Radiotherapy Versus Low-Dose Tamoxifen Following Breast Conserving Surgery for Low-Risk Breast Ductal Carcinoma in Situ

[Background] Radiotherapy (RT) following breast conserving surgery (BCS) is commonly used in ductal carcinoma in situ (DCIS) of breast to decrease local recurrence. Previous retrospective studies suggested that a substantial proportional of low-risk DCIS patients who underwent BCS alone will not develop a subsequent invasive breast cancer over time.

To further identify a population of patients with low-risk DCIS in whom adjuvant RT could be safely omitted, two prospective trials and one randomized trial have been published the ipsilateral breast tumor recurrence (IBTR) rate in patients with low-risk breast DCIS, including smaller size, larger margin width, and lower grade. The 5-year IBTR rate in Dana Farber/Harvard Cancer Institute (low to intermediate grade disease and a margin width of 10 mm) was 12% for breast DCIS patients who underwent BCS alone. The Eastern Cooperative Oncology Group (ECOG) E5194 reported a 12-year IBTR rate of 14.4% (7-year IBTR rate, 10.5%) for patients with low-risk DCIS (cohort 1 criteria: mammographically detected low- or intermediate-grade DCIS, measuring less than 2.5 cm with margins ≥ 3 mm) who underwent BCS alone. The risk for DCIS and invasive cancer increased steadily over time, without any plateau.

Another randomized trial, Radiation Therapy Oncology Group (RTOG) 9804, showed the IBTR rates from that at 5 years (3.5%) to that at 7 years (6.7%) in post-BCS patients with low-risk breast DCIS (similar criteria of ECOG E5194 trial, cohort 1) who did not receive RT. In contrast, the 7-year IBTR rate was low (0.9%) for patients who underwent BCS and received RT in RTOG 9804 trial. Our retrospective study showed that the 7-year IBTR rate was 5.6% in breast DCIS patients who had criteria of cohort 1 of ECOG E5194 trial and underwent BCS alone. Taken together, two prospective clinical trials and our retrospective study disclosed that the 7-year IBTR for BCS alone group (no RT) ranged from 5.6% to 10.5%, even if patients whose clinicopathologic features met the criteria of ECOG E5194 cohort 1. These results suggest that RT following BCS is indicated for these patients who have low-risk breast DCIS.

Another randomized trial, UK/ANZ [UK, Australia, and New Zealand] DCIS trial], including high-risk and low-risk breast DCIS patients, demonstrated a significant benefit of tamoxifen (20 mg every day for 5 years) in terms of reducing the ipsilateral (Hazard ratio [HR], 0·77; 95% confidence interval [CI], 0·59-0·98; P = 0·04) and contralateral breast events (HR, 0·27; 95% CI, 0·12-0·59; P = 0·001) in the BCS alone group, and this benefit of tamoxifen was not observed in the BCS plus RT group for ipsilateral events (HR, 0·93; 95%, 0·50-1·75; P = 0·8). These findings suggest that even if patients with low-risk clinicopatholgical features of DCIS (relatively low-risk of IBTR), the addition of tamoxifen in patients who had received RT may not reduce IBTR than those receiving RT alone. However, the estrogen receptor (ER) status among the patients with DCIS enrolled in the aforementioned three prospective randomized trials (ECOG E5194 trial, RT9804 trial, and UK/ANZ trial) was initially unknown.

In a retrospective analysis of the relationship between ER status and response to tamoxifen in 732 patients (41%) who comprised the original NSABP B-24 population (76% positive for ER), the significant effect of tamoxifen in reducing ipsilateral and contralateral breast events was demonstrated in ER-positive DCIS but not in ER-negative tumors. Our retrospective study also showed that age < 40 years and negative ER status in tumors were closely associated with the higher IBTR rate. Among our patients with cohort 1 criteria of ECOG E519 study, the 7-year IBTR rate for ER-positive group and ER-negative group was 5.0% and 8.0%, respectively.

[Rationale] Although the results obtained from ECOG E5194 (cohort 1) and RTOG 9804 trial (the same enrolled clinicopathological features to cohort 1 of ECOG E5194 trial) demonstrated that the 7-year IBTR ranged from 5.6% to 10.5% for low-risk DCIS patients, the aforementioned two studies included a proportional patients who had young age and negative ER status tumor. Previous studies and our studies revealed that age < 40 years and ER-negative status in tumor were independent prognostic factor for recurrence of breast DCIS irrespective of tumor characteristics.

The UK/ANZ randomized trial, enrolling high-risk and low-risk clinicopathologic features of DCIS, demonstrated that a benefit of tamoxifen in terms of reducing the IBTR is observed in the BCS alone group but not found in the BCS plus RT group. In a recent published data of a randomized trial of comparing low-dose tamoxifen (5 mg QD) for 3 years with placebo in prevention of recurrence of women with hormone-positive DCIS or lobular carcinoma in situ, low-dose tamoxifen was demonstrated to significantly decrease local recurrence when compared with placebo arm. These findings indicate that tamoxifen at the dose of 5 mg/day can decrease the incidence of recurrence in women with operated hormone sensitive DCIS with a limited toxicity. However, the effect of the administration of low-dose tamoxifen is similar to the RT effect in terms of reducing IBTR for patients who had the criteria of ECOG E5194 cohort 1 and positive ER status remains unclear.

[Hypotheses] Based on the aforementioned results, we hypothesized that the administration of tamoxifen is not inferior than the prescription of RT in terms of reducing the IBTR for DCIS patients who had age more than 40 years, the pathological features meeting the ECOG E5194 cohort 1 criteria, and positive ER status in tumors.

To approve the hypothesis, we will design a randomized non-inferiority trial to assess whether the effect of administration of tamoxfien (5 mg per day) for 10 years following BCS is not inferior in terms of reducing IBTR when comparing RT following BCS for patients who had low-risk clinicopathologic features and positive-ER status of breast DCIS.

[Study Design] We will design a randomized non-inferiority trial to assess whether the effect of administration of tamoxfien (5 mg) for 10 years following BCS is not inferior in terms of reducing IBTR when comparing RT (in terms of 50 Gy in 25 fractions or 40.05 Gy in 15 fractions) following BCS for patients who had age more than and equal 40 years, low-risk clinicopathological features (ECOG E5194 cohort 1 criteria), and positive-ER status of breast DCIS.