Detecting an MCI and Amyloid Digital Neuro Signature(DNS) Using Altoida's Multimodal Digital Biomarkers.

Overview

The purpose of this study is to establish multiple points of clinical validity for the Altoida digital biomarker assessment in individuals with a clinical diagnosis of cognitively normal (CN) and Mild Cognitive Impairment (MCI). Participants will use the Altoida app and the de-identified sensor data captured by the device will be used to train specific machine-learning algorithms to recognize early symptoms of cognitive decline, such as MCI or MCI with likelihood of amyloid pathology, as measured by digital biomarkers (T1 - Visit 1). Participants will be invited for an additional visit to evaluate test-retest reliability (T1' - Visit 2). Optionally, an updated variation of the Altoida app will be tested over the course of two additional visits to ensure optimal digital assessment delivery based on best practices in neuropsychological testing, user experience design, and data collection integrity (T2 - Visit 3 and T2' - Visit 4).

Digital biomarkers are indicators of a person's health status as measured by a digital device. They are gaining traction in neurology research for their capacity to offer accurate, accessible, and continuous measures of cognitive performance that could enable earlier diagnoses. For individuals at risk of developing dementia, early and differential diagnosis is key to streamlining patient management, benefiting patients and healthcare systems alike. While conventional neuropsychological assessments remain the gold standard for assessing cognitive and functional decline, these evaluations are lengthy (90-120 min), require a trained specialist, and are not free of bias and practice effects. In this context, digital biomarkers that enable the continuous and objective evaluation of multidimensional features assessing activities of daily living may have the potential to capture subtle changes in cognition and functional ability before the onset of cognitive decline. The Altoida Digital Biomarker Platform enables an objective evaluation of an individual's cognitive and functional impairment. The Altoida platform consists of two parts: 1) a participant-facing assessment (tablet-based) and 2) a site-facing analytics and reporting web portal. The assessment evaluates cognitive and functional skills based on a series of motor and augmented reality (AR) tasks that mirror the engagement of the brain during activities of daily living (Figure 1). These activities include tapping and tracing shapes, as well as placing and finding virtual objects while faced with a distractor task. The assessment takes an average of 10 minutes (average cognitively normal) to 18 minutes (average MCI) to complete. The dashboard provides real-time analytics and integration of study data into clinical workflows. The platform is currently intended for investigational use only. It has not received FDA pre-marketing clearance or approval. The platform evaluates multi-modal features, including micro-movements, speed, reaction times, and navigation trajectories, which are used to train specific machine-learning models, termed Digital Neuro Signature (DNS). Using machine learning, the digital biomarkers extracted by the Altoida assessment can be used to measure a patient's cognitive performance and to identify distinct clinical outcomes, such as MCI and MCI with likelihood of amyloid pathology in an ecological manner. The assessment also generates scores of specific brain domains of cognition defined by the DSM-V, such as learning and memory, executive function, complex attention, and perceptual-motor coordination derived from specific digital features scored with normative models (age and sex-adjusted). These are derived from specific digital features scored with normative models. In previous studies, Altoida's digital biomarkers were found to be useful in detecting early cognitive decline and also in predicting progression to dementia. In recent years, the Altoida assessment has been used across several global research studies, confirming the ease of use, non-invasiveness, potential to identify cognitive impairment as well as correlations with neuropsychological assessments. Early clinical recognition of Alzheimer's disease (AD) is critical. There is currently no software-based tool approved by regulatory authorities to adjunctively diagnose individuals with MCI and amyloid positivity, which is a population with a greater likelihood of progressing to full AD dementia. Early and differential diagnosis could create opportunities for participation in clinical trials of disease-modifying therapies and assist drug developers with accelerating the enrollment of the right patients for the right therapies.