Advancing methods for pharmacological fMRI to improve discovery and development of treatments for mental health

This project aims to advance pharmacological functional magnetic resonance imaging (pharmaco-fMRI) as a robust and reliable tool for accelerating the clinical development of novel mental health treatments and to develop pharmaco-fMRI-based precision approaches for predicting individual responses to drugs. Treatment effectiveness in psychiatry is often unpredictable due to varied responses and subjective diagnoses, contributing to high drug trial failure rates. Funders and regulators now promote objective measures, such as functional neuroimaging, to improve precision drug development. Pharmaco-fMRI is a validated method for quantifying brain function changes in response to drugs, but small sample sizes have limited its impact. This fellowship will aggregate existing datasets, already pledged by over 15 leading PIs, harmonise the largest pharmaco-fMRI dataset to date, and establish a consortium to facilitate sharing with researchers. Using this dataset, this fellowship will develop novel methods to characterise subject-specific drug effects free from confounds like scanner and acquisition sequence, enabling drug response prediction from resting-state brain function. This will be done using a new foundation model of brain function, BrainLM. The project will also validate advanced methods like Receptor Enriched Analysis of Connectivity by Targets (REACT), which provides molecular insights from fMRI scans without the need for costly PET scans.