Watch time: 5 minutes
“Our research focused on how clinical trial designs could be improved. Conducting clinical trials for Alzheimer’s disease has traditionally been very difficult. Unfortunately, it hasn’t been very fruitful enough to yield more results: 99% of clinical trials in the last 20 years have failed in Phase 2 and 3 trials in the Alzheimer’s area.”
Enrolling patients who are less likely to show significant cognitive decline on placebo may make it more difficult to demonstrate the benefits of active treatment on cognition. data from a placebo group, showing that predictive machine learning models may increase sensitivity to treatment effects and reduce sample size requirements in clinical trials.1
In total, 1982 patients were included in the pooled placebo analysis and no significant cognitive decline was observed in 42% to 58% of individuals at the end of the study. Using predictive machine learning models, positive predictive values were approximately 12% to 25% higher than the sample rate of significant cognitive decline. Notably, the negative predictive value of the model was approximately 15% to 24% higher than the baseline rate for patients who had stable cognition at the end of the study.
Ali Ezzati, MD, Assistant Professor of Neurology at the Albert Einstein College of Medicine and Montefiore Medical Center, will be presenting the Dementia Experimental Therapies Session at the 2023 American Academy of Neurology (AAN) Annual Meeting on April 22nd. published this study in 27 years old, Boston, Massachusetts.During the meeting, Ezati sat with neurology liveIn an interview, ® discusses the reasons behind the difficulties and failures of Alzheimer’s disease (AD) clinical trials. He also discussed findings from his research that were presented and suggestions for using machine learning predictive models to improve exam design.
Click here for more information on AAN 2023.
