Determining the indications for post mastectomy radiotherapy: moving from 20th century clinical staging to 21st century biological criteria.

In the era of effective systemic therapy, the decision on whether or not to recommend post-operative adjuvant radiotherapy for intermediate risk breast cancer remains a point of discussion in many multidisciplinary tumour board meetings. Variations in advice regarding post-mastectomy radiotherapy have been demonstrated within the context of guideline-based practice [1]. A recent survey of nodal irradiation policy in European Organisation for Research and Treatment of Cancer (EORTC) centres also demonstrated a wide variation in practice [2]. On the one hand, there is level 1 evidence for a significant survival benefit from the addition of post mastectomy radiotherapy, including the chest wall from the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) overview for patients with nodal metastases, including low-volume N1 disease [3]. There is also increasing evidence for the addition of peripheral nodal irradiation in early breast cancer from two recently reported trials. One is the EORTC internal mammary chain and medial supraclavicular radiation (IM-MS) trial, reported by Poortmans et al. and the other is the National Cancer Institute of Canada Clinical Trials Group NCIC CTG M.20 trial of comprehensive regional nodal irradiation in patients treated with breast conserving therapy (Whelan et al, both papers currently in press in the New England Journal of Medicine). On the other hand, the increasing use of effective systemic therapy, often not available or applied to the patients included in the aforementioned trials, has improved the overall survival in patients with pN1 disease, i.e. one to three nodal metastases. The EBCTCG overview is dominated by the data from the Danish Breast Cancer Cooperative Group trials (1441 of 2801 patients with one to three positive nodes). Even in the subgroup analysis of the DBCG 82b and 82c trials in which at least eight nodes were removed from the axilla, the locoregional recurrence rate was very high (27%) in the non-irradiated group [4]. In addition, the adjuvant chemotherapy regime of cyclophosphamide, methotrexate and 5-FU (CMF) used in the 82b trial is currently considered suboptimal. Anthracycline-based chemotherapy with or without taxanes has replaced CMF as the current standard for women with one to three positive nodes and confers a significant survival advantage over CMF [5]. Five years of Tamoxifen or longer has been shown to improve survival compared with shorter treatment duration [6, 7], and Her-2 neu inhibition was not available at the time of the trials included in the overview. Therefore, although the relative risk reduction from radiotherapy might be significant, the absolute gains may only be modest. Punglia et al. [8] have suggested a bell-shaped curve to describe the effect of improving local control with local therapy in relation to the effectiveness of systemic therapy, a concept adapted by Poortmans [9]. The basic concept is that with increasing effectiveness of systemic therapy, the survival gain from improving local control with radiotherapy improves as the metastatic risk decreases, but only up to a certain level, and beyond that it lowers again. Furthermore, any benefits have to be weighed against potential long-term toxicity of the treatment. As a result, many national evidence-based guidelines set caveats in their advice regarding adjuvant post mastectomy radiotherapy, or radiotherapy to the regional lymph nodes, or do not advise it at all [10, 11]. The article by Nordenskjöld et al. ‘No clear effect of postoperative radiotherapy on survival of breast cancer patients with 1–3 positive nodes’ in this issue is very relevant to the current debate, as the results described are non-concordant with the results of the EBCTCG overview, and the recently published nodal radiotherapy trials. This epidemiological study describes a population-based analysis of 4448 patients treated in the years 1989–2006 in two Swedish regions, with/without adjuvant chest wall or regional nodal radiotherapy. Apart from the radiotherapy policy, the treatment protocol of these two regions was identical. Patients with one to three positive nodes on axillary dissection all received optimal systemic therapy, based on the St Gallen breast cancer panel recommendations at the time. The relative survival analysis demonstrated no difference in survival between the two regions related to addition or omission of post mastectomy radiotherapy, or regional nodal radiotherapy. This is contradictory to the overview results, and the authors suggest that the use of modern systemic therapy may, in part, explain these results. We hope that some of the issues raised will be resolved with data from the BIG 02-04 MRC EORTC SUPREMO trial, which recruited >1600 patients between April 2007 and May 2013 [12]. This trial included patients with intermediate risk stage II breast cancer after mastectomy who also received modern systemic therapy. Patients were prospectively randomized to receive chest wall radiotherapy or not. However, the final results of this trial are expected in 2023 at the earliest. In the meantime, the policy regarding adjuvant radiotherapy for intermediate risk breast cancer is becoming more difficult to extrapolate from trials such as those included in the overview, all of which included some degree of axillary dissection. The staging axillary dissection is increasingly being replaced by the sentinel node procedure. The results of the recent EORTC AMAROS trial [13] will probably lead to an increased use of primary axillary radiotherapy for patients with a positive sentinel node, instead of axillary clearance. In addition, neoadjuvant systemic therapy is increasingly used for stage II disease, making decisions based on the post (chemo-)therapy staging complex. Should we determine radiotherapy indications based on pre-treatment TNM classification, or should we adapt radiotherapy policy based on the downstaging achieved with neoadjuvant systemic treatment? In the SUPREMO trial, patients with neoadjuvant and post-operative adjuvant systemic therapy were both included in the randomisation, so the results will hopefully be relevant for future treatment scenarios as well. Annals of Oncology editorials