gene expression in muscle biopsies as an outcome measure. However, as discussed at the 2007 NIH Workshop on Translational Research in Muscular Dystrophy (http://www.ninds.nih.gov/news_and_events/proceedings/Translational_Research_in_Muscular_Dystrophy.htm), dozens of potential therapeutic strategies have shown promise in animal studies and may mature to clinical trials. The number of experimental therapeutics potentially ready for clinical trials in MD has increased dramatically in recent years and is likely to continue to grow.
Strategies aimed at downstream targets in the pathogenic pathways, not at the root cause of MD, may yield treatments for only some of the symptoms. Some therapeutics may have positive synergistic effects. Therefore, clinical trials testing combinations of potential therapeutics may be necessary to improve overall efficacy. The current directions of MD research point at the need for clinical testing of a large number of potential therapeutics individually and in combination. The development and validation of more efficient outcome measures for use in these trials can accelerate progress toward effective treatments.
Current State of the Science
Outcome measures currently used in trials in MD or other muscle diseases include serum biomarkers of muscle breakdown (e.g., serum creatine kinase), expression and localization of specific proteins analyzed in muscle biopsies, measures of muscle function (strength or resistance to fatigue), and measures of quality of life and psychosocial health. While each has merits and may be most appropriate for specific trials, additional outcome measures that are noninvasive and sensitive to early changes in muscle structure, composition, or physiology can accelerate clinical studies for a wide range of muscular dystrophies and other muscle diseases.
Participants at this meeting presented data from pilot and ongoing observational studies of DMD subjects. Imaging and analysis have been performed successfully on boys as young as five years. It was stressed that the quality of imaging data acquired depends heavily on subject compliance. Attention to the comfort of the patients is an important step in assuring their compliance. Training the children on a model of the MR instrument, having them watch entertaining videos during the scan, and having a parent close by can help keep them calm and still to improve the quality of data. Most should be able to spend 30 to 60 minutes in the magnet, allowing for data collection from several MR modalities. It was recognized that great advances have been made over the last few years with fast imaging protocols, which can reduce scan time. However, care must be taken not to sacrifice the quantitative nature of the scan in favor of short scan times. Meeting participants indicated that fibrosis and contractures, often present in patients older than 14 years of age, increase the discomfort associated with MR scans and decrease the quality of data that can be achieved through MR analysis.
Over the course of this meeting, participants discussed the strengths and weaknesses of various MRI and MRS modalities as they relate to skeletal muscle and detection of the fat, water, fibrous tissue, and inflammation within the muscle. Each brings a tradeoff between coverage and detail, as well as resolution or sensitivity and the time to acquire the data.
- T1w images were considered by participants to be most appropriate for anatomical mapping and for determining muscle cross sectional area or volume.
- Gadolidium contrast can provide additional information on tissue permeability, but its intravenous administration increases the complexity of the protocol and adds a risk of adverse reactions. Some participants considered the use of contrast agents in their dystrophy patients not to be advisable.
- T2w images can quantify fat within specific muscles or muscle groups.
- 3PD MRI was recommended as an approach that provides both anatomical detail and quantitation of lipid content. 3PD is a relatively fast technique, and it could be used to monitor changes in muscle fat over weeks or months, which may be a good predictor of disease progression. 3PD MRI was considered preferable to T1w imaging for lipid quantification.
- Quantification of lipid, creatine, and water concentrations within smaller, selected regions of interest of muscle can be done with 1 H MRS. Although this method is more time consuming than 3PD, it has the advantages of greater sensitivity and specificity.
- 31 P MRS was discussed as an effective approach for measuring energetic metabolites such as ATP, phosphocreatine, inorganic phosphate, and metabolites of cell membrane synthesis and breakdown. Therefore, it can indicate changes in muscle physiology that may be associated with early events in disease progression. The limitations of 31 P MRS are low signal-to-noise ratio, low spatial resolution even with long scan times, and the need for exercise before or during data acquisition to follow energetics.
- DTI is being explored as a method for quantifying muscle fiber size, orientation, and integrity. Although DTI has been used in only a few observational studies in skeletal muscle, it has been used extensively in brain imaging, and it will likely be a sensitive measure of muscle degeneration due to the highly ordered structure of this tissue.
- Short Tau Inversion Recovery MRI is being explored as a method for distinguishing inflammation/edema from fat infiltration, potentially an early marker of disease progression.
Meeting participants recommended T1w, T2w, 3PD, and 1 H MRS as preferred modalities for acquiring a standardized minimal data set informative of muscle disease status in young patients capable of spending a limited amount of time in the magnet. These measures could be conducted with the instruments available at most clinical research centers.
Imaging data presented at this meeting demonstrated significant variability among DMD patients at the same age and among muscles in the same patient. In the legs, the gastrocnemius and soleus muscles appear more affected than the tibialis muscles. Recent physical activity or stretching may also introduce variability in the sensitive MR measures. Additional studies will be necessary to assess variability of MR measurements along the length of individual muscle, which could reinforce the potential benefits of MR over muscle biopsies with regard to sampling error.
With technology that is currently available in most clinical research institutions, MR approaches can quantitate skeletal muscle cross-sectional area, volume, and lipid and water content. Additional, more specialized approaches can measure muscle fiber length and orientation, injury, sarcolemma integrity, metabolism, and energetics. Preliminary studies in MD patients suggest that MR measures of muscle structure and composition may be more sensitive than functional outcome measures in assessing disease-related changes over short periods of time.
MR modalities offer great potential to accelerate studies of muscle diseases and disorders, both as outcome measures for clinical studies and as powerful tools to explore pathophysiology and mechanisms of disease. The trajectory of clinical trials for the muscular dystrophies draws attention to the need for more efficient and sensitive outcome measures. The wide range of muscle parameters that can be directly measured non-invasively by MR increases the likelihood that early surrogate markers of disease progression can be identified. For example, changes in intramuscular lipid composition may be a predictor of subsequent changes in muscle function in DMD.
Additional studies are needed to determine the basis of variability of these measures and to establish standardized approaches for minimizing variability. Studies to compare MR data with histology/histopathology in biopsies from animal models or patients would provide important validation of the MR approach. If well-designed and carefully controlled studies in one form of MD demonstrate that MR outcome measures are sensitive, accurate, and valid measures of disease progression, the techniques may quickly find applications in studies of other dystrophies and other, more common muscle diseases and disorders.
Meeting participants discussed a wide range of factors that should be considered in designing studies aimed at further developing MR outcome measures for use in MD clinical trials.
- To conduct studies with the highest power and the smallest number of subjects, variability in outcome measures must be minimized.
- Enrolling patients in early or middle stages of disease progression and working closely with the patients and parents can increase data quality.
- Study design would likely involve normal controls, and including disease controls may help to assess specificity.
- To minimize risk to the subjects and simplify institutional review board approval, sedation, repeated biopsies, and intravenous administration of contrast agents should be avoided.
- MR outcomes should be compared in the same studies to functional outcome measures such as timed walking tests, composite measures of physical function, and quantitative force assessments. Quality-of-life measures should also be included in these studies.
- Instrument capability and compatibility should be considered.
- Since the object is to develop MR outcome measures that can be used effectively in multisite clinical trials, studies aimed at developing and validating the measures should, if possible, utilize the same types of instruments and expertise that would be available at trial sites.
- Standards for calibration phantoms should be established and implemented.
- All of the investigators in a study involved in collecting and handling the data at multiple sites should receive standardized training to assure consistent, high quality data.
- Multisite studies should include a central site for data collection, quality monitoring, and consistent data processing.
- International coordination, communication, and sharing of data among investigators are necessary for the development of skeletal muscle MR outcome measures for muscular dystrophy clinical trials. The use of these measures in clinical trials will help accelerate progress toward effective treatments for these and other important muscle diseases and disorders.
For more information about this meeting or to discuss NIH support for research on this topic, contact Dr. Glen Nuckolls, Program Director for Muscle Disorders and Therapies, NIAMS (firstname.lastname@example.org, 301-594-4974).
Stefan Bluml, Ph.D., University of Southern California
Jane Zheng Fan, M.D., University of North Carolina Chapel Hill
Richard S. Finkel, M.D., The Children's Hospital of Philadelphia
Vicente Gilsanz , MD. , Ph.D., University of Southern California
Penny Garrood, M.D., University of Newcastle upon Tyne
Martin J. Kushmerick M.D., Ph.D., University of Washington
Steven A. Moore, M.D., Ph.D., University of Iowa
Shantanu Sinha, Ph.D., University of California , San Diego , School of Medicine
Lee Sweeney, Ph.D., University of Pennsylvania Health Systems
Krista Vandenborne, Ph.D., P.T., University of Florida
Glenn A. Walter, Ph.D., University of Florida
Amanda Boyce, Ph.D., NIAMS
Jonelle K. Drugan, Ph.D., M.P.H., NIAMS
Laurie Gutmann, M.D., NINDS
Shahnaz Khan, M.P.H., NIAMS
Gayle Lester, Ph.D., NIAMS
Joan McGowan, Ph.D., NIAMS
Glen Nuckolls, Ph.D., NIAMS
Giovanna M. Spinella, M.D., NIH Office of Rare Diseases
Madeline Turkeltaub, R.N., Ph.D., C.R.N.P., NIAMS