Interfacing with the brain for therapy

The treatment of diseases of the brain remains a major challenge. This is partly because drugs ignore the moment-to-moment interactions between nerve cells that determine brain function and dysfunction. Electrical stimulation therapies, however, can intervene fast enough to modify these interactions. Our central hypothesis is that we can improve therapies based on electrical stimulation by titrating stimulation according to the state of neural circuits at any given instance. In this research programme we will focus on patients with two exemplar conditions, Parkinson’s disease and Essential Tremor, and develop feedback controlled forms of deep brain stimulation, and of non-invasive stimulation techniques where stimulation is delivered by electrodes on the skin to the surface of the brain or to nerves in the limbs. Using these approaches we seek to develop treatment approaches that serve to increase efficacy and lessen side-effects.

The failure rate for new drugs targeting important central nervous system diseases remains high and investment in drug discovery in this field continues to diminish. A key problem is that current therapeutic approaches tacitly assume a static equilibrium in brain state. In contrast, we seek to identify and correct those circuit interactions that, through their moment-to-moment dynamics, underpin disease symptoms related to aberrant movement. To this end, we will focus on proof-of-principle studies in patients with two exemplar conditions, Parkinson’s disease and Essential Tremor, and leverage electrical stimulation in its different forms. Common to the therapeutic interventions that we plan to develop, whether invasive or non-invasive, is that they increase selectivity by limiting stimulation according to feedback. Increased selectivity should translate in to improved efficacy and reduced side-effects, and these will be our ultimate measures of success. The feedback used to control stimulation may be neural, as in the case of the amplitude or phase of oscillations recorded in the EEG or in depth recordings, or peripheral, as in the case of the amplitude or phase of tremor oscillations recorded by accelerometers. Non-invasive stimulation, like transcranial alternating current or peripheral nerve stimulation, or invasive stimulation delivered through deep brain electrodes, is then controlled according to the amplitude or phase of the feedback. However, the most appropriate choice of feedback signal and control policy can vary according to changing context, such as sleep or gait, and we will therefore develop flexible closed-loop control procedures that change between key contexts. In parallel, we will investigate the functional role of oscillatory activities in the cortical-basal ganglia network so as to inform our targeted therapy development for brain diseases.