Ten-year Results of a Randomized Trial of Internal Mammary Chain Irradiation after Mastectomy

Reviewer: Christine Hill-Kayser, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 9, 2009

Presenter: P. Romestaing
Presenter's Affiliation: Centre De Radiotherapie Charcot, Ste Foy Les Lyon, France
Type of Session: Plenary


  • Radiation delivery to nodal regions following mastectomy for breast cancer remains controversial.
  • Among clinical discussion points remains the question of the need for delivery of radiotherapy to the internal mammary lymph nodes.
  • Although, directed delivery of radiation to the internal mammary chain may increase the comprehensive nature of nodal radiation, recent work has also demonstrated significantly increased risk of cardiac disease following radiation for breast cancer. Radiation of the internal mammary nodes may significantly impact scatter to the lungs and heart.
  • Surgical studies have demonstrated that risk of internal mammary node involvement with tumor may be increased by other clinical tumor factors – these include central/inner tumor location, associated with a 10% risk of internal mammary node involvement, and central/inner tumor location with axillary node involvement, associated with a 40% risk of internal mammary node involvement.
  • The Early Breast Cancer Cooperative Trialists’ Group (EBCCTG) demonstrated in a recent meta-analysis (Lancet, 2005) a benefit of post-mastectomy radiation in women with positive axillary lymph nodes having undergone mastectomy (PMRT). Post-mastectomy radiation was associated with decreased risk of loco-regional recurrence and improved overall survival with 15 years of follow-up. The majority of studies included in this meta-analysis (24/25) did include internal mammary irradiation.
  • Prior findings thus demonstrate that women having undergone mastectomy may benefit from radiation of the internal mammary nodes, that this may be associated with increased toxicity, and that certain subsets of women may be at particular risk for internal mammary node involvement.
  • Although these findings are hypothesis-generating, the questions of potential risk and benefits of internal mammary chain radiotherapy have not been definitively answered as of yet.
  • The trial described here was designed to evaluate the impact of internal mammary node radiotherapy on overall survival in breast cancer patients treated with mastectomy.

Materials and Methods

  • This study was designed as a randomized phase III trial that evaluated chest wall, axillary, and supraclavicular radiation with and without internal mammary radiation.
  • The study’s primary endpoint was overall survival at 10 years.
    • The expected benefit was 10% at 10 years.
    • The accrual goal was 1200 patients to achieve 90% power.
  • Inclusion criteria were as follows:
    • 75 years of age or younger
    • Stage I or II adenocarcinoma of the breast with the following stipulations:
      • Central/inner tumor location with or without pathologic axillary disease (pN0-N1).
      • Outer quadrant location with pathologic axillary disease (pN1).
      • Tumor size at least 1 cm.
  • All patients had undergone mastectomy with available pathologic information. Patients were stratified by tumor region within the breast (outer versus inner), nodal status (pN0 vs. pN1), and systemic treatment (chemotherapy versus none).
  • Patients were randomized to receive internal mammary irradiation in addition to chest wall, axillary, and supraclavicular radiation versus chest wall, axillary, and supraclavicular radiation only.
  • Internal mammary radiation consisted of mixed photon/electron irradiation delivered to the first 5 intercostal spaces with a depth of 2-3.5 cm. The inner field border did not extend past the sternal midline.
    • Patients received 12.5 Gray (Gy) in 5 fractions using photons, and 32.5 Gy in 13 fractions using electrons.
  • Remaining fields were delivered according to individual practice guidelines.
  • Systemic treatment was delivered according to the discretion of the treating physician.


  • Between January, 1991 and December, 1997, 1407 patients from 12 centers were enrolled on this study. Of these, 83 were excluded, so a total of 1334 patients were included in the analysis.
  • Patient characteristics were balanced between the two study arms, with a mean age of 56.5 years across the study cohort. 75% of patients had pathologically positive lymph nodes, and 86% received some form of systemic treatment (chemotherapy, hormones, or both).
  • Median follow-up was 10 years, and a total of 536 deaths were observed.
    • Of these, 370 were documented to be due to breast cancer progression.
    • 54 were due to other causes.
    • Cause of death was unknown in 112 cases.
  • Overall survival did not differ significantly between the two groups, with 10-year overall survival being 62.6% in the group receiving internal mammary irradiation and 59.3% in the group not (p = 0.8). Overall survival rates at 3 and 5 years in the two groups were 88.7%/ 87.1% and 79.2%/ 75.6%, respectively, with no significant difference between the groups.
  • Several subgroup analyses were performed, and no difference in overall survival was described between the group receiving internal mammary irradiation and the group not in the following subgroups: those with medial tumors and negative lymph nodes, those with medial tumors and positive axillary nodes, and those with outer tumors and positive axillary nodes. All subgroups were further broken down to patients receiving chemotherapy versus those not, and still no difference in overall survival was observed.
  • No increase in grade III-IV heart, skin, or lung toxicity was observed in either group.

Author's Conclusions

  • The authors conclude that no survival benefit to irradiation of the internal mammary chain after mastectomy appears to exist, and that no subgroup examined in this study appears to benefit from this treatment.
  • They also note that adding internal mammary node irradiation did not appear to increase cardiac toxicity.
  • They acknowledge that radiotherapy plans were not subjected to centralized review, and that this may be a weakness of the study presented here.

Clinical/Scientific Implications

  • This is a well-designed, interesting study that attempts to answer an important and controversial question in the breast literature.
  • In this large cohort of patients, no survival difference was detected with addition of internal mammary node irradiation following mastectomy. Several pieces of information remain incomplete, however, and will be of interest as the data matures:
    • First, the median follow-up of 10 years is relatively limited in the setting of breast cancer. Indeed, the Early Breast Cancer Cooperative Trialists’ Group’s recent meta-analysis did not show a clear benefit of post-mastectomy radiotherapy until 15 years of follow-up. An update of this data will thus be of interest.
    • Additionally, the impact of post-mastectomy radiotherapy on survival appeared to be greatest in younger patients in the meta-analysis. A subset analysis of the patients enrolled on this study by age would thus be of interest and could potentially elucidate a group of patients for whom benefit would be achieved with the addition of internal mammary irradiation.
    • Along the same lines, cardiac toxicity may not be manifest with this length of follow-up, and further information will be of interest in the future.
  • A further point is that the study presented here was designed to detect a 10% survival benefit to the addition of internal mammary irradiation at ten years. Within the oncologic literature, a 10% survival benefit from a single treatment is uncommonly detected, and a lesser benefit may exist but would not be detected with the power of this study.
  • Finally, the authors provide no information on loco-regional control in their presentation. Chest wall recurrence may be extremely difficult to manage, and may detract significantly from quality of life in both the curative and palliative settings. Information regarding rates of chest wall recurrence in the two groups is important for clinical decision-making.
  • The authors have provided an interesting addition to the breast cancer literature. As their data matures, it may be practice-changing. At this point, however, we cannot eliminate the possibility that a subset of, potentially younger, patients may benefit from internal mammary irradiation with longer follow-up. We can also not decisively conclude that this technique will not result in increased risk of cardiac toxicity.