BACKGROUND: To date, beneficial effects of multimodal exercise programmes on Parkinson's disease (PD) have focused on motor symptoms and little attention has been paid to the potential effects of such programmes on the non-motor symptoms of PD, which are now universally known as one of the key drivers of quality of life and a key unmet need. We aim to explore clinical effectiveness of a ballet-based dance programme in addressing non-motor and motor symptoms of Parkinson's disease across all stages of progression. METHODS: A randomised, single-blind, controlled trial of 160 people with Parkinson's across all motor stages (Participants will be stratified into three groups of motor advancement: Hoehn and Yahr (HY) stages I and II being Mild Group, HY Stage III being Moderate Group and HY Stages IV and V being Severe Group) will be randomly allocated to either an intervention or a control group using an independent randomisation body. The primary outcome is an improvement in non-motor symptoms as measured by the Movement Disorders Society Non-Motor Scale (MDS-NMS). The intervention protocol consists of 12 one-weekly dance sessions led by English National Ballet. Each session is followed by a 'tea and biscuit' social time. Control group follows standard clinical pathway and joins the 'tea and biscuit' to control for any positive effects of social interactions. All participants are assessed at baseline, immediately after completion of the intervention and 3-6 months later to explore any potential longitudinal effects. DISCUSSION: To our knowledge, no adequately powered study has explored the effects of a dance-based intervention on non-motor symptoms of Parkinson's disease, assessing these on both holistic and granular levels. We also aim to stratify participants in accordance with their motor state as assessed by. HY staging to explore specific effects on the symptoms at the initial, moderate and complex stages of the disease. If successful, this trial provides first evidence on clinical effectiveness of a ballet-based dance intervention for symptoms of Parkinson's disease, assessed in a robust, rigorous manner. TRIAL REGISTRATION: NCT04719468.