CURATE.AI COR-Tx Trial for Post Brain Radiotherapy Patients

Overview

Cognitive deficit is common in patients who have undergone whole brain or partial brain radiotherapy. To counteract intellectual deterioration, the conventional strategies includes drug- based treatments such as donezepil and memantine, which have shown to only provide marginal improvement and, cognitive training regimens, both of which are usually administered at fixed dose/intensities often leading to sub-optimal responses. This study aims to address this clinically relevant problem by harnessing the CURATE.AI platform to identify N-of-1cognitive training profiles the can enhance learning trajectories through individualised calibration and training regimens. CURATE.AI is a phenotypic personalised medicine (PPM) platform that correlates a patient's phenotypic response (cognitive performance) to a certain input (training intensity) based exclusively on the patient's data. This PPM platform is independent of biological system or interventional agent and can be applied to any disorder treatment where dosing/intensity could be better personalised. CURATE.AI is expected to optimise/personalise cognitive training in post-brain radiotherapy patients by dynamically modulating the intensity of a digital cognitive test battery that measures executive processing, multitasking and perceptual learning tasks. In addition, this clinical feasibility trial aims to assess this cognitive test battery as a potential analogous or complementary diagnostic tool as compared to traditional cognitive evaluations performed by a clinician.

Brain radiotherapy causes downstream cognitive deficits. Drug-based cognitive decline treatments show little improvement and side effects may reduce patient compliance. Regimens are usually administered at a fixed dose that doesn't incorporate high patient variability, leading to sub-optimal responses. Effective cognitive training can improve cognitive performance. Artificial intelligence platforms show great potential for training personalisation. The CURATE.AI platform can be used to identify N-of-1 (single subject) training profiles that can be used to optimise learning trajectories through individualised calibration and training regimens, potentially leading to improved outcomes compared to standard static or adaptive training strategies. CURATE.AI uses only a subject's own performance data to identify the intensity needed for his/her best output. As the subject evolves during the course of intervention, the training intensities are dynamically modulated to maintain performance within a given range. Here the investigators propose to test the feasibility of CURATE.AI, with a digital cognitive test battery as the interface, as an adaptive training platform for cognitive training addressed to improve brain cancer radiotherapy patients' cognitive performance. The acceptability, implementation and limited efficacy of the digital intervention (DI) will be explored. In addition, the investigators propose to test the feasibility of the digital cognitive test battery potential as a digital diagnostic (DD) tool as compared to traditional cognitive evaluations performed by a clinician. User experience and usability will also be explored.