Evidence synthesis to guide systematic candidate drug prioritisation for repurposing in clinical trials of neurodegenerative diseases.

Abstract

Together, neurodegenerative diseases are a leading and increasingly common cause of disability in our aging population. Despite extensive research and investment, and numerous drugs with promising laboratory data, treatment options remain limited and cure elusive. Drug development has been characterised by high costs and failure rates, and lengthy timelines. With innovations in clinical trial designs promising more efficient clinical evaluation of pipelines of candidate drugs, it is also crucial to rethink how candidate drugs are selected for trials. With the use of derisked compounds, drug repurposing potentially provides time and cost efficiencies in drug development. To inform systematic evidence-based approaches in prioritising candidate drugs for repurposing in neurodegenerative diseases, we identify, evaluate, and report evidence from: (i) published literature via machine learning assisted systematic reviews of clinical studies of Alzheimer’s disease (AD), frontotemporal dementia, Huntington’s disease, motor neuron disease (MND), multiple sclerosis (MS), Parkinson’s disease; animal studies; and in vitro studies, including using Systematic Online Living Summaries (SOLES); (ii) in-house experimental high throughput drug screening; (iii) pathway and network analysis; (iv) mining drug and trial databases. We report living evidence summaries with interactive web applications and drug CVs to inform expert panel decisions. We applied earlier versions of this framework to drug selection for progressive MS in MS-SMART (ISRCTN28440672) and for MND in the first three arms of MND-SMART (NCT04302870). We plan to continue to develop and improve the framework and inform ongoing drug selection in multi-arm multi-stage trials in various conditions including MND, AD, progressive MS and cerebral small vessel disease. Incorporation of additional data domains and more efficient artificial intelligence tools will enable more efficient, broad-scope, scalable evidence synthesis.