In conclusion, during one year of natural course, muscle strength declined about 2% as measured by CIDD, and 76 N as measured by HHD. These findings were supported by comparable worsening of the secondary outcome measurements Handgrip, NSS, Vignos Scale and HMAS. Arguably, scores derived from manual muscle testing do not represent interval data; thus one cannot accurately measure percentage change. However, using more quantitative measures of strength demonstrated a similar lack of benefit as did the other secondary outcome measures. Additionally, Vignos and HMAS are fairly rough scales, not suitable for identifying more subtle changes over a one year period in a slowly progressive muscle disease. Vignos Scale and HMAS both showed worsening, slightly in favor of placebo. The Hammersmith Motor Ability Score thereby showed overall worsening, not restricted to walking ability. Interestingly, timed tests did not detect relevant changes illustrating that tests measuring changes in complex activities in milliseconds may not perfectly mirror and have an impact on changes in daily life. Additionally, QSMT by torque measurement die not prove useful to detect reliable changes. Patient-reported quality of life assessments remained fairly unchanged during this first year or improved slightly, mainly in the dimensions, which were impaired in this study population (physical functioning, mental health, standardized physical component scale).
Interestingly, CK levels remained stable during placebo administration, and decreased during deflazacort treatment, but non-corresponding with clinical improvement. This finding confirms that steroids are not just ineffective in dysferlinopathy, but show different activities, thereby enhancing the internal validity of this clinical trial.
Non-recoverable loss of strength in patients with dysferlin-deficiency, who were initially misdiagnosed as inflammatory myopathy and treated with corticosteroids, had been suggested by Hoffmann et al.  in an uncontrolled series of 20 patients prior to this study.
Interestingly, we detected reduced dysferlin levels in patients with inflammatory myopathies (polymyositis, inclusion body myositis) treated with steroids (unpublished data). If a reduction in dysferlin-expression would occur in dysferlinopathy patients, then it would affect patients with residual or partially functional dysferlin protein more than patients without any protein. However, we did not obtain muscle biopsies of the patients in our study during or after treatment.
In our trial, patients did not improve during the deflazacort period of treatment. In contrast, under deflazacort treatment, there was a trend to muscle strength worsening, which recovered after discontinuation of the study drug during the wash-out period. In patients with residual dysferlin protein levels, the negative effect of deflazacort was more pronounced than in patients with absent dysferlin protein. However, due to the small sample size of the subgroups, this finding is mainly hypothetic and has not been described before, but we may speculate that steroids have a negative effect on dysferlin expression or function.
In contrast, Belanto et al.  hypothesized that a key therapeutic benefit of glucocorticoids may be the up-regulation of dysferlin as an important component of glucocorticoid-enhanced myogenic differentiation. However, this was based on experimentation in cultured C2C12 cells that may not be an adequate model for adult, dystrophic muscle. Further work is required to understand the exact molecular action of steroids on dysferlin expression or function in mammalian muscle, whether it is a direct or an indirect effect.
The SF-36 scores in subscales physical role function and general health perception as well in the standardized physical component scale at baseline showed similar values in dysferlinopathies as in other chronic diseases, and improved slightly during the 1-year natural history phase, possibly due to better patient support and care during the ongoing clinical trial. No impaired quality of life compared to the German general population were observed in the emotional subscales of the SF-36. During deflazacort treatment - in line with the results of the other outcome measures, e.g. strength testing - SF-36 ratings worsened in all subscales, but remained stable or improved slightly during placebo administration, most pronounced for mental health, emotional role and vitality subscales and the standardized mental component scale. The CGI global severity score did not change comparing starting- and endpoint of the trial, suggesting that relevant differences of disease severity were not fully recognizable by the investigator within 27 months of the trial duration.
During deflazacort treatment, we observed a broad spectrum of known side effects. Our study shows that deflazacort is not an effective therapy for dysferlinopathies, and may even harm patients. Therefore, off-label use is not warranted. Even if this result is not what we would have hoped for, it is nevertheless an important finding, since there is good reason now to definitely withhold steroid treatment from patients with dysferlinopathy.
In this study, we were able to establish data regarding the natural course of dysferlinopathies that improve our understanding of the clinical problems these patients are confronted with and may serve as reference for future studies. As measures of muscular function in our sample of juvenile and adult ambulatory patients, CIDD, HHD and NSS have proven useful in showing a change in muscle strength and daily activities within one to two years, while timed tests seemed not suitable. Additionally it became clear that a quality of life measure, specifically adapted on patients with muscle diseases would be warranted, since SF-36 and CGI do not adequately mirror the specific problems of muscular dystrophy patients.
Although no major therapeutic breakthrough has been achieved and curative treatment modalities are not yet applicable, life expectancy and quality of life of dysferlinopathy patients could be remarkably improved by establishing a drug therapy, capable of delaying the dystrophic process and improving muscle strength and function. Unfortunately, no such therapy is available yet, and treatment of dysferlinopathies is mainly based on symptomatic treatment. Therefore, the results of this clinical trial – even being negative - are warranted and may influence further guidelines for steroid treatment in dysferlinopathies. Furthermore, our assessment of the natural history of the disease will provide new insights in the clinical understanding of dysferlinopathies .