Effect of tamoxifen and radiotherapy in women with locally excised ductal carcinoma in situ: long-term results from the UK/ANZ DCIS trial

Abstract

Background: Initial results of the UK/ANZ DCIS (UK, Australia, and New Zealand ductal carcinoma in situ) trial suggested that radiotherapy reduced new breast events of ipsilateral invasive and ductal carcinoma in situ (DCIS) compared with no radiotherapy, but no significant effects were noted with tamoxifen. Here, we report long-term results of this trial.

Methods: Women with completely locally excised DCIS were recruited into a randomised 2×2 factorial trial of radiotherapy, tamoxifen, or both. Randomisation was independently done for each of the two treatments (radiotherapy and tamoxifen), stratified by screening assessment centre, and blocked in groups of four. The recommended dose for radiation was 50 Gy in 25 fractions over 5 weeks (2 Gy per day on weekdays), and tamoxifen was prescribed at a dose of 20 mg daily for 5 years. Elective decision to withhold or provide one of the treatments was permitted. The endpoints of primary interest were invasive ipsilateral new breast events for the radiotherapy comparison and any new breast event, including contralateral disease and DCIS, for tamoxifen. Analysis of each of the two treatment comparisons was restricted to patients who were randomly assigned to that treatment. Analyses were by intention to treat. All trial drugs have been completed and this study is in long-term follow-up. This study is registered, number ISRCTN99513870.

Findings: Between May, 1990, and August, 1998, 1701 women were randomly assigned to radiotherapy and tamoxifen, radiotherapy alone, tamoxifen alone, or to no adjuvant treatment. Seven patients had protocol violations and thus 1694 patients were available for analysis. After a median follow-up of 12·7 years (IQR 10·9-14·7), 376 (163 invasive [122 ipsilateral vs 39 contralateral], 197 DCIS [174 ipsilateral vs 17 contralateral], and 16 of unknown invasiveness or laterality) breast cancers were diagnosed. Radiotherapy reduced the incidence of all new breast events (hazard ratio [HR] 0·41, 95% CI 0·30-0·56; p<0·0001), reducing the incidence of ipsilateral invasive disease (0·32, 0·19-0·56; p<0·0001) as well as ipsilateral DCIS (0·38, 0·22-0·63; p<0·0001), but having no effect on contralateral breast cancer (0·84, 0·45-1·58; p=0·6). Tamoxifen reduced the incidence of all new breast events (HR 0·71, 95% CI 0·58-0·88; p=0·002), reducing recurrent ipsilateral DCIS (0·70, 0·51-0·86; p=0·03) and contralateral tumours (0·44, 0·25-0·77; p=0·005), but having no effect on ipsilateral invasive disease (0·95, 0·66-1·38; p=0·8). No data on adverse events except cause of death were collected for this trial.

Interpretation: This updated analysis confirms the long-term beneficial effect of radiotherapy and reports a benefit for tamoxifen in reducing local and contralateral new breast events for women with DCIS treated by complete local excision.

Funding: Cancer Research UK and the Australian National Health and Medical Research Council.

Figure 1

Trial profile T=tamoxifen. R=radiotherapy. C_T=control group for tamoxifen. C_R=control group for radiotherapy.

Figure 2

Kaplan-Meier curve for cumulative incidence and annual hazard plot of all breast events in the tamoxifen comparison

Figure 3

Forest plot for new breast events in the tamoxifen comparison stratified by whether or not patients received radiotherapy HR=hazard ratio. DCIS=ductal carcinoma in situ.

Figure 4

Kaplan-Meier curve for cumulative incidence and annual hazard plot of all breast events in the radiotherapy comparison

Figure 5

Forest plot for new breast events in the radiotherapy comparison stratified by whether or not patients received tamoxifen HR=hazard ratio. DCIS=ductal carcinoma in situ.