Improving the efficiency of Alzheimer’s disease clinical trials

Today Alzheimer’s Research & Therapy publishes research to improve the sensitivity of the Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-Cog) in clinical trials. Here, co-author of the work reveals more about what they did.

Alzheimer’s disease is the most common form of age-related dementia and affects over 44 million elderly people worldwide. The prevalence is expected to dramatically increase over the next few decades as life expectancy continues to improve across the world.

In Alzheimer’s disease, patients suffer from progressive damage to their brains leading to much faster cognitive decline than typically observed in the normal course of aging.

Why research Alzheimer’s?

Currently, there are no treatments available for preventing or slowing the progression of Alzheimer’s disease in the brain. Over the last decade, more than 400 treatments were studied in clinical trials for Alzheimer’s disease.

While the treatments showed promise in animal studies, they were unable to demonstrate a conclusive evidence of slowing disease progression in human trials. This puts the failure rate of Alzheimer’s clinical trials as the highest in comparison to any other disease including cancer.

A primary reason behind the high failure rate is believed to be the lack of a sensitive tool for measuring effects of treatments in clinical trials. Currently, a neuropsychological rating scale called the Alzheimer’s Disease Assessment Scale-Cognitive subscale (ADAS-Cog) is used as the standard outcome measure in clinical trials.

However, ADAS-Cog has been shown to suffer from poor sensitivity in tracking progression of Alzheimer’s disease in patients. As a consequence, even if a treatment did slow down Alzheimer’s progression, ADAS-Cog may not be able to detect it in clinical trials.

ADAS-Cog was originally developed nearly three decades ago.

ADAS-Cog was originally developed nearly three decades ago. Due to limited psychometric tools available at the time, a comprehensive analysis of the rating scale was not conducted.

Since then, the field of psychometric theory has undergone rapid advances and significantly impacted the development and the administration of rating scales, especially in the fields of education and psychology.

An example of such application is in intelligence testing such as Graduate Record Examinations, where respondents are scored on a standard intelligence scale using computer-administered adaptive tests.

What did we do?

In our study, we comprehensively studied the ADAS-Cog rating scale to identify the limitations that lead to low sensitivity in clinical trials. Our analysis indicated several problems associated with the items on the rating scale and the methodology used for scoring the items.

While improving the items on the ADAS-Cog is a longer term challenge and a topic of ongoing research, we hypothesized that simply addressing the limitations associated with its current scoring methodology may improve the performance of the ADAS-Cog in clinical trials.

Using a modern psychometric technique called as item response theory, we developed a new scoring methodology for the ADAS-Cog, which addresses most of the major limitations associated with its current scoring methodology.

We evaluated the new scoring methodology using a large dataset of over 5000 patients collected across three major cohorts of Alzheimer’s Disease Neuroimaging Initiative, Alzheimer’s Disease Cooperative Study and the Coalition Against Major Diseases, which included data from 12 previous clinical trials.

Using simulated clinical trials and real clinical trials, our results suggested that simply improving the scoring methodology of the ADAS-Cog leads to significant improvements in its sensitivity in clinical trials.

Using simulated clinical trials and real clinical trials, our results suggested that simply improving the scoring methodology of the ADAS-Cog leads to significant improvements in its sensitivity in clinical trials.

When data from a real negative trial of Huperzine A was retrospectively analyzed using the new scoring methodology, a significant treatment effect was detected which was missed in the original trial.

The findings of Huperzine A are supported by the fact that subsequent investigations of Huperzine A revealed similar evidence of a treatment effect.

Improved efficiency

This research has wide implications for future trials until a new improved rating scale is developed for use as an outcome measure.

The new scoring methodology helps in significantly improving the efficiency of clinical trials, which is highly desirable for rapid testing of future treatments in the critical question for a treatment for Alzheimer’s disease.

We recommend use of the new scoring methodology as part of the secondary efficacy analysis to further evaluate and establish its significance over the current scoring methodology.

View the latest posts on the On Medicine homepage

Comments