Digital biomarkers of cognitive decline could be the antidote to the slop in the subjective tests on which clinical neurology programs often hinge. With proof of concept established in small clinical trials and bigger readouts on the way, data should emerge in the next year or two on the feasibility for digital readouts to replace standard endpoints in clinical trials for cognition.
The gold standard for assessing cognitive function in the clinic has been investigator-administered surveys that score patients on a battery of questions that assess memory, problem solving and other abilities.
But the subjective nature of these surveys means they suffer from variability on multiple fronts, both from the same patient across different office visits and between different patients and physicians. In addition, because the surveys are administered once a month at most, they only capture snapshots of a patient’s behavior - usually in the artificial setting of the doctor’s office - and may not accurately reflect her or his day-to-day cognitive function.
Despite these shortcomings, investigator-administered surveys still dominate as endpoints used to decide if a compound should advance to large late-stage trials or if it is worthy of FDA approval.
“People have pointed out that there is a problem with the translatability