Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomized controlled trial

Authors: Joan Houghton, et al.
Affiliation: UKCCCR DCIS Working Party / DCIS trialists in the UK, Australia, and New Zealand
Source:Lancet. 2003 Jul 12;362(9378):95-102

Introduction

The establishment of breast screening programs in the UK, Australia, and New Zealand resulted in an increase in the diagnosis of ductal carcinoma in situ (DCIS). At that time, randomized trials had demonstrated that breast conservation therapy (BCT) for invasive cancer was a safe alternative to mastectomy. However, it was still unclear whether breast-conserving alternatives were sufficient for DCIS. This study (contemporary with NSABP B-17, EORTC 10853, and NSABP B-24) was designed to investigate the role of breast irradiation as well as adjuvant tamoxifen for locally excised DCIS.

Methods

• Between 1990 and 1998, patients with DCIS and microinvasive carcinoma detected by screening and deemed suitable for BCT were offered participation
• The study excluded patients with lobular carcinoma in situ, uncertain pathological margins of resection, Paget's disease, and reduced life expectancy
• Complete excision, defined by free margins on histological examination and radiology of the surgical specimen, was required
• Re-excision was allowed to establish clear margins
• Randomization, using a 2x2 factorial design, was independent for each of the two treatments (radiotherapy and tamoxifen)
• Patients who either expressed a preference within one of the 2-way randomizations or were not equally suitable for the options within one of the randomizations were only randomized and analyzed on the other 2-way pathway
• Radiotherapy consisted of supervoltage radiation to the whole breast through tangential fields to a dose of 50 Gy (at isocenter) in 25 fractions over 5 weeks
• Tamoxifen was prescribed at a dose of 20 mg daily for 5 years
• Patients were followed up at least once a year
• Analysis was confined to patients who were randomly allocated to each main treatment comparison
• All analyses were on an intention-to-treat basis and stratified according with whether the alternate treatment had been given or not, and whether this was by choice or because of random treatment
• All p values are two-sided

Results

• 1694 patients were analyzed (1701 entered minus 7 excluded):
• • 912 participated in the 2x2 randomization
  • 782 refused randomization in one of the 2-way pathways:
  • • 664 chose whether or not to have radiation, with 603 choosing no radiation. They only participated in the tamoxifen randomization.
    • 118 chose whether or not to have tamoxifen, with 29 choosing no tamoxifen. They only participated in the radiation randomization.
  • Median follow-up was 52.6 months
  • 3% had micro-invasion and 8% were on HRT (well balanced among arms)
  • Recurrence in these subgroups did not differ from that in the overall group
  • 10% were younger than 50 years at randomization
  • 16% developed new breast events (9% in situ and 6% invasive)
  • • 85% of these were ipsilateral
    • The event rate per 100 woman-years was highest in the group who received no adjuvant treatment (4.77), and second highest in the group who received tamoxifen only (3.86)
    • Women who received radiotherapy had the lowest event rates
• 1576 were randomized/analyzed for the tamoxifen comparison: 782 controls and 794 in the treated group
• • New breast events were noted in 14% of patients randomized to tamoxifen and 18% in those randomized to no tamoxifen (not significant)
  • • Invasive events: 7% vs. 6% (not significant)
    • In situ events: 7% vs. 11% (less with tamoxifen, p=0?03)
    • There were too few contralateral events for analysis
• 1030 were randomized/analyzed for the radiotherapy comparison: 508 patients in the control group and 522 in the treated group
• • New breast events were noted in 7% of patients randomized to radiotherapy and 16% in those randomized to no radiotherapy (highly significant reduction with radiotherapy)
  • The rate of contralateral events was not affected by radiotherapy
  • Ipsilateral invasive events: 3% vs. 6% (less with radiotherapy, p=0?01)
  • Ipsilateral in situ events: 3% vs. 7% (less with radiotherapy, p=0?0004)
• 2% of patients developed a new non-breast cancer
• • 8 were gynecological tumors; 7 among those who had tamoxifen and 1 among those who did not have tamoxifen (not significant)
• 45 patients died during follow-up:
• • In 23 the cause of death was breast cancer
  • There were too few deaths for comparison between treatment arms
• With respect to the occurrence of new events, the effect of tamoxifen was not altered by radiotherapy and vice-versa
• Because only 160 (9.5%) patients were younger than 50 years at randomization, meaningful analysis of the effect of radiotherapy and tamoxifen stratified by age (=50 years vs. >50 years) was not possible

Discussion

• Ipsilateral invasive disease was not reduced by tamoxifen.
• The only significant reduction associated with tamoxifen in the absence of radiotherapy was in the rate of ipsilateral DCIS.
• For patients who also received radiotherapy, the reduction in these events was not significant.
• These results differ from the only other trial to investigate this drug in completely excised DCIS:
• • In the NSABP B-24 trial, the number of new breast events was significantly lower in the tamoxifen group than in the control group
  • • Significant reduction for invasive disease; non-significant trend for non-invasive disease
  • The number of new breast events was similar in both trials (256 for this trial vs 214 in the B-24 trial), indicating that the discrepancies are not likely to be caused by differences in power.
  • Only 33% in the tamoxifen comparison received radiotherapy compared with 100% in the B-24 trial
  • • However, sub-group analyses showed the same pattern of results for tamoxifen both in the absence and in the presence of radiation.
  • The differences might be partly explained by the fact that in the B-24 trial 33.5% of patients were younger than 50 years, compared with only 9.5% in this trial.
  • In the B-24 trial, the reduction in ipsilateral event rate with tamoxifen was more obvious for patients under the age of 50 years, compared with those older than 50 years.
  • Similarly, there was a trend in this study in support of greater risk reduction with tamoxifen for women less than 50 years than for those older than 50 years.
• This study confirms the reduction in ipsilateral events with the use of radiotherapy demonstrated by the NSABP B-17 trial and the EORTC trial.
• • However, the results of the present study differ from those trials in that there was a greater risk reduction for in situ than for invasive disease.
  • Also, the absolute recurrence rate was lower in this trial than in either of the other two ductal carcinoma in-situ trials.
  • The fact that the UK trial had older patients than did NSABP B-17, and most of these patients had disease detected through a screening program might explain these differences.
• The effects of radiotherapy in this trial were much the same whether or not patients received concurrent tamoxifen and irrespective of their age.
• Radiotherapy had no effect on contralateral incidence.

Interpretation

• The results of this trial confirm the previously reported benefit of radiotherapy after complete local excision for DCIS.
• There are conflicting data for the use of tamoxifen in women with DCIS.
• The following issues remain unclear:
• • Is there a subgroup at low risk of recurrence in which radiotherapy can be safely avoided?
  • Is the use of adjuvant tamoxifen for DCIS warranted outside the setting of randomized trials? What about in those > 50 years?
  • Are the results with tamoxifen applicable to newer endocrine agents such as the aromatase inhibitors?
  • Is hormone receptor status needed to assess the true role of endocrine agents in DCIS?
• Longer follow-up of all patients randomized into these trials is needed.