Nearly 60% of ALS Patients are Excluded from Clinical Trials, Stressing the Need for New Eligibility Criteria, Study Shows

The majority of patients with amyotrophic lateral sclerosis (ALS) are excluded from participating in clinical trials, a large study has revealed.

This finding raises questions on the extrapolation of trial results for this patient population. The researchers also stressed the need for individualized risk-based criterion to balance the gains in trial design, and loss in generalizability.

The study, “Refining eligibility criteria for amyotrophic lateral sclerosis clinical trials,” was published in the journal Neurology.

ALS is a neurodegenerative disease that can have a very diverse clinical presentation and progression profile. Its inherent heterogeneity (symptom variability from patient to patient) can complicate the process of trial design.

To increase the efficacy of an investigational therapy, it is common for researchers to enroll a more homogeneous population. This approach allows better protocol adherence, but potentially excludes patients who are unlikely to benefit from the investigational treatment.

For ALS, this often means that patients who have had the disease longer, or those who are unlikely to survive the follow-up period will be excluded from trials. However, such strict criteria may compromise the final results, with low potential to provide useful information on safety and efficacy for the general patient population.

To further explore this issue, researchers at the University Medical Center Utrecht, in the Netherlands, reviewed the eligibility criteria of 38 ALS targeted clinical trials. Then they applied the eligibility criteria of these studies to a group of 2,904 patients, who were diagnosed with ALS in the Netherlands between January 2006 and December 2016.

The team found that, on average, 59.8% of these ALS patients would have been excluded from the clinical studies on the day of diagnosis. Changes in eligibility criteria over time led to some alteration in this exclusion rate, which increased from 53.3% between 2000 and 2010, to 65.5% between 2010 and 2017.

The most common reason for trial exclusion was patients not meeting a specific El Escorial category (23% were excluded), followed by their respiratory function as determined by forced vital capacity (17% were excluded), and required disease duration (12% were excluded).

The differences between the patients who were enrolled in the clinical trials and the actual Dutch ALS population were remarkable. Enrolled patients were mainly younger, male, and with a slower disease progression rate. They also had a better risk profile, with a longer survival prognosis than the general ALS population.

Collectively, these findings suggest that “despite the already applied eligibility criteria only a selective subset of the eligible patients will participate in trials,” the team wrote. “This finding is indicative of an additional latent selection process.”

The researchers could reduce the exclusion rates using the European Network to Cure ALS (ENCALS) survival model, a personalized prediction model, rather than the standard evaluation approaches.

Selection of patients with ENCALS risk scores between 0.55 and 3.3 — meaning that the risk of dying during follow-up is between half and three-fold the risk for the average ALS patient — would lead to a sample size reduction of 34%, which is similar to what happened in the trial for Radicava (edaravone) (NCT01492686).

With the ENCALS single selection criterion, it would be possible to increase the eligibility rate by five-fold, compared with the standard eligibility approaches used in the Radicava trial, the study said.

“People with ALS rely on us to use the best tools available to conduct rapid, efficient, effective trials. Incorporating prediction models into patient selection appears to be one of these tools,” Chafic Karam, MD, and James D. Berry, MD, MPH, experts in the ALS field, stated in an editorial published in the journal Neurology.

“As prediction-based algorithms and other sophisticated computational techniques make their way into clinical trials, perhaps we can look forward to ever more efficient trial designs,” they said.