More Evidence That Locoregional Radiation Therapy Improves Survival: What Should We Do?

In this issue of the Journal, Ragaz et al. ( 1 ) report the long-term results (20 years of follow-up) of a randomized trial of locoregional radiation therapy compared with no further treatment after mastectomy among axillary lymph node–positive premenopausal patients with breast cancer treated with adjuvant intravenous chemotherapy (cyclophosphamide, methotrexate, and 5-fluorouracil). In this follow-up, radiation therapy reduced isolated locoregional recurrence, distant recurrence, deaths due to breast cancer, and overall mortality. The previous report of this study ( 2 ) demonstrated similar findings, but the reduction in mortality was not statistically significant. This update is important because it demonstrates that in patients with lymph node–positive breast cancer, survival is improved with locoregional radiation therapy (relative risk RR = 0.73, 95% confidence interval CI = 0.55 to 0.98). This update also reports that the improvement in survival appears to be the same for patients with one to three positive lymph nodes (RR = 0.76, 95% CI = 0.50 to 1.15) or for patients with four or more positive lymph nodes (RR = 0.70, 95% CI = 0.46 to 1.06). There was no suggestion that the treatment effect was less for patients at lower risk of locoregional recurrence, although the study was relatively underpowered for such comparisons. The trial also provides some long-term data on toxicity; an increased rate of lymphedema persisted, but the number of non–breast cancer deaths, in particular cardiac deaths, was not statistically significantly different in the two arms. However, these results on long-term toxicity should be viewed with caution because the overall rates of non–breast cancer deaths were low and because the relative increase in non–breast cancer deaths after radiation therapy tended to be similar to that observed in other trials ( 3 ). Such a proportional increase, if real, may be relevant for older women at higher risk for cardiac disease. The results of the British Columbia study and of a similar trial from the Danish Breast Cancer Cooperative Group (DBCG) ( 4 ) generated a considerable amount of controversy because previous postmastectomy radiation therapy trials demonstrated that radiation therapy decreased the risk of locoregional recurrence but failed to have an effect on overall survival ( 5 ). These studies and a companion study ( 6 ) by the DBCG group in postmenopausal patients demonstrated that locoregional radiation therapy after mastectomy in patients treated with adjuvant systemic therapy (chemotherapy or hormonal therapy) increased survival. At the time, the generalizability of these results was questioned ( 7 , 8 ). The locoregional recurrence rate in the control arms of the British Columbia and Danish trials was higher than expected for patients treated with standard surgical approaches and modern systemic therapy and was attributed to the use of less intensive chemotherapy and limited axillary lymph node dissection. Concerns about the potential for increased risk of late cardiac morbidity associated with locoregional radiation therapy were also raised ( 9 ). Despite these concerns, these results did lead to a paradigm shift in our thinking about breast cancer and supported the concept that aggressive locoregional treatment could prevent systemic failure and deaths from breast cancer. As a result, national guidelines were developed to recommend the use of locoregional radiation therapy for patients at high risk of locoregional recurrence (i.e., patients with primary tumors of more than 5 cm in diameter or with more than four positive axillary lymph nodes) after mastectomy ( 10 , 11 ). Substantial uncertainty, however, remained about the management of patients at lower risk for locoregional recurrence (i.e., those with fewer than three positive axillary lymph nodes). Consequently, randomized trials were initiated to evaluate the role of locoregional radiation therapy after mastectomy or breast-conserving therapy in such patients at moderate risk for locoregional recurrence ( 12 , 13 ). Since the first report from the British Columbia trial appeared, investigators from single centers or from multicenter collaborative groups have documented substantial locoregional recurrence rates for patients treated with anthracycline-based chemotherapy ( 14 , 15 ). A meta-analysis of all trials of locoregional radiation therapy after mastectomy in patients treated with systemic therapy has confirmed the findings of the original trials ( 16 ), and subgroup analyses have suggested that radiation therapy technique, timing of treatment, and type of chemotherapy may affect the efficacy of radiation therapy. An increased risk of cardiac toxicity has not been detected with more modern radiation therapy techniques but increased rates of lymphoedema and shoulder dysfunction have been reported ( 17 , 18 ). Other studies ( 19 , 20 ) have confirmed an increased risk of second malignancies associated with chest wall irradiation. Where are we now, 7 years after the first report of the British Columbia trial? Most clinicians continue to use locoregional radiation therapy after mastectomy in patients at high risk of recurrence, but its use is less widely accepted for patients at lower risks. For patients treated with breast-conserving surgery and whole breast irradiation, the role of additional regional radiation therapy should be clarified when the results of the following two trials are reported. The European Organisation for Research and Treatment of Cancer trial 22922-10925 ( 12 ) randomly assigned patients with medial tumors or with lymph node–positive breast cancer to receive medial supraclavicular and internal mammary lymph node irradiation or no lymph node irradiation. Accrual for this trial has been completed. The National Cancer Institute of Canada Clinical Trials Group MA-20 trial ( 13 ) is randomly assigning patients with lymph node–positive or high-risk lymph node–negative breast cancer either to supraclavicular, high axilla, and internal mammary lymph node irradiation plus breast irradiation or to breast irradiation alone. More than half of the targeted number of participants has been accrued for this trial, and accrual is continuing on a steady basis. What should we do for patients at moderate risk of locoregional recurrence (one to three positive axillary nodes and high-risk lymph-node-negative disease) after mastectomy? Unfortunately, high-quality evidence to address this issue will not be forthcoming. A large, randomized, controlled trial supported by the North American Breast Intergroup closed because of poor accrual, but a United Kingdom/European study that is being planned may provide such evidence (Kunkler I: personal communication). Some may argue that because fewer patients are receiving a mastectomy, this question is less relevant. However, up to 50% of patients in some parts of the United States continue to receive a mastectomy ( 21 ), and mastectomy remains an important treatment option for younger women who have a family history of breast cancer. In the absence of other evidence, we may be forced to rely on subgroup analyses, such as that provided by Ragaz et al. ( 1 ) and the DBCG ( 4 ), which suggest that women with one to three positive lymph nodes are likely to receive relative benefits in breast cancer outcomes and survival that are relatively similar to those for patients with four or more positive lymph nodes. For some physicians and their patients, these relative benefits will translate into absolute benefits important enough for treatment. This, however, is not an ideal situation. Subgroup analyses may be affected by lack of power to discern important clinical differences. With the increasing use of more effective and potentially cardiotoxic agents (such as anthracyclines and traztuzumab), it is not clear that the same relative benefits will be realized. To avoid toxicity, locoregional radiation therapy is now often given after chemotherapy without treatment of the internal mammary lymph nodes, which also may potentially decrease its effectiveness. Given these uncertainties, a recent survey suggests that radiation oncologists differ among themselves on whether locoregional radiation therapy should be given after mastectomy to patients with one to three positive lymph nodes ( 22 ). Breast cancer is the most common cancer in women, affecting more than 200 000 women per year in the United States and more than 1 000 000 women per year worldwide. Consequently, our treatment policies, even if they are directed at small subgroups, are likely to affect thousands of women per year. We owe it to our patients that our management strategies use the highest possible level of evidence, i.e., that from large, randomized trials. Although randomized trials are difficult to conduct, they may provide additional benefits, including the ability to use modern genetic-based correlative science to identify patients most likely to benefit from a certain treatment ( 23 ). Because uncertainties continue about the effectiveness of locoregional radiation therapy after mastectomy in moderate-risk patients, we strongly urge that a randomized controlled trial be mounted to resolve these uncertainties.