To the Editor:

I read the interesting article titled “The Role of Irradiation of the Internal Mammary Lymph Nodes in High-Risk Stage II to IIIA Breast Cancer Patients After High-Dose Chemotherapy: A Prospective Sequential Nonrandomized Study,” by Dr Stemmer et al.¹ The authors have analyzed 100 patients retrospectively and compared two groups of patients receiving internal mammary (IMN) irradiation or not and concluding a better disease-free survival and borderline overall survival increase in the IMN irradiated group.

I have couple of concerns regarding the article. There are some methodological flaws, as stated by the authors in the discussion. The patients have been irradiated by a standard 9 to 12 MeV electron beam directed to 1 cm depth under the sternum. However, as Recht et al pointed out, internal mammary lymph node location is not standard, and he and his colleagues found out that they are in midline and lie at first 3 cm depth from the skin in 78% of the patients.² Along with these findings, Recht et al recommended individualized computerized determination of the location of the internal mammary lymphatics. So it can be deduced that Stemmer et al have not actually properly treated more than 20% of the patients in the IMN irradiated group.

Another point we do not know is the status of internal mammary lymph nodes— whether they are diseased or not. The authors did not mention whether radiologic or pathologic evaluation has been done in the whole group. It should be noted that probability of IMN lymph node involvement is critical for this article. The authors are expected to stratify the patients according to the involvement risk criteria of the IMN along with the primary sites and sizes of the tumor. However, as I read the article, I understood that the authors have not reported the location sites of the tumor in the breast. The literature data showed that location of the tumor is important for internal mammary lymphatic involvement besides the number of involved axillary lymph nodes. An inner quadrant–located tumor with positive axillary lymph nodes has a probability of 44% to 65% internal mammary lymph node involvement, whereas an outer quadrant location has a probability of 19% to 42%.^3,4 The stratification of the patients according to tumor location may also help in elimination of potential biases. The other important issue is T stage of the patients. Although the authors reported axillary lymph node status, they did not report the tumor size, which greatly affects the probability of internal mammary involvement. The risk of IMN involvement for T1–T2 and T3 are 15% to 31%, and 28% to 48% respectively.^4,5

Regarding the Results section, it is hard to believe that IMN irradiation increases disease-free survival and leads to a borderline increase in overall survival. Clinical failure in IMN site is so rare that it is not that easy to identify the benefit of irradiation of this particular site in such a small patient group. Because the authors have not particularly reported the boundaries of DFS concept and rates of internal mammary relapse in both groups, such a conclusion seems unlikely.