History of Surgical Clinical Trials and Cooperative Group Trials

NSAPB B-04

The National Surgical Adjuvant Breast Project (NSABP) B-04 trial began in August 1971. In this trial, 1665 women with breast cancer with disease confined to the breast and axilla, with clinically mobile disease, were enrolled and stratified into clinically node-negative and node-positive groups. Clinically node-negative patients (n = 1079) were randomized to radical mastectomy (n = 362) versus total mastectomy with regional radiation (n = 352) versus total mastectomy without radiation (n = 365). Clinically node-positive patients (n = 596) were randomized to radical mastectomy (n = 292) or total mastectomy with regional radiation (n = 294). No patients received adjuvant systemic therapy, and this trial preceded the identification and reporting of receptor subtypes.

The primary end point was disease-free survival (DFS), and there was no difference observed according to treatment groups in either clinical node-negative or node-positive patients. Similarly, there was no difference in overall survival according to treatment in node-positive patients or in node-negative patients. DFS was significantly reduced in patients with clinically node-positive disease compared to clinically node-negative disease, including both locoregional and distant recurrence. There was no difference in locoregional or distant recurrence according to treatment. In both node-negative and node-positive patients the incidence of locoregional recurrence plateaued by 3 years, while distant recurrence continued to occur out to 10 years. In patients assigned to radical mastectomy, there was a 40% incidence of pathologically positive lymph nodes in patients with clinically node-negative disease. In these clinically node-negative patients, the incidence of the first recurrence in the axilla was 4% in the radical mastectomy group, 6% in the total mastectomy group, and 4% in the total mastectomy plus radiation group.

This landmark trial demonstrated that there is no role for the morbid Halsted radical mastectomy, as it did not improve recurrence or survival over less invasive management. Postmastectomy radiation did not statistically reduce local, regional, or distant recurrence in this trial. Additionally, treatment of the axilla (nodal dissection or radiation) did not reduce local recurrence or distant recurrence or have any impact on survival in clinically node-negative patients despite a 40% rate of microscopic nodal positivity in patients randomized to axillary dissection. In clinically node-positive patients, axillary lymph node dissection (ALND) and axillary radiation were also equivalent for local, regional, and distant recurrence and survival. This trial revolutionized breast cancer surgery by putting to rest the need for radical mastectomy and laid essential groundwork for future trials on deescalation of surgery in the breast and in the axilla.

NSABP B-06

The NSABP B-06 trial was designed to address the oncologic safety of lumpectomy compared to mastectomy and determine the role of radiation after breast-conserving surgery. In this trial, 1851 women with primary breast cancer <4 cm, with or without clinically positive lymph nodes, were randomized to lumpectomy (n = 634), lumpectomy + radiation therapy (RT) (n = 628), or total mastectomy (n = 589). All patients underwent ALND, and patients with node-positive disease were treated with adjuvant melphalan and fluorouracil.

There was no difference in DFS, distant DFS, or overall survival between the three treatment groups. Looking specifically at the incidence of ipsilateral breast recurrence, the addition of radiation to lumpectomy resulted in a significant reduction in events compared to lumpectomy alone. At 20 years, lumpectomy alone resulted in a cumulative incidence of breast recurrence of 39% compared to 14% in patients who had a lumpectomy followed by radiation. The benefit of radiation was larger in patients with node-positive disease who were at higher risk for recurrence, 44% ipsilateral breast recurrence without radiation, compared to 9% with radiation at 20 years. This trial demonstrated that the addition of whole-breast radiation to lumpectomy improves breast recurrence, but does not impact disease-free, distant disease-free, or overall survival. Additionally, this trial showed that lumpectomy with radiation is equivalent to mastectomy for local recurrence, distant recurrence, and survival.

NSABP B-14, B-21

The recognition that most breast cancers are driven by upregulation of the estrogen receptor (ER) and estrogen-dependent signaling led to the development of antiestrogen therapeutics. The NSABP B-14 trial investigated whether adjuvant tamoxifen improved recurrence and survival compared to placebo in patients with breast cancer. A total of 2644 patients with clinically lymph node-negative ER-positive (ER+) breast cancer were enrolled. Patients were initially treated with modified radical mastectomy, however, after the publication of NSABP B-06 were permitted to have a lumpectomy with ALND and adjuvant radiation. At 4 years, tamoxifen increased DFS from 77% in the placebo group to 83% in tamoxifen-treated patients. There was no benefit observed in overall survival, which at 4 years was 92% in the placebo group and 93% in patients receiving tamoxifen. This trial demonstrated, for the first time, that targeted therapeutics inhibiting estrogen signaling improve oncologic outcomes in patients with ER+ breast cancer.

The NSABP B-21 trial showed that, in patients with small ER+ breast cancers less than 1 cm, the addition of radiation to tamoxifen reduced in-breast recurrence with no effect on distant recurrence and survival, much like B-06 demonstrated a local recurrence benefit without survival benefit. Due to the long course of recurrence and mortality for ER+ breast cancer, it was hypothesized that older patients may not derive the benefit from all treatment modalities.

Milan I

The Milan I trial randomized over 700 pT1 breast cancer patients into either radical mastectomy or quadrantectomy with axillary dissection and breast radiotherapy. The first results were published in 1981 and showed no difference in DFS and OS. A 20-year follow-up reported a crude cumulative incidence of local recurrence in the breast conservation and mastectomy groups of 8.8% and 2.3%, respectively ( P < 0.001). As in the initial report, there was no statistically significant difference in contralateral breast cancer events, overall survival, or breast cancer–specific survival.

CALGB 9343

The CALGB 9343 trial asked if the subgroup of older women treated with adjuvant tamoxifen could be safely omitted from postoperative radiation. Women aged 70 years or older with clinical stage I ER+ invasive carcinoma underwent lumpectomy with ALND and were randomized to adjuvant tamoxifen alone (n = 319) or tamoxifen with radiation (n = 317). There was no difference between the two groups in distant metastasis-free survival, which was over 90% in both groups at 10 years. A difference in locoregional recurrence-free survival was observed, which was 90% at 10 years in the tamoxifen alone group and 98% in patients who received both tamoxifen and radiation. Most of these patients were treated with repeat breast-conserving surgery, and so no difference in mastectomy-free survival was shown between patients who received radiation and those who did not. Finally, there was no difference in overall survival according to the receipt of radiation, with 66% of patients alive at 10 years in the tamoxifen arm and 67% alive in the tamoxifen + radiation arm. A noninferiority calculation did not demonstrate the advantage of radiation for the prevention of mastectomy or breast cancer–specific survival. The authors report that only 6% of the deaths in the trial were due to breast cancer and did an analysis comparing survival in the trial to age-matched women in the general population without cancer, which was equivalent. This trial demonstrated that, in older women with small ER+ breast cancers, the cancer itself is not a major driver of overall mortality, and radiation can be safely omitted in properly selected patients without compromising survival. However, it is important to note that radiation did reduce in-breast recurrence, and all patients underwent ALND, so the applicability of these findings in the absence of nodal clearance has not been shown, and nodal staging should not be omitted in these patients.

Prime II

A similar trial in women aged 65 years or older (PRIME II) with small, ER+, node-negative cancers showed similar results. Patients were treated with an adjuvant aromatase inhibitor, and fewer local recurrences were seen compared to the CALGB 9343 trial. That trial also demonstrated no improvement in overall survival with radiation, further supporting the safety of this strategy in a patient population where the cancer is not a driver of mortality and showing that with more efficacious endocrine therapy a reduction in the magnitude of absolute benefit of radiation on local recurrence.

EBCTCG 1081

While reduction of in-breast recurrence with radiation after breast-conserving surgery is well established, individual trials failed to demonstrate a benefit for overall or disease-specific survival, perhaps due to a lack of power and length of follow-up necessary to show a later manifesting effect. The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) performed a meta-analysis of 10,801 patients from 17 randomized controlled trials of breast-conserving surgery with or without adjuvant radiation. Of note in this meta-analysis, all patients included underwent ALND.

Analyzing the entire population, the meta-analysis demonstrated, consistent with individual trial results, a reduction in any first recurrence in patients who had radiation compared to those who did not, as well as a cumulative incidence of any recurrence at 10 years of 35.0% without radiation and 19.3% for patients who did undergo radiation. Breast cancer–specific death was 25.2% at 15 years in patients who did not receive radiation and 21.4% in patients who did. Therefore a 15.7% absolute difference in recurrence was seen at 10 years and a 3.8% improvement in death, indicating that, for every four recurrences prevented at 10 years, one death was prevented at 15 years.

Radiation reduced the risk of recurrence by about 50%, and this relative treatment effect was constant across a number of subgroups analyzed. While the relative effect was constant, subgroup analysis demonstrated that the absolute differences in recurrence and in breast cancer–specific death were larger in node-positive patients than in node-negative patients, indicating that the downstream effects on survival are directly related to the absolute benefit of reduction of local recurrence, which is dependent on the risk of that outcome. Enriching the population at higher risk for local recurrence reduces the number needed to treat with radiation and improve the risk-benefit of adjuvant radiation after breast-conserving surgery.

NCIC Clinical Trials Group MA.20

Whole-breast radiation had been shown to reduce in-breast recurrence following breast-conserving surgery however, the effect of regional nodal irradiation (defined as including internal mammary, supraclavicular, and axillary lymph nodes) in addition to whole-breast radiation was not known. Patients with node-positive or high-risk node-negative (defined as primary tumor >5 cm or >2 cm with fewer than 10 axillary nodes removed, and at least one of the following: grade 3, ER-negative, or lymphovascular invasion) treated with breast-conserving surgery and adjuvant systemic therapy were enrolled in the MA.20 trial. All patients with a positive sentinel node underwent ALND. Patients were randomized to either whole-breast radiation (WBRT, n = 916) or whole breast radiation plus regional nodal radiation (WBRNRT, n = 916).

The primary outcome was overall survival, and at 10 years there was no significant difference between the WBRT (81.8%) and WBRNRT (82.8%) groups. The trial was positive for the prespecified secondary end point of DFS, which was 77% in the control group and 82% in WBRNRT patients. The effect on reduction of recurrence was dependent on the site of recurrence and was larger for isolated locoregional DFS, hazard ratio (HR) 0.59; however, a statistically significant improvement in distant DFS was also observed (HR 0.76, confidence interval [CI] 0.60–0.97).

This MA.20 trial demonstrated a benefit of regional nodal field radiation in patients with positive lymph nodes or otherwise at high risk of higher nodal burden. Importantly, this effect, including on local-regional disease, was observed in addition to axillary dissection and WBRT. These findings indicate a benefit of regional nodal field radiation, which treats both the axilla and supraclavicular/internal mammary nodes that are inaccessible to surgery. Taken in conjunction with the AMAROS trial, which showed equivalence for regional recurrence in patients with T1–2 N1 tumors of nodal field RT to axillary dissection, these results show that regional nodal radiation is the preferred strategy for regional control. Radiation reduces treatment morbidity and distant recurrence in patients without clinically involved lymph nodes while maintaining an equivalent rate of nodal recurrence.

Targit-A

While whole-breast and even regional nodal radiation was becoming the standard of care in breast cancer after breast conservation, many questioned the utility of radiating the whole breast considering that 90% of in-breast tumor recurrences occurred in the same quadrant as the tumor bed. Concurrently, the availability and safety of intraoperative local radiation were established in multiple tumor types. The TARGIT-A trial sought to compare targeted intraoperative radiotherapy with the conventional policy of whole breast external beam radiotherapy.

The study randomized over 2000 patients to either whole-breast irradiation (WBI) or intraoperative targeted radiation. Local recurrence at 4 years was 1.20% (95% CI 0.53–2.71) in the targeted intraoperative radiotherapy and 0.95% (0.39–2.31) in the external beam radiotherapy group (difference between groups 0.25%, −1.04 to 1.54; P = 0.41). Fourteen percent of patients did receive both intraoperative radiation and whole-breast radiation due to positive margins after lumpectomy. The intraoperative RT alone group had lower report radiation toxicity and overall good cosmesis, thus opening the door to not only partial breast irradiation but also intraoperative radiation.

Eliot

Given the increasing use of accelerated partial breast irradiation (APBI), intraoperative radiotherapy with electrons was trialed to determine if a single intraoperative dose could be an effective treatment for patients with early breast cancer. The ELIOT trial randomized participants to either electron intraoperative radiotherapy (ELIOT) or conventional WBI. The study included 1305 women with T1/2 (<2.5 cm) N0 breast cancer who were deemed good breast conservation candidates. Local recurrence at 12 years was 11% in the ELIOT group and 2% in the WBI group with a total of 54 additional ipsilateral breast tumor recurrences (IBTRs) in the ELIOT group (HR 4.62, 95% CI 2.68–7.95; P < 0.0001). Although the local recurrence rate was significantly higher in the ELIOT group, this did not translate into an overall survival benefit. The use of ELIOT was not shown to be equivalent to WBI even though it was within the prespecified equivalence margin. The total number of IBTR events in the WBI group was less than anticipated. An unplanned subset analysis did identify a group of patients within the ELIOT group with a very low IBTR (<1.3% at 10 years). This group included patients with small (<1 cm) luminal A tumors that had a Ki67 proliferative index of less than 14%. These results helped fashion the current guidelines for APBI use in breast cancer treatment and highlighted the importance of patient selection when considering this approach.

NSABP B-51/RTOG 1304

The NSABP B-51/Radiation Therapy Oncology Group (RTOG) is a phase III trial evaluating postmastectomy chest wall + nodal irradiation and postlumpectomy regional nodal radiation in patients who convert from node positive to node negative following neoadjuvant chemotherapy (NCT). The primary end point will determine whether the proposed treatment significantly reduce the rates of events for invasive breast cancer recurrence-free interval. A target accrual of 1636 participants was planned. This study is currently closed to accrual, and results have not yet been published.

Margin Guideline Meta-Analysis and Consensus for Invasive Cancer

NSABP B-06 demonstrated that breast conservation surgery followed by whole-breast radiation is equivalent to mastectomy. Other studies confirmed these results and breast conservation became a standard treatment option. Until recently, however, the margin width constituting a negative margin was the subject of great debate. In 2013 the Society of Surgical Oncology (SSO) and the American Society for Radiation Oncology (ASTRO) organized a multidisciplinary consensus panel of experts to perform a literature review and meta-analysis. Between 1965 and 2013, 33 studies meeting the defined eligibility criteria were included, consisting of 28,162 patients. After review, the panel concluded that a negative margin defined as “no ink on tumor” effectively minimized the risk of IBTR, and obtaining wider margins did not significantly further reduce this risk. This was found to be true regardless of invasive tumor type, biologic subtype, or young age. For patients with invasive breast cancer or mixed pathology (invasive plus ductal carcinoma in situ [DCIS]), the recommendation for a negative margin is “no ink on tumor.”