Outcome Measures in Neuromuscular Diseases

Introduction: What are Outcome Measures?

Good outcome measures are a critical component of any successful therapeutics evaluation program. Although some pediatric neuromuscular diseases are now seeing discoveries translated into early-phase clinical trials, there remains a great need for better therapeutics. Bringing these therapeutics to patients requires clinical trials using outcome measures that ideally are easy to implement, are sensitive to change, and reflect an important aspect of the disease.

Outcome measures can be biomarkers, meaning that they measure a biological characteristic of the disease in question. Biomarkers can be metabolites measured in blood or cerebrospinal fluid, electrophysiological measures, or imaging markers. Outcome measures can also be clinical measures, meaning that they measure something that can be reported by observers such as clinicians or caregivers, or by the patient directly.

The goal of using these biomarkers or clinical outcome measures is to assess changes that occur over the course of a clinical trial. Biomarkers or clinical measures can also be used in different contexts, such as when diagnostic or prognostic markers are sought. However, the overview in this chapter will focus only on measures of disease activity or progression, and will discuss them in the context of their potential usefulness in clinical research.

To set the stage for successful clinical trials, researchers need not only patients who are ready to participate, but also high-quality datasets on the natural history of the disease using outcome measures suitable for clinical trials. Collecting natural history data requires significant resources, and an uncommon degree of altruism on the part of patients and their families, because there is no potential for direct benefit to the participant. To lessen the burden added by the study visits, it can be helpful if outcome measures can be collected at the same time that clinical care is provided. Conversely, collecting a small set of outcome measures during clinical visits can also help patients and their care team with charting disease progression. The outcome measures information can help guide care, and can help better understand how the patient’s disability affects function.

Developing and validating outcome measures requires time and resources, and is associated with burden to patients and research teams. Therefore, progress in a given field can be accelerated when research teams can agree on the use of a common set of outcome measures. Using a common set of measures across sites leads to larger datasets being available in a shorter timeframe, so that outcome measures can be developed and tested more efficiently. Avoiding the development and use of multiple outcome measures focused on the same concept and indication can save time and money. In situations when no good outcome measure is available to help address a question, it is preferable if researchers turn to existing outcome measures rather than starting anew in developing one. For example, using an existing scale, or part thereof, can lead to more rapid validation of a potential outcome measure, and can also create opportunities to address important questions across diseases, for example by comparing datasets for two different neuromuscular diseases.

Outcome measures are better delineated for some neuromuscular diseases than others; to a degree, this reflects how far along the respective fields are in terms of therapeutics development. This overview will focus on spinal muscular atrophy (SMA) and Duchenne muscular dystrophy (DMD), but several of the principles and in some cases even the outcome measures discussed may be applicable to a wider range of pediatric neuromuscular disorders.

Outcome Measures in Different Stages of Therapeutics Development

The hallmark of pediatric neuromuscular disorders is muscle weakness. Muscle weakness can be directly measured using manual or quantitative muscle testing. It can also be measured more indirectly in terms of its effect on motor function. For pivotal clinical trials, an outcome measure that is inherently related to how a patient functions or feels in their everyday life is more meaningful than a measure that quantitates strength but cannot be interpreted in terms of its functional consequences. Weakness can affect muscles other than skeletal muscles; respiratory muscle weakness, for example, is of interest as an outcome in most neuromuscular diseases. Similarly, involvement of bulbar muscles may prompt use of outcome measures assessing bulbar function. In a subset of pediatric neuromuscular disorders, cardiac muscle can be affected as well.

When muscle function or strength cannot be easily measured, or when it is not clear if clinical measures of muscle function would change within the observation period of a trial, then biomarkers may be useful. This is particularly relevant in early phase (Ib or II) trials. In this scenario, it is hoped that biomarkers may be more sensitive to early treatment effect than are clinical measures.

In rare instances in medicine, biomarkers can be surrogates of downstream clinical outcomes. Hypertension, for example, is thought to predict the downstream risk of cardiovascular events. In most instances, this kind of strong association cannot be established. The expectation for pediatric neuromuscular disorders at this time is not to establish a surrogate marker in the foreseeable future, but rather to use biomarkers in a therapeutics development program as an early signal of activity, to be followed by later stage trials with observations using clinical markers.

Another type of biomarker is not specific to the disease and its hallmark features, but is specific to the presumed mechanism of action of the intervention. These kinds of markers of biological activity can be very useful in early proof of mechanism trials when it is important to establish that the intervention has the desired biological effect in patients.

Clinical outcome measures ideally reflect an aspect of the disease that matters to patients. In other words, clinical outcome measures should reflect how a patient feels or functions. This information is best obtained from the patient directly. When assessments of functional status are obtained from an observer, such as a relative, caregiver, or clinician, then the outcome measure captures not only the actual function of the patient, but also the subjective perception and opinion of the observer. Therefore, any measure that captures information from the patient as a source is preferable to one that captures information indirectly.

Clinical outcome measures are, ideally, easily interpretable and clinically meaningful. When an outcome measure focuses on a function relevant to the patient’s daily living activities, the clinical meaning and relevance of any change can be easily understood and communicated. To ensure the meaningfulness of an outcome, experts in outcomes development often start with focus groups to determine which domains are most important to patients, rather than starting with what clinicians think is most important. When outcome measures contain a large number of disparate items, or when they mix functional information with clinical observation and physical examination, then the interpretation of any change is often unclear.

Different outcome measures have variable places in the spectrum of drug development. In early human trials (phase 0 and Ib), biomarkers of mechanism, to demonstrate that the intervention hits its target, are useful. In early exploratory efficacy trials (phase IIa), biomarkers that can signal an early effect of the intervention can be useful. Clinical efficacy trials (phase IIb or phase III) mandate clinical outcomes that ideally indicate how a patient feels, functions, or survives. Figure 54.1 summarizes the drug development process.

Clinical Outcome Measures

The ideal clinical outcome measure is age appropriate, easily administered, functionally meaningful, and imposes minimal patient burden. In rare neuromuscular conditions where enrollment in clinical trials requires multiple international sites, ideal outcome measures are those which require minimal training and equipment to allow participation of investigators with limited clinical resources.

It is best to build on existing tools when choosing an outcome measure because one can take advantage of any data available for the existing tools, compare data to published cohorts in the same or another condition, and take advantage of any regulatory pathway already established for the outcome measure. Outcome measures designed specifically for a particular condition are ideal but not mandatory since children with weakness have similar functional limitations.

There is a broad clinical phenotype both across different pediatric neuromuscular conditions as well as within individual disorders. For example, in SMA, there is a spectrum of disease severity and functional ability, ranging from infants and children never achieving the ability to sit unsupported, to those only able to sit but unable to stand and walk, and to those who can walk but have weakness limiting their ability to rise from the floor, run, and jump. It is difficult to envision a single motor performance-based measure assessing functionally meaningful abilities for all SMA types. Similarly, boys with DMD transition from an ambulatory stage to a nonambulatory stage in which gross motor function is significantly reduced, and upper limb function is a more relevant potential outcome measure. One must consider if the population of patients enrolled in a clinical trial will lose function during their participation that would not be captured on the outcome measure they choose because of a floor effect. For example, if a boy with DMD loses ambulation during a clinical trial where a walking test is the primary outcome measure, without a nonambulatory assessment any further changes in the participant’s clinical function would not be identified.

We have categorized the different motor function outcome measures below into those for ambulatory and nonambulatory children. Careful consideration with regards to the best clinical outcome measure based on age, disease severity, functional level, and rate of progression must be made when designing a clinical trial.