Shortcomings in ALS Research: Strategies for Improvement

Share this content:
Challenges include poorly understood genetic complexity of amyotrophic lateral sclerosis, inadequate animal models in preclinical studies, and poor clinical trial design.
Challenges include poorly understood genetic complexity of amyotrophic lateral sclerosis, inadequate animal models in preclinical studies, and poor clinical trial design.

Despite their limited efficacy and survival benefit, riluzole and edaravone remain the only 2 drugs approved by the US Food and Drug Administration for the treatment of amyotrophic lateral sclerosis (ALS).

ALS is a progressive neurodegenerative disease that faces the well-documented inherent challenges of new drug development for very rare diseases. In a recent review by Nakul Katyal, MD, and Raghav Govindarajan, MD,1 the authors identified 5 key challenges that collectively hinder the design of ALS clinical trials and the development of new treatments. These challenges include the poorly understood genetic complexity of ALS, inadequate animal models in preclinical studies, poor clinical trial design, lack of sensitive biomarkers, and diagnostic delays.

Genetic Models

Mutations linked to 5 genes have been identified as being associated with familial ALS only in 5% to 10% of patients, with the remainder categorized as sporadic. The challenge, however, is that characteristics of familial and sporadic ALS overlap, often resulting in disease misclassification that can lead to errors in patient stratification and recruitment into clinical trials. Improved understanding of the ALS genotype, and designing clinical studies focused on a specific genetic model, is an approach suggested by the authors to improve the development of therapeutic agents. At this time, it is unclear whether specific tools exist or are in development to enable identification and stratification of patients with ALS according to their genotype, and thus enable the design of clinical studies focused on the individual genetic model. Further, considering the rarity of ALS and the relatively small patient population with familial ALS (5%-10%), there is a need for appropriate clinical design focused on investigating specific ALS genetic models.

Preclinical Models

Mice, rats, and canine models have been used in preclinical studies of ALS. Although disease characteristics and genetic mutations have been demonstrated in these models, the authors attest that the animal model is methodologically flawed. For example, in the superoxide dismutase 1 mice models, treatment was often started before the onset of symptoms, unlike patient cases where treatment is initiated after symptom onset.2 The neuroprotection provided by the advanced treatment in the animal models may result in overestimation of clinical effect. Further, differences in sex presentation in the superoxide dismutase 1 mice model, with earlier symptom onset and earlier deaths among males compared with females, were not taken into consideration. Failure to stratify the data according to sex can result in drawing conclusions that may be incorrectly associated with a possible drug effect. To limit potential erroneous and misleading conclusions from poorly designed clinical trials, the authors summarized recommendations from the ALS Therapy Development Institute trial, which include separating male and female mice, gene tracking (because not all disease-causing genes are transferred to subsequent generations), and randomization and blinding to minimize spurious conclusions. Despite the recommendations, the value of animal models in ALS drug development remains questionable, given that the authors acknowledge that the current animal models have several methodological flaws. A discussion may be warranted on how animal models can be improved, or on alternative models to replace them.

Clinical Trial Design

Clinical trial design presents significant challenges to the development of new ALS therapies. To date, not 1 of the 18 different candidate drugs tested has shown clinical efficacy. The challenges identified by the authors include appropriateness and sufficiency of drug dosing, poor penetration of the blood-brain barrier, and failure to take into consideration confounding variables of drug interactions from prior treatment such as riluzole. In addition, there is inherent bias for the publication of positive results, which can result in unnecessary repetition of poorly designed trials and negative results. The authors recommend modification of trial design and route of drug delivery by optimizing intrathecal and intramedullary routes to bypass the blood-brain barrier while separating patients according to their phenotype. The authors also recommend modification of the traditional study design by dividing clinical trials into learning phase (focused on exploring toxicity, drug interaction, and pharmacokinetic and pharmacodynamic parameters of the investigational agent) and confirmatory phase (focused on the traditional phase 3 randomized controlled trial design).

It is, however, unclear how the learning phase differs from the traditional preclinical and early-phase clinical trial designs that focus on establishing toxicity, drug interaction, pharmacokinetic and pharmacodynamic parameters, and dosing. It is also unclear how a separation of learning phase and confirmatory phase will improve trial efficacy and statistical power. Given the small patient population that challenges traditional randomized controlled trials, alternative trial designs should be considered, such as open-label, factorial designs, N-of-1 studies, or crossover studies.


There are currently no specific ALS biomarkers or surrogate endpoints to reliably monitor disease progression and treatment response. Although several biological fluid biomarkers have been identified with potential benefits for disease prognosis, progression, and pharmacodynamic properties, interpretation of results can be challenging and errors are known to occur. Further, several confounding factors including age, ethnicity, sex, and comorbidity, as well as differences in sample collection, processing, and storage, can affect the clinical relevance of biomarkers. Electrophysiological markers have also been used to extensively diagnose and monitor ALS progression and treatment response. The challenge, however, is that electrophysiological markers are not reliably consistent, with variation in results depending on the electrode size, limb temperature, and positioning.

To address the challenges associated with the poor reliability of the currently used ALS biomarkers and other markers, the authors suggest adopting the US Food and Drug Administration guidelines for the development of biomarker assays. In addition, the authors advocate for collaborative efforts between pharmaceutical companies and academic centers to develop an ALS biomarker consortium, similar to the Parkinson Progression Markers Initiative and the Parkinson Disease Biomarkers Program.

Disease Awareness

The clinical presentation of ALS varies widely from predominantly upper or lower motor neuron symptoms to nonmotor neuron symptoms, which can mimic other diseases, posing significant challenges to early differential diagnosis. Lack of clinical awareness and delayed referral to neurologists contribute to these challenges. The accuracy of the currently used El Escorial and Awaji diagnostic criteria3,4 is limited because of the wide variation of ALS symptoms. As a consequence, ALS diagnosis is often delayed, and treatment, including entry into clinical trials, is often initiated after the disease is significantly advanced. The authors recommend a collaborative effort between primary care physicians and neurologists, together with the involvement of patient organizations, to raise public awareness of ALS. Awareness of early symptoms of the disease, prompt referral to a neurologist for diagnosis and treatment, and increasing availability and access to clinical trial centers, including a universal enrollment and monitoring criteria for affected patients, should be strongly considered as approaches to improving ALS treatment and outcomes.


  1. Katyal N, Govindarajan R. Shortcomings in the current amyotrophic lateral sclerosis trials and potential solutions for improvement. Front Neurol. 2017;8:521.
  2. Shibata N. Transgenic mouse model for familial amyotrophic lateral sclerosis with superoxide dismutase-1 mutation. Neuropathology. 2001;21(1):82-92.
  3. Brooks BR, Miller RG, Swash M, Munsat TL. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000;1(5):293-299.
  4. Costa J, Swash M, De carvalho M. Awaji criteria for the diagnosis of amyotrophic lateral sclerosis: a systematic review. Arch Neurol. 2012;69(11):1410-1416.
You must be a registered member of Neurology Advisor to post a comment.

Sign Up for Free e-newsletters

CME Focus