Introduction: Active surveillance is a strategy for managing low-risk, localised prostate cancer, where men are observed with serial prostate-specific antigen assessments to identify signs of disease progression. Currently, there are no strategies to support active surveillance compliance nor are there interventions that can prevent or slow disease progression, ultimately delaying transition to active treatment before it is clinically required. Recently, we proposed that exercise may have a therapeutic potential in delaying the need for active treatment in men on active surveillance. Methods and analysis: A single-blinded, two arm, multicentre randomised controlled trial will be undertaken with 168 patients randomly allocated in a ratio of 1:1 to exercise or usual care. Exercise will consist of supervised resistance and aerobic exercise performed three times per week for the first 6 months in an exercise clinical setting, and during months 7–12, a progressive stepped down approach will be used with men transitioning to once a week supervised training. Thereafter, for months 13 to 36, the men will self-manage their exercise programme. The primary endpoint will be the time until the patients begin active therapy. Secondary endpoints include disease progression (prostate specific antigen), body composition and muscle density, quality of life, distress and anxiety and an economic analysis will be performed. Measurements will be undertaken at 6 and 12 months (postintervention) and at 24 and 36 months follow-up. The primary outcome (time to initiation of curative therapy) will be analysed using Cox proportional hazards regression. Outcomes measured repeatedly will be analysed using mixed effects models to examine between-group differences. Data will be analysed using an intention-to-treat approach. Ethics and dissemination: Outcomes from the study will be published in peer-reviewed academic journals and presented in scientific, consumer and clinical meetings.