Abstract
Background:
A Congenital Muscular Dystrophy (CMD) Working Group (WG) consisting of international experts reviewed common data elements (CDEs) previously developed for other neuromuscular diseases (NMDs) and made recommendations for all types of studies on CMD.
Objectives:
To develop a comprehensive set of CDEs, data definitions, case report forms and guidelines for use in CMD clinical research to facilitate interoperability of data collection, as part of the CDE project at the National Institute of Neurological Disorders and Stroke (NINDS).
Methods:
One working group composed of ten experts reviewed existing NINDS CDEs and outcome measures, evaluated the need for new elements, and provided recommendations for CMD clinical research. The recommendations were compiled, internally reviewed by the CMD working group, and posted online for external public comment. The CMD working group and the NIH CDE team reviewed the final version before release.
Results:
The NINDS CMD CDEs and supporting documents are publicly available on the NINDS CDE website (https://www.commondataelements.ninds.nih.gov/CMD.aspx#tab=Data_Standards). Content areas include demographics, social status, health history, physical examination, diagnostic tests, and guidelines for a variety of specific outcomes and endpoints. The CMD CDE WG selected these documents from existing versions that were generated by other disease area working groups. Some documents were tailored to maximize their suitability for the CMD field.
Conclusions:
Widespread use of CDEs can facilitate CMD clinical research and trial design, data sharing and retrospective analyses. The CDEs that are most relevant to CMD research are like those generated for other NMDs, and CDE documents tailored to CMD are now available to the public. The existence of a single source for these documents facilitates their use in research studies and offers a clear mechanism for the discussion and update of the information as knowledge is gained.
INTRODUCTION
The congenital muscular dystrophies (CMDs) are a group of rare genetic muscle diseases that present at birth or during infancy with hypotonia and weakness [1–3]. Batten [4, 5] first described the clinical and pathological features of CMD in 1903 and 1904 and Howard [6] proposed the term Dystrophia Muscularis Congenita in 1908. As a group, CMDs are genetically determined [1, 2] and were classified based on clinical or ethnic characteristics. The original classical descriptions include congenital atonic-sclerotic muscular dystrophy (Ullrich CMD) [7], Fukuyama CMD (FCMD) [8], rigid spine syndrome [9], muscle-eye-brain (MEB disease) [10], and Walker-Warburg syndrome (WWS) [11, 12]. With the identification of specific disease causing mutations over the past two decades, further variants have been recognized and there has been clarification of genotype/phenotype correlations [1, 2]. CMDs overlap and are allelic with some limb girdle muscular dystrophy phenotypes wherein symptoms begin in late childhood or even adulthood. Diagnosis depends on muscle biopsy and DNA analysis. In the last decade, there have been several peer reviewed publications addressing the cross sectional natural history of several CMD variants. There is also increasing interest in translational studies of CMD, and progress in these areas may be facilitated by efforts to standardize data collection and reporting.
The effort to standardize data collection by NINDS began in 2005 with the development of CDEs to assist NINDS-funded investigators in collecting neuroscientific clinical research studies data. The CDEs are a guide for research data to be collected in a standard and consistent fashion [13]. The primary goals of this NINDS CDE Project are to: 1) Disseminate standards for the collection of data from participants enrolled in studies of neurological diseases; 2) Create easily accessible and affordable tools for investigators to collect study data; 3) Encourage focused and simplified data collection; and 4) Improve data quality by providing uniform data descriptions and tools across NINDS-funded clinical studies [13]. The CDEs are content standards that can be applied to various data collection models and are intended to be dynamic and evolve over time. It is important to note that the CDE project is not a database – rather it is a collection of metadata and data standards which consist of identified common definitions and standardized case report forms (CRFs) and outcome measures. Data collection using CDEs allows more straightforward comparisons between different studies and groups and facilitates useful meta-analyses increasing the ability to conduct translational studies in CMD.
To date, the NINDS CDE website contains metadata with data standards that identify common definitions and standardized CRFs and outcome measures for 24 neurological diseases, including CMD. This paper reviews the process by which the CMD Working Group (WG) adapted the CDE documents that were available for other disease states for use in the CMD field. The usefulness of each of these documents was discussed by an international panel of experts in areas such as neurology, psychology, rehabilitation, neuropathology, genetics, and biochemistry. The subset of CDE documents pertaining to CMD were then reviewed and tailored to be suitable for data collection in the CMD patient population.
METHODS
Data element development
NINDS CDEs are based on the International Organization for Standardization and by the International Electrotechnical Commission (ISO/IEC) 11179 metadata registry standard. The naming convention is also based on international standards, ISO/IEC 11179 and provides the needed foundation for interoperability, in other words, data sharing so that information from different sources may be able to be pooled or compared. The resulting meaningful metadata assists in the process to better store, organize, analyze, search, share and publish data.
Development of CMD-specific CDEs
The first set of CMD CDEs was developed at the request of researchers in the CMD clinical research community who were interested in using them for a planned natural history study and clinical trial [14]. Between 2008 and 2010, a series of meetings (Table 1) were held that established the need for CDEs in the field of CMD. This initial set of recommended CDEs, developed in collaboration with Cure CMD, were only for CMD natural history studies and were not merged or added to the overall Neuromuscular Disease (NMD) recommendations. These initial meetings formed a basis for CDE generation, review, and administrative support through the collaboration between CMD specialists (called the “CMD CDE Working Group”) and the NIH CDE team [14–17].
Meeting history in the development of current NINDS CMD CDEs
The NIH CDE team released the Neuromuscular Disease (NMD) NINDS CDEs in 2012. In 2014, a new CMD CDE Working Group was developed and members provided expert commentary on the NMD CDEs relevant to CMD. From May to November 2014, the CMD CDE Working Group (See Table 2) identified common data elements/sets and supporting documentation such as data dictionaries, form templates and manuals of procedures tailored for CMD research to assist investigators who are conducting CMD studies. The CMD CDE Working Group met with the NIH CDE team monthly to move this effort forward, with the purpose of developing CDE documents relevant to the CMD patient population.
NINDS congenital muscular dystrophy common data element working group (2014 to present)
NINDS: National Institute of Neurological Disorders and Stroke; NIH: National Institutes of Health. 1Also participated in November 2008, March 2009, March 2010, June 2010, September 2010, and October 2010 CMD Meetings listed in Table 1. 2Also participated in March 2009, June 2010, and September 2010 CMD Meetings listed in Table 1. 3Also participated in November 2008 and September 2010 CMD Meetings listed in Table 1. 4Also participated in November 2008 and March 2010 CMD Meetings listed in Table 1. 5Also participated in March 2010, September 2010, and October 2010 CMD Meetings listed in Table 1. 6Also participated in the November 2008 CMD Meeting listed in Table 1.
CMD CDE working group
Members of this CMD CDE Working Group were selected based on their prior work in the CMD field and/or based on their involvement in NINDS CDE efforts related to other neuromuscular disorders. Once all areas of specialization were adequately represented, each CMD CDE Working Group member was assigned a subset of CDE documents for review and comment. They reviewed all the existing neuromuscular diseases, Duchenne/Becker muscular dystrophy (DMD/BMD), myasthenia gravis (MG), spinal muscular atrophy (SMA), amyotrophic lateral sclerosis (ALS), and Friedreich ataxia (FA) CDE recommendations. Elements of the review included 1) whether the CDE document or measurement was truly suitable for the CMD population, 2) whether the individual data elements in each document were suitable for data collection in the CMD population, and 3) the importance of each individual data element (Table 3). These determinations were then discussed and debated via teleconference with the entire CMD Working Group and edited into a final format based on the consensus of the group.
Characteristics used to make CDE outcome measure recommendations:
Data element classification
With respect to determining the importance of individual data elements, the NINDS CDE effort has generated a grading scale to label elements as “Core,” “Supplemental – Highly Recommended,” applicable to any study and might generally be required information. Such elements are very rare, given elements correspond to tests that are highly useful within the CMD field and are highly recommended if those sorts of data are being tracked, but they are not considered “Core” to any study in the CMD field. This classification is highly useful when determining the most suitable test or endpoint for collecting a certain type of data from CMD patients. The “Supplemental” group of elements can be very useful in CMD clinical research studies, but the importance of these elements depends on their relevance to the study’s design (i.e., clinical trial, cohort study, etc.). “Exploratory” elements are those that may be useful in CMD research, but have not been sufficiently studied and validated.
After a final internal review by the Working Group, the CMD CDEs were posted for public review from November 2014 through January 2015. Comments and suggestions from the public review were considered by the Working Group, and revisions were made. CMD CDE Version 1.0 was released on the NINDS CDE website in March 2015.
RESULTS
From an initial pool of 213 outcome measures that had been reviewed in other Neuromuscular diseases such as MG, SMA, and DMD/BMD, the CMD CDE Working Group narrowed the list to approximately 75 outcome measures. Of these the WG requested additional information on 21 outcome measures.
Early meetings of the CMD CDE WG focused on the identification of relevant documents, with considerable discussion on whether specific assays or endpoints were useful in the CMD population. A list of documents specific to CMD are found in Table 4 and outcome measures that were classified as Supplemental – Highly Recommended are found in Table 5. For a complete listing of the CMD CDE recommendations, please see the NINDS CDE website (https://www.commondataelements.ninds.nih.gov/#page=Default). CDE documents that were excluded from the CMD CDE list (e.g., Swallowing-Slurp Test, French Motor Function Measure (MFM) Scale) included tools that were not used in the CMD field. These typically were either unsuitable for use in the age range of CMD patients or were associated with phenotypes not seen in CMD.
CMD-specific case report forms (CRFs), measures, and scales
Supplemental– highly recommended outcome measures for congenital muscular dystrophy
*Special circumstances apply.
Once the key documents were identified, individual data elements were assessed for their suitability in CMD. In some cases, this resulted in the addition of data elements beyond what was originally included in the CDE document. Examples of such additions and changes include: Intake Medical History: Data elements included in this form collect CMD-specific data points in the body system categories. For example, specific data elements were expanded upon for Ophthalmologic (i.e., strabismus, retinal detachment); Pulmonary (i.e., BiPAP use, difficulty breathing during meals); and Musculoskeletal (i.e., number of broken bones, joint dislocation) body system categories. Interval Medical History: The data elements in this form are collected to assess changes in disease progression or health status from study visit to study visit. Like the intake medical history form, CMD-specific data points are assessed in the body system categories. Prenatal and Perinatal History: Given that CMD often affects patients in infancy, Working Group members suggested the inclusion of additional data fields to provide a greater investigation of maternal and prenatal history. Elements for “Respiratory status at birth” and “Muscle tone at birth” were added to this form. Muscle Biopsies and Autopsy Tissue: Additional data elements were added to this form, based on feedback from numerous diagnosticians while publishing a paper related to the use of a CDE-based muscle biopsy reporting form [18]. There were no elements that were added specifically for the CMD population. Peripheral Nerve Biopsies: A more complete clinical history section (like the section included on the “Muscle Biopsy and Autopsy Tissue” document) was added.
Additional discussions focused on the classification of individual data elements as Core, Supplemental – Highly Recommended, Supplemental, or Exploratory. Once again, changes were reflective of whether a given element was relevant to the CMD population and the likelihood that a test or element would be used in a high-quality CMD clinical research study. Overall, there were remarkably few elements defined as Core elements, partially because the clinical course of CMD is extensively heterogeneous and thus there are few elements that would be essential and applicable to all CMD patients (Table 6). In general, the list of Core elements includes identifying or demographic data, as such information is required in medical records to allow appropriate billing and quality control practices. Examples of topics discussed include: Surgical History: The “Total number of surgeries in lifetime” field was considered a Core element, with all other elements on this form (relating to specifics of these surgeries) designated as Supplemental. Prior and Current Medications: The question “Did the participant/subject take any medications (Insert number of days here) days before or during the study” was designated as a Core element to studies where medication use was/is tracked. Biopsy/Autopsy-Related Documents: Given the limited clinical information that is often available on diagnostic biopsies, the only elements here that could truly be considered “Core” are required elements that would be needed for patient and specimen tracking (such as patient age and gender). All other elements were considered Supplemental or Supplemental – Highly Recommended for studies including biopsy or autopsy tissue. Echocardiogram: All CDEs in these documents were classified as Exploratory. CMD core common data elements (CDE)
DISCUSSION
The NINDS CDE project has focused on the identification of critical data elements in clinical research related to neurological disease and the construction of documents for the standard collection of these data elements. General (nonspecific) CDEs have now been generated through this effort, as well as CDEs focused on ALS, cerebral palsy, Chiari I Malformation, epilepsy, FA, headache, Huntington disease, mitochondrial disease, multiple sclerosis, neuromuscular diseases (including CMD, DMD/BMD, facioscapulohumeral muscular dystrophy, MG, myotonic dystrophy, SMA), Parkinson disease, spinal cord injury, and traumatic brain injury [13, 19–34]. Developing an expertise in the documentation of CMD and other disease specific CDEs goes beyond just cataloging CDEs. The NINDS CDE project is a resource for helping researchers/clinicians to obtain current versions of outcome measures and identify how to utilize in their research or clinical settings. Furthermore, the existence of a central reference site for CDE acquisition and discussion will facilitate the development of “living documents” that evolve as our state of knowledge in this area progresses. The CDE philosophy of data collection is increasingly prevalent in clinical research as this program has progressed, and public access to CDE documents has streamlined the efforts of new investigators when designing research studies in these fields.
This report describes a collaborative international effort to identify critical aspects of data collection in the CMD field. The process of identifying CDEs suitable for CMD clinical research was dramatically streamlined by the abundance of already-designed CDE documents created by groups focused on other neurological disease states. While there was an initial expectation that new CDE documents would be required, review of these documents identified that only minor changes were necessary in some forms to make them suitable for CMD studies. This speaks to the value of the CDE project as a centralized effort where duplicative effort was avoided due to the CDE development by other neurological disease study groups. As CDE collection efforts progress, the establishment of data collection standards should become less cumbersome as other standards continue to be designed and shared.
There is currently little published data on the use of the posted CDE standards in active research, but we have anecdotal experience in the use of the “muscle biopsy” CDE forms and elements in teaching and research settings. While there is little evidence that experienced muscle pathologists have switched to the use of the CDE checklist format for the reporting of clinical muscle biopsies, we have received feedback that the list of findings on the form is very useful in resident and fellow education. Additionally, several gene therapy clinical trials have incorporated the use of a modified CDE checklist in their muscle biopsy analysis plans, as this format allows a complete and unbiased set of potential findings while also improving the interoperability of data collection by biopsy review committees. The checklist format also allows the additional advantage of well-defined potential entries, which facilitates the development of specific muscle biopsy evaluation modules within clinical trial data entry software. As the trials implementing these checklists have just recently begun, it remains to be seen whether this approach is useful or if it will be judged to be unnecessarily restrictive.
The NINDS CMD CDE project was comprised of a WG of researchers from the CMD community. The WG reviewed a few of the CDEs related to patient history and recommended the addition of minor components. The WG suggested the use of the CDE lists that were compiled for muscle biopsies, nerve biopsies, and intraepidermal nerve fiber densities. Due to there being no standard template (Note: there are indeed standards for pathologists’ reports via CLIA certification, but no agreed upon template for muscle/nerve biopsies) for pathological reporting of muscle or nerve biopsies, those forms were generated through a process of personal experiences and suggestions from the neuromuscular pathology community.
The process of compiling CDE’s was useful in the identification of gaps of knowledge in the CMD field, and hopefully the establishment of these resources will facilitate studies in these areas. The paucity of natural history studies in CMD made it difficult to base our recommendations on prior published work. However, it did offer the opportunity to provide recommendations on data collection in future studies. As the currently proposed CDE’s are put into use, there will surely be experience-based recommendations to improve their usefulness and the efficiency of data collection. The centralized nature of CDE document distribution and discussion will facilitate the evolution of these elements as we learn more about improving the documentation of data collection.
One key area of improvement that has already been implemented is the inclusion of patients and patient advocate representatives on current CDE discussion committees. While patient inclusion was not a feature of the CMD CDE discussion process several years ago, it has become apparent that the patient/family perspective is extremely useful when judging the usefulness of specific data elements. Thus, patients and advocates have now been included in the development of CDE’s on the more recent committees and we hope that they will also be active in the process of evolving these tools to optimize their usefulness in clinical research.
CMD WORKING GROUP MEMBERS
University of Michigan Hospital and Health Systems, Ann Arbor, MI, USA
National Institutes of Health, Porter Neuroscience Research Center, Bethesda, MD, USA
Cincinnati Children’s Medical Center, Cincinnati, OH, USA
The Hospital for Sick Children, Toronto, Ontario, Canada
University of Texas Southwestern Medical Center, Dallas, TX, USA
Medical College of Wisconsin, Children’s Hospital of Wisconsin, Milwaukee, WI, USA
University of California – San Diego, San Diego, CA, USA
University of Iowa Children’s Hospital, Iowa City, IA, USA
University College London, Institute of Child Health-Dubowitz Neuromuscular Centre, London, United Kingdom
Levine Children’s Hospital, Charlotte, NC, USA
NIH CDE TEAM
NINDS CDE Project Officer, National Institutes of Health/National Institute of Neurological Disorders and Stroke, (NIH/NINDS), Bethesda, MD, USA
Program Director, Office of Clinical Research, National Institutes of Health/National Institute of Neurological Disorders and Stroke, (NIH/NINDS), Bethesda, MD, USA
Program Director, Office of Clinical Research, National Institutes of Health/National Institute of Neurological Disorders and Stroke, (NIH/NINDS), Bethesda, MD, USA
Program Director, Extramural Research Program, National Institutes of Health/National Institute of Neurological Disorders and Stroke, (NIH/NINDS), Bethesda, MD, USA
Director, Patient Resources for Clinical and Translational Research, Office of Rare Diseases Research, National Institutes of Health, National Center for Advancing Translational Sciences (NIH/NCATS), Bethesda, MD, USA
Program Director, Intellectual and Developmental Disabilities Branch, Center for Developmental Biology and Perinatal Medicine, National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NIH/NICHD), Bethesda, MD, USA
The EMMES Corporation, Rockville, MD, USA
The EMMES Corporation, Rockville, MD, USA
CONFLICT OF INTEREST
Dr. Lawlor is a member of the scientific advisory boards for Audentes Therapeutics, and A Foundation Building Strength, and has been supported by sponsored research agreements by Audentes Therapeutics, Solid Biosciences, and Demter/Ichorion Therapeutics. He is also a scientific collaborator with Acceleron Pharma, Pfizer, and Valerion Therapeutics.
Dr. Muntoni is member of the DSMB of a clinical trial on CMD involving a drug developed by Santhera Pharmaceutics; and has received consulting fees from PTC Therapeutics, Sarepta Therapeutics, BioMarin, Roche, Biogen, Italfarmaco, Akashi Therapeutics, Avexis, Pfizer, Trivorsan and Catabasis.
Dr. Iannaccone serves on advisory boards (paid) for Sarepta, Santhera and Avexis and she receives research support from Biogen, Avexis, Sarepta, FibroGen and PTC Therapeutics as well as the MDA and NIH.
Dr. Mathews, Ms. Ala’i-Hansen, Ms. Odenkirchen, and Ms. Feldman have no conflicts of interest to report.
Footnotes
ACKNOWLEDGMENTS
The views expressed here are those of the authors and do not represent those of the National Institutes of Health (NIH), the National Institute of Neurological Disorders and Stroke (NINDS) or the US Government. Logistics support for this project was provided in part through NIH Contract HHSN271201200034C. The development of the NINDS CMD CDEs was made possible thanks to the great investment of time and effort of WG members and the members of the NIH CDE Project team participating from 2007 to 2015.
Michael Lawlor’s work in this program was supported by his involvement in the Congenital Muscle Disease Tissue Repository, which is supported by A Foundation Building Strength, Cure CMD, Where There’s A Will There’s A Cure, Audentes Therapeutics, Solid Biosciences and several additional private donors.
Francesco Muntoni is supported by the National Institute of Health Research Biomedical Research Centre at Great Ormond Street Hospital for Children NHS Foundation Trust and University College London. The support of Muscular Dystrophy UK and of the National Commissioning to the Dubowitz Neuromuscular Centre for the national role as the CMD Centre for UK is also gratefully acknowledged.