Multicentre randomised controlled trial examining the cost-effectiveness of contrast-enhanced high field magnetic resonance imaging in women with primary breast cancer scheduled for wide local excision (COMICE)

L. W. Turnbull, S. R. Brown, C. Olivier, I. Harvey, J. Brown, P. Drew, A. Hanby, A. Manca, V. Napp, M. Sculpher, L. G. Walker, S. Walker, COMICE Trial Grp

Research output: Contribution to journalArticlepeer-review


Objectives: To determine whether the addition of magnetic resonance imaging (MRI) to current patient evaluation by triple assessment would aid tumour localisation within the breast and thus reduce the reoperation rate in women with primary breast tumours who are scheduled for wide local excision (WLE), and to assess whether the addition of MRI would be cost-effective for the UK NHS. Design: A multicentre, randomised controlled, open, parallel group trial with equal randomisation. The main design was supplemented with a qualitative study to assess patients' experiences of the treatment process and care pathway, and involved the development of a non-scheduled standardised interview (NSSI). Setting: The study took place at 45 hospitals throughout the UK. Participants: Women aged 18 years or over with biopsy-proven primary breast cancer who had undergone triple assessment, were scheduled for WILE, and were capable of providing written informed consent. Interventions: Patients were randomised to receive MRI or no MRI. Randomisation was performed using minimisation, incorporating a random element. All MRI was performed at 1.5T or 1.0T with a dedicated bilateral breast coil. Main outcome measures: The primary end point of the trial was the reoperation rate. Secondary outcome measures included discrepancies between imaging and histopathology, and the effectiveness of using both procedures; change in clinical management after using MRI; the clinical significance of MRI-only-cletected lesions; the rate of interventions; the ipsilateral tumour recurrence rate; patient quality of life (QoL); and cost-effectiveness. Results: From a total of 1623 patients, 816 were randomised to MRI and 807 to no MRI. No differences in reoperation rates were found between the two groups of patients [MRI patients 18.75%, no MRI 19.33%, difference 0.58%,95% confidence interval (CI) -3.24 to 4.40].Therefore, the addition of MRI to conventional triple assessment was not found to be statistically significantly associated with a reduced reoperation rate (odds ratio = 0.96,95% Cl 0.75-1.24, p = 0.769 1). The best agreement between all imaging modalities and histopathology with regard to tumour size and extent of disease was found in patients over 50 years old with ductal tumours NST and who were node negative. In the imaging arm, mastectomy was found to be pathologically avoidable for 16 (27.6%) out of 58 patients who underwent the procedure. There were no significant differences between the groups regarding the proportion of patients receiving chemotherapy, radiotherapy or additional adjuvant therapies, as well as for local recurrence-free interval rates and QoL. An acceptable NSSI was developed for use in this population of patients. Economic analysis found no difference in outcomes between the two trial arms. Conclusions: The addition of MRI to triple assessment did not result in a reduction in operation rates, and the use of MRI would thus consume extra resource with few or no benefits in terms of cost-effectiveness or HRQoL. However, MRI showed potential to improve tumour localisation, and preoperative biopsy of MRI-only-detected lesions is likely to minimise the incidence of inappropriate mastectomy.

Original languageEnglish
Pages (from-to)1-182
Number of pages182
JournalHealth technology assessment
Issue number1
Publication statusPublished - Jan 2010


  • MRI

Cite this