Wide local excision followed by external beam radiation therapy (EBRT) to the whole breast has become the standard of care for most patients with localised 'early' breast cancer in the UK, Europe, and the USA. Local relapse rates are low, and overall survival figures have improved during the past decade, with the advent of more effective systemic endocrine- and chemo-therapy. A policy of EBRT for every patient undergoing breast conserving surgery (BCS) is however associated with a number of practical difficulties, acute radiation side effects and longer term toxicity, all of which detract from the obvious benefits of EBRT. In addition, with a disease as common as early breast cancer and a treatment programme typically requiring sophisticated radiation planning and many fractions of treatment, the policy of BCS plus EBRT has enormous resource implications within departments of oncology, greatly contributing to lengthy pre-treatment delays. For all these reasons, we and others have developed an increasing interest in techniques of partial breast irradiation, with an emphasis in our own Department on the emerging technique of intra-operative radiotherapy (IORT), which we initially employed as a boost to the tumour bed for use in conjunction with EBRT to the whole breast. To test the possibility of replacing the whole of the EBRT 3-6 week programme by a single application of IORT at the time of surgery, we and others have commenced a large scale prospectively randomised clinical trail in selected patients. Nine international centres are currently participating, and 350 patients have now been randomised to receive either IORT as part of the initial surgical excision or conventional EBRT with a pragmatic dose policy according to the preference of the contributing centre. The majority of patients undergoing IORT receive this at the time of initial surgery but it is also permissible within the trial programme to randomise suitable patients after the excised specimen has been histologically examined, thus avoiding any unsuitable patients - for example, those with a lobular carcinoma. These patients will be stratified and assessed separately from the 'pre-pathology' group, whose surgery and IORT is completed within a single session; if the latter patients are found to have unfavourable histology we have the facility, within the trial, to add EBRT. The trial is ongoing and our early experience has been encouraging. We have also recently assessed the long term local failure rate in patients offered IORT as a tumour bed boost, in conjunction with conventional EBRT. This methodology will also be the subject of a future randomised clinical trial.