FormalPara Key Summary Points

Why carry out this study?

Adults with spinal muscular atrophy (SMA) experience significant fatigue.

The objective of this study was to measure fatigue in an adult SMA population using a variety of instruments.

What was learned from the study?

The instruments used to measure perceived fatigue did not find a consistent relationship between SMA severity and fatigue.

An SMA-specific scale is needed to evaluate differences in fatigue impact across the SMA population.

Introduction

SMA is an autosomal recessive neuromuscular disease characterized by progressive muscle weakness and atrophy [1,2,3]. SMA is caused by the loss or mutation of the survivor of motor neuron (SMN) 1 gene. There is a wide range of clinical severity in SMA, and the key determinant of disease phenotype is the copy number of SMN2, a nonfunctional variant of the SMN1 gene [4]. SMA has been historically classified into four types based on severity and age of symptom onset [3, 5,6,7,8,9,10]. The most severe and common type (accounting for approximately 60% of SMA births), type I, presents within the first six months of life. According to historical type classifications that predate approved disease-modifying therapies, babies with type I never achieve the ability to sit and usually require both ventilatory and feeding support, with the eventual use of permanent ventilation or death prior to the age of two years [11]. SMA type II, which accounts for about 30% of SMA cases, presents symptoms between 6 and 18 months of age, and while children may achieve the ability to sit independently, they will not be able to walk independently without treatment. About 10% of SMA cases are classified as SMA type III, in which symptoms appear after 18 months of age. Those with SMA type III may stand and walk independently, but lose these abilities over time [1, 12]. Lastly, those with SMA type IV, the rarest and least severe type of SMA, typically have onset of weakness in the second or third decade of life and experience mild motor impairment [13].

While SMA is often associated with pediatric patients—including infants and very young children—more than one-third of the global SMA population is estimated to be 18 or older [14]. This proportion is likely to increase over time as novel disease-modifying therapies improve outcomes and extend life spans.

There are currently three U.S. Food and Drug Administration (FDA) approved treatments for SMA. These include the antisense oligonucleotide nusinersen, approved in December 2016 for pediatric and adult SMA patients [15]; the gene therapy onasemnogene abeparvovec-xioi, approved in May 2019 for children under the age of 2 [15, 16]; and the small-molecule drug risdiplam, approved in August 2020 for patients 2 months of age and older and now approved for all ages [17]. Clinical trial data for all of these drugs have demonstrated improved survival and motor function [18].

Despite these advances, adults with SMA continue to experience significant mental and psychosocial impacts associated with their disease. While these impacts remain less well studied or understood [19] than those for pediatric patients, one oft-cited challenge is fatigue [20,21,22,23]. With fatigue, it is helpful to distinguish between the physical construct of fatigability—which is defined as “magnitude or rate of change in a performance criterion relative to a reference value over a given time of task performance or measure of mechanical output”—and perceived fatigue [24]. Perceived fatigue can be characterized by an overwhelming sense of tiredness, increasing sense of effort, lack of energy and motivation, and a feeling of exhaustion [25,26,27,28]. Previous studies have reported significant levels of perceived fatigue in adults with SMA. In one study, 81% of SMA patients complained of disabling fatigue and had higher severity scores than normal controls [29] and another study found that more than half of SMA patients had abnormal or severe levels of perceived fatigue [22].

There are many instruments available for measuring perceived fatigue, but only one developed specifically for use in SMA [30, 31]. The objectives of this study were to measure perceived fatigue in adults with SMA using five different fatigue instruments, evaluate the reliability of these instruments, and describe whether differences in fatigue levels were predicted by demographics, SMA type, SMN2 copy number, and treatment experience. Quantifying fatigue in an adult SMA population using a variety of instruments can provide a more complete perspective on its effects as well as baseline measures for future studies that assess perceived fatigue.

Methods

The survey was developed by Cure SMA and the Cure SMA Industry Collaboration (SMA-IC). Cure SMA is an SMA patient advocacy organization based in the United States that provides support and funding for the care and treatment of SMA and hosts the largest self-reported SMA membership database worldwide [32]. The SMA-IC was established in 2016 to leverage the experience, expertise, and resources of pharmaceutical and biotechnology companies, as well as other nonprofit organizations involved in the development of SMA therapeutics to address a range of issues. At the time of the study, the SMA-IC included Novartis Gene Therapies, Biogen, Genentech/Roche Pharmaceuticals, Scholar Rock, and SMA Europe.

All people with SMA who were in the Cure SMA membership database and over the age of 18 in December 2020 were invited to participate in the survey via email. The online survey was hosted on Alchemer, a cloud-based integrated feedback platform. Demographics, self-reported SMA type and SMN2 copy number, use of SMA treatments, and fatigue and quality of life measures were captured. Institutional review board (IRB) approval was obtained from Western IRB on November 24, 2020 (IRB ID: 20203852). Survey participants were informed of the intention to publish the anonymized results before they began the survey. All survey respondents provided informed consent to participate in the study on the survey landing page prior to being able to start the survey. The study procedures were in accordance with the 1964 Declaration of Helsinki and its later amendments. Data were de-identified before analysis.

In order to prevent survey fatigue, a randomization feature on Alchemer was utilized so that each respondent randomly received three of the following instruments: the Modified Fatigue Impact Scale (MFIS), Multidimensional Fatigue Inventory (MFI), Fatigue Severity Scale (FSS), PedsQL™ Multidimensional Fatigue (PedsQL MF) scale, and Spinal Muscular Atrophy Health Index (SMA-HI) fatigue and sleep modules [33,34,35,36]. The instruments were selected based on previous use in SMA and/or availability in the public domain. Table 1 describes each instrument.

Table 1 Fatigue instruments included in the study
Full size table

Statistical analyses

Cronbach’s alpha and inter-item correlations were computed to evaluate the reliability of each instrument except the SMA-HI. These values were not calculated for the SMA-HI due to missing SMA-HI short-form questions (see below). However, the SMA-HI was designed for use specifically in SMA and has been validated previously.

Mean scores and standard deviations (SD) were calculated for all scales and subscales. Since the SMA-HI short form was not implemented in this study, the SMA-HI sleep and fatigue subscales were each missing one item that would typically be included in the calculation of subscale scores. Using a predetermined statistical plan for missing data, we utilized the average response from the completed items to estimate the response for the missing items in each subscale prior to calculating subscale scores.

Multivariable regression models were used to evaluate the relationship between the fatigue scores and demographics, SMA type, SMN2 copy number, and SMA treatment status. SMA type was categorized into types I, II, and III, and unknown/other. Unknown/other SMA type included those with distal SMA, type IV SMA, and unknown SMA type. SMN2 copy number was categorized into ≤ 2 copies, 3 copies, 4 or more copies, and unknown. Treatment status was categorized as ever treated with an SMA disease-modifying therapy (DMT) and never treated with an SMA DMT. Type of DMT used was not captured in the survey.

Results

A total of 253 adults completed the online survey (5.93% with type I, 44.66% with type II, 44.27% with type III, and 5.14% other/unknown) (Table 2). The mean (SD) age at SMA diagnosis within the total sample was 8.13 years (12.86), with many having lived with SMA for years. The majority of participants (75.49%) reported having used an SMA DMT. The respondent subgroups for each scale or subscale were relatively similar according to the demographic and disease-related characteristics assessed.

Table 2 Characteristics of survey respondents
Full size table

Reliability statistics and mean (SD) fatigue scores for each scale or subscale are presented in Table 3. Cronbach’s alpha was greater than 0.90—which is the minimum accepted value for scales used for measurement on individuals—for the full MFIS, MFI, FSS, and PedsQL MF scales. However, the MFIS was the only scale with alpha values of 0.90 or above for all subscales with at least three items. Inter-item correlations varied widely (ranging from 0.29 to 0.74 for the scales and subscales). The relatively high number of IICs close to or exceeding 0.50 suggests that the items on several of the scales and their subscales may be too closely related and somewhat redundant.

Table 3 Scale reliability and scores
Full size table

Mean fatigue scores for respondents were higher than general population scores for all instruments for which comparators were available (see Table 3 vs. Table 1). Table 4 summarizes the factors associated with fatigue by instrument. Bold values indicate significant associations (p < 0.05) between fatigue and the independent variables.  Higher income was associated with lower fatigue for the MFIS total score, the MFI general score, and all three sub-scores of the PedsQL MF. Gender was not significantly associated with fatigue scores for any of the five instruments. Among SMA-specific outcomes, less severe SMA type was associated with lower fatigue levels in the cognitive MFIS score, but higher fatigue levels in the FSS and the general PedsQL MF. Treatment with an SMA DMT was associated with lower fatigue levels in the physical, cognitive, and total MFIS scores; the physical MFI score; the general PedsQL MF score; and the fatigue SMA-HI score.

Table 4 Factors associated with subscale fatigue score by instrument
Full size table

Discussion

The scales evaluated in this research demonstrated reasonable levels of reliability for use in SMA, although performance varied, and adaptations for the community may yield better performance. High inter-item correlations within certain scales and subscales suggest that their items may be repetitive, and scale modification could be appropriate for the SMA community. Future research may be useful in further confirming the validity and reliability of the scales, by more carefully and systematically examining the four dimensions of validity as well as test–retest reliability and inter-rater reliability. Ultimately, research to develop new or refined instruments that are tailored to the SMA community may yield the best understanding of fatigue within this population.

Examining predictors of perceived fatigue yielded a mixture of expected and surprising results. Given the impact of DMTs, it was not surprising that patients who had been treated with an SMA therapy had significantly lower fatigue scores than those who had never received treatment in the physical, cognitive, and total MFIS scores, as well as in the physical MFI score. However, given the significant variation in disease severity based on SMN2 copy number and SMA type, it was surprising that not all scales demonstrated differences in fatigue by SMA severity. One possible explanation is that the instruments used may be more likely to pick up on sociodemographic differences (e.g., age, education, quality of life) rather than SMA-specific distinctions [41].

Several aspects of our results have similarities to previous studies. For example, the physical MFI results correspond with those from a previous study by Binz et al. in which 75% of participants were abnormally fatigued, with the highest scores in the physical dimensions, followed by general fatigue and reduced activity [41]. Additionally, the FSS results correspond with those from a previous study by Dunaway et al. in which all type II and type III SMA patients reported perceived fatigue [38].

Overall, however, the inability to tease out major distinctions among subgroups suggests several potential conclusions. It is possible that there are no differences in perceived fatigue among various subgroups within the adult SMA population despite the varying impact that SMA has on individuals’ ability to pursue activities of daily living and enjoy a full life. Another conclusion is that motor function—and moreover differences in motor function by SMA type and copy number—is not related to perceived fatigue. This was also suggested by Dunaway Young et al. [22] when evaluating perceived fatigue in ambulatory SMA patients and reporting that they did not find differences in perceived fatigue between SMA type II and type III patients. On the other hand, however, given the heterogeneity among this population, it seems more likely that the lack of statistically significant differences identified in this study is due to insufficient specificity relating to fatigue within the available measurement instruments for this patient population.

Study Limitations

One significant limitation of this study is that current motor function was not collected in our survey. As a result, we could not evaluate perceived fatigue levels based on motor function—only based on SMA type. Future studies should evaluate current motor function by perceived fatigue levels, as variation and loss in motor functions exist across SMA types in all ages including adults. An additional limitation of this study pertains to the cohort of survey respondents. When considering how representative the sample is of the overall SMA adult population, it appears that this specific group of survey respondents (which includes a significant number of type I patients) may be healthier, older, and experiencing longer life with SMA and therefore potentially less significant perceived fatigue than the general adult SMA population. Natural history data indicate that survival among SMA type I is generally less than 2 years [11].

Conclusion

Fatigue is a significant factor in the lives of people living with SMA. Understanding the impact of fatigue upon patients’ ability to live their lives to the fullest is important to help focus efforts by the SMA community to reduce the burden of this disease. This effort by Cure SMA marks a step in the direction of developing this evidence base and defining a roadmap for progress. That said, to develop a more definitive understanding of how adult SMA patients perceive fatigue, it would be useful to develop a detailed, SMA-specific fatigue scale that could effectively tease out differences across subgroups within the SMA patient population as well as treatment effects. Of the five instruments used in this survey, only one was specific to the SMA population. Further, even that one instrument was not a complete tool focused on fatigue but rather the two modules from a broader survey aimed at assessing a variety of aspects of living with SMA. Further analysis of the data from this study is being contemplated to better understand the similarities and differences of specific tools and which items best represent perceived fatigue in the SMA population. These efforts are especially important to advance improved quality of life as more patients’ lives are extended well into adulthood with the development and approval of effective disease-modifying therapies.