M. Waite

Results of a two-year pilot study of clinical outcome measures in collagen VI- and laminin alpha2-related congenital muscular dystrophies

Potential therapies are currently under development for two congenital muscular dystrophy (CMD) subtypes: collagen VI-related muscular dystrophy (COL6-RD) and laminin alpha 2-related dystrophy (LAMA2-RD). However, appropriate clinical outcome measures to be used in clinical trials have not been validated in CMDs. We conducted a two-year pilot study to evaluate feasibility, reliability, and validity of various outcome measures, particularly the Motor Function Measure 32, in 33 subjects with COL6-RD and LAMA2-RD. In the first year, outcome measures tested included: Motor Function Measure 32 (MFM32), forced vital capacity (FVC) percent predicted sitting, myometry, goniometry, 10-meter walk, Egen Klassification 2, and PedsQL(TM) Generic and Neuromuscular Cores. In the second year, we added the North Star Ambulatory Assessment (NSAA), Hammersmith Functional Motor Scale (HFMS), timed functional tests, Measure of Activity Limitations (ACTIVLIM), Quality of Upper Extremity Skills Test (QUEST), and Patient-Reported Outcomes Measurement Information System (PROMIS) fatigue subscale. The MFM32 showed strong inter-rater (0.92) and internal consistency (0.96) reliabilities. Concurrent validity for the MFM32 was supported by large correlations (range 0.623-0.936) with the following: FVC, NSAA, HFMS, timed functional tests, ACTIVLIM, and QUEST. Significant correlations of the MFM32 were also found with select myometry measurements, mainly of the proximal extremities and domains of the PedsQL(TM) scales focusing on physical health and neuromuscular disease. Goniometry measurements were less reliable. The Motor Function Measure is reliable and valid in the two specific subtypes of CMD evaluated, COL6-RD and LAMA2-RD. The NSAA is useful as a complementary outcome measure in ambulatory individuals. Preliminary concurrent validity of several other clinical outcome measures was also demonstrated for these subtypes.

English Cross-Cultural Translation and Validation of the Neuromuscular Score: A System for Motor Function Classification in Patients With Neuromuscular Diseases

OBJECTIVE To develop and validate an English version of the Neuromuscular (NM)-Score, a classification for patients with NM diseases in each of the 3 motor function domains: D1, standing and transfers; D2, axial and proximal motor function; and D3, distal motor function. DESIGN Validation survey. SETTING Patients seen at a medical research center between June and September 2013. PARTICIPANTS Consecutive patients (N=42) aged 5 to 19 years with a confirmed or suspected diagnosis of congenital muscular dystrophy. INTERVENTIONS Not applicable. MAIN OUTCOME MEASURES An English version of the NM-Score was developed by a 9-person expert panel that assessed its content validity and semantic equivalence. Its concurrent validity was tested against criterion standards (Brooke Scale, Motor Function Measure [MFM], activity limitations for patients with upper and/or lower limb impairments [ACTIVLIM], Jebsen Test, and myometry measurements). Informant agreement between patient/caregiver (P/C)-reported and medical doctor (MD)-reported NM scores was measured by weighted kappa. RESULTS Significant correlation coefficients were found between NM scores and criterion standards. The highest correlations were found between NM-score D1 and MFM score D1 (ρ=-.944, P<.0001), ACTIVLIM (ρ=-.895, P<.0001), and hip abduction strength by myometry (ρ=-.811, P<.0001). Informant agreement between P/C-reported and MD-reported NM scores was high for D1 (κ=.801; 95% confidence interval [CI], .701-.914) but moderate for D2 (κ=.592; 95% CI, .412-.773) and D3 (κ=.485; 95% CI, .290-.680). Correlation coefficients between the NM scores and the criterion standards did not significantly differ between P/C-reported and MD-reported NM scores. CONCLUSIONS Patients and physicians completed the English NM-Score easily and accurately. The English version is a reliable and valid instrument that can be used in clinical practice and research to describe the functional abilities of patients with NM diseases.

Electrical impedance myography in individuals with collagen 6 and laminin α-2 congenital muscular dystrophy: a cross-sectional and 2-year analysis: EIM in COL6 and LAMA2 CMD

Electrical impedance myography (EIM) is a noninvasive electrophysiological technique that characterizes muscle properties through bioimpedance. We compared EIM measurements to function, strength, and disease severity in a population with congenital muscular dystrophy (CMD).

715. Giant Axonal Neuropathy - The Role of Natural History Studies in Gene Transfer Therapy Trial Design

OBJECTIVE: Giant axonal neuropathy (GAN) is a rare childhood onset autosomal recessive neurodegenerative disorder affecting the central and peripheral nervous system. Mutations in the GAN gene cause loss of function of gigaxonin, a cytoskeletal regulatory protein, clinically leading to progressive sensorimotor neuropathy and neuropathy, reduced coordination, slurred speech, seizures, and progressive respiratory failure leading to death by the 2nd to 3rd decade of life. We are in the midst of a first-in-human intrathecal AAV9 mediated gene transfer trial for GAN. Given the ultra-rare nature of the disease, it is critical to build disease appropriate outcome measures and have lead in safety and outcome data to feed into the clinical trial. Our aim here was to identify and develop targeted quantitative markers of disease severity in GAN. METHODS: This natural history study evaluated measures of motor, neurophysiologic, and ophthalmologic function as well as exploratory neuroimaging markers in genetically confirmed GAN patients seen at the National Institutes of Health (NIH). The primary aim was to correlate a quantitative motor scale (MFM32) with a semi-quantitative Neuropathy Impairment Score (NIS) in cross sectional analysis of GAN patients. Secondary aims included evaluation of strength (myometry) and motor nerve amplitude (nerve conduction) compared to motor function (MFM32). Quantitative ophthalmologic testing included retinal nerve fiber layer thickness analysis correlated to visual acuity, MFM32, and motor nerve amplitude. Additional exploratory measures of spinal cord volume, brain DTI, quantitative MR spectroscopy, and biochemical (in serum and CSF) measures are underway. RESULTS: 15 GAN patients were evaluated at the NIH. Data analysis (Pearson correlation) showed the following: 1) NIS and MFM32 are significantly correlated (p<0.0001); 2) Muscle strength of knee extension, knee flexion, hip abduction, and hand grip each correlate significantly with MFM32 (p values from <0.0001 to 0.039) ; 3) Median motor nerve amplitude correlates significantly with MFM32 (p=0.0016). There was no significant correlation between elbow extension or elbow flexion strength and MFM32. Retinal nerve fiber layer corresponds strongly with MFM32 and the median motor nerve amplitude (p<0.0001 and p=0.005, respectively). CONCLUSIONS: The phase I intrathecal gene transfer study for GAN is a novel clinical trial, and will set a precedent by proof of principle for future gene transfer trials for neurodegenerative disorders. The primary outcome is safety, but given that the trial entails administration of a presumed effective dose, careful selection to outcome measures was essential in the protocol design. Through this natural history study, we have identified markers such as the MFM32, NIS, median motor amplitude, and retinal nerve fiber layer thickness, as well as spinal cord volumetric data that clearly correlate with disease progression. These markers are being followed longitudinally in natural history and were designed into the phase I trial. Such broad data capture allows for the incorporation of more targeted disease specific clinically meaningful outcome measures and data on natural rate of progression of such markers over short durations of follow up. Such models of trial design will be crucial to the development of genetic therapeutic trials for rare neurodegenerative disorders.

Correlation of phenotype with genotype and protein structure in RYR1-related disorders

Variants in the skeletal muscle ryanodine receptor 1 gene (RYR1) result in a spectrum of RYR1-related disorders. Presentation during infancy is typical and ranges from delayed motor milestones and proximal muscle weakness to severe respiratory impairment and ophthalmoplegia. We aimed to elucidate correlations between genotype, protein structure and clinical phenotype in this rare disease population. Genetic and clinical data from 47 affected individuals were analyzed and variants mapped to the cryo-EM RyR1 structure. Comparisons of clinical severity, motor and respiratory function and symptomatology were made according to the mode of inheritance and affected RyR1 structural domain(s). Overall, 49 RYR1 variants were identified in 47 cases (dominant/de novo, n = 35; recessive, n = 12). Three variants were previously unreported. In recessive cases, facial weakness, neonatal hypotonia, ophthalmoplegia/paresis, ptosis, and scapular winging were more frequently observed than in dominant/de novo cases (all, p < 0.05). Both dominant/de novo and recessive cases exhibited core myopathy histopathology. Clinically severe cases were typically recessive or had variants localized to the RyR1 cytosolic shell domain. Motor deficits were most apparent in the MFM-32 standing and transfers dimension, [median (IQR) 85.4 (18.8)% of maximum score] and recessive cases exhibited significantly greater overall motor function impairment compared to dominant/de novo cases [79.7 (18.8)% vs. 87.5 (17.7)% of maximum score, p = 0.03]. Variant mapping revealed patterns of clinical severity across RyR1 domains, including a structural plane of interest within the RyR1 cytosolic shell, in which 84% of variants affected the bridging solenoid. We have corroborated genotype-phenotype correlations and identified RyR1 regions that may be especially sensitive to structural modification.

Upper extremity outcome measures for collagen VI-related myopathy and LAMA2-related muscular dystrophy

Congenital muscular dystrophy (CMD) comprises a rare group of genetic muscle diseases that present at birth or early during infancy. Two common subtypes of CMD are collagen VI-related muscular dystrophy (COL6-RD) and laminin alpha 2-related dystrophy (LAMA2-RD). Traditional outcome measures in CMD include gross motor and mobility assessments, yet significant motor declines underscore the need for valid upper extremity motor assessments as a clinical endpoint. This study validated a battery of upper extremity measures in these two CMD subtypes for future clinical trials. For this cross-sectional study, 42 participants were assessed over the same 2-5 day period at the National Institutes of Health Clinical Center. All upper extremity measures were correlated with the Motor Function Measure 32 (MFM32). The battery of upper extremity assessments included the Jebsen Taylor Hand Function Test, Quality of Upper Extremity Skills Test (QUEST), hand held dynamometry, goniometry, and MyoSet Tools. Spearman Rho was used for correlations to the MFM32. Pearson was performed to correlate the Jebsen, QUEST, hand-held dynamometry, goniometry and the MyoSet Tools. Correlations were considered significant at the 0.01 level (2-tailed). Significant correlations were found between both the MFM32 and MFM Dimension 3 only (Distal Motor function) and the Jebsen, QUEST, MyoGrip and MyoPinch, elbow flexion/extension ROM and myometry. Additional correlations between the assessments are reported. The Jebsen, the Grasp and Dissociated Movements domains of the QUEST, the MyoGrip and the MyoPinch tools, as well as elbow ROM and myometry were determined to be valid and feasible in this population, provided variation in test items, and assessed a range of difficulty in CMD. To move forward, it will be of utmost importance to determine whether these upper extremity measures are reproducible and sensitive to change over time.

T.P.42

Comparisons of clinical outcome measures of motor performance to be used in clinical trials in Congenital Muscular Dystrophy subtypes are limited. However, redundancy, discriminant validity, and ceiling and floor effects of motor function measures are an important consideration in optimizing the use of outcome measures clinical trials. Methods: We conducted a study to compare the Motor Function Measure 32 (MFM32), the Hammersmith Functional Motor Scale (HFMS), and the North Star Ambulatory Assessment (NSAA) in 29 subjects with COL6-RM and LAMA2-RD. We analyzed concurrent validity using Spearman rank order correlation, discriminant analysis using Wilcoxon Mann Whitney test, and floor and ceiling effects using a cut off of 20% for each item. We hypothesized that the HFMS and NSAA would expand the motor function assessment of the MFM32 in both extremes of functional limitations and disease severity (NSAA for higher functioning/less severe and HFMS for lower functioning/more severe). Concurrent validity for the MFM32, HFMS, and NSAA was strong as evidenced by large correlations (ranging 0.76–0.93, p p

P.10.10 The Rasch-scaled motor function measure for patients with congenital disorders of muscle

Given recent progress toward therapeutic interventions in the congenital disorders of muscle (CDM), the validation of appropriate outcome measures for patients with congenital muscular dystrophies and myopathies becomes a high priority in order to precisely characterize natural history and the effects of treatments in future clinical trials. The aim of our study was to examine whether the Motor Function Measure (MFM), an ordinal functional scale designed for neuromuscular diseases, would fulfill all Rasch model expectations in a sample of 289 patients with CMD and to propose changes to improve its reliability and sensitivity to change. Three distinct analyses were performed considering the 3 domains (D1 Standing position and transfers, D2 Axial and proximal and D3 Distal motor function) of the MFM using RUMM 2030 software with a partial credit model. The original MFM failed to meet the model expectations in each domain. The reduction to a three-categorie response (0,1,2) scale in 20 items with disordered thresholds, produced ordered response thresholds for 18 items. Seven misfitting items were identified as not fitting and removed. Finally, the Rasch-scaled 10-item MFM D1, 9-item MFM D2 and 6-item MFM D3 conformed adequately to the Rasch measurement model. The Rasch-scaled-MFM D1 was well targeted to the patient population with quite similar mean location for items (0.000) and person (0.316) whereas the Rasch-scaled-MFM D2 and D3 were better targeted for more severely affected patients, as expected from the domains covered. Reliability coefficients of the Rasch scaled MFM suggest sufficient ability for the total score D1 (0.9) and D2 (0.8) and D3 (0.7) to distinguish between groups for research use. The results provide evidence for the measurement properties of the Rasch-scaled MFMCDM promoting its use as outcome measures in clinical trials for patients with congenital disorders of muscle. A raw score-to-Rasch person measure conversion is supplied for investigators.