Radiotherapy Dose Received by the Internal Mammary Chain Lymph Nodes in Cases with Relapse at this Site: A Case-Control Study

Abstract

Listen

Translate article

<h3>Purpose/Objective(s)</h3> Studies of sentinel lymph node biopsy of internal mammary nodes (IMN) in breast cancer show a 15% positivity rate, if untreated these may serve as a reservoir for regional recurrence (RR) and/or distant metastases. Randomized clinical trials examining the benefit of IMN radiotherapy (RT) haven't demonstrated an overall survival or disease-free survival benefit. However, in one study an ad hoc sub-analysis of patients with inner/central tumors showed a 7-year breast cancer mortality of 10.2% without IMNRT and 4.9% with IMNRT (Kim Y. JAMA Oncol 2022;8(1):96-105). IMNs often get some incidental dose from tangent RT, but this dose isn't known and we hypothesize that it might diminish the apparent benefit of deliberate IMNRT. <h3>Materials/Methods</h3> Patients diagnosed with non-metastatic breast cancer between 2005 and 2014, who relapsed with IMN disease, were identified in a population-based database. They were matched 1:3 by age, diagnosis year, pathological features and treatment received. An IMN CTV was outlined on planning CT scans covering the internal mammary vessels in the first 3 intercostal spaces. The equivalent dose in 2 Gy fractions to the IMNs was calculated to account for different RT prescriptions. A two-sample t-test and Cox regression analysis were used to evaluate the effect of IMN dose as a continuous variable on RR and breast cancer-specific survival (BCSS). <h3>Results</h3> 70 patients with IMN relapse were identified and 210 cases without relapse were matched. Median follow up was 10.8 years and median age 59 years. The matching of relapse and non-relapse cases, respectively, gave well balanced cohorts: most were pT2 (47.1% and 47.6%), pN2 and pN3 comprised 25.7% and 20.5% and tumor location was inner/central in 35.3% and 35.2%. RT was delivered to 43 (61.4%) relapse patients and 130 (61.9%) matched patients with 11 (15.7%) and 52 (24.8%) receiving intentional IMN coverage. The mean IMN dose was 44.9 Gy (94.1%) for intentional coverage and 13.3 Gy (28.3%) for incidental coverage. The average minimum dose to the IMNs was 6.8 Gy (14.4%) for the relapse cases and 14.7 Gy (31.0%) for the matches (p = 0.008). The average mean dose to the IMNs was 18.4 Gy (38.7%) for the relapse cases and 26.9 Gy (56.7%) for the matches (p <0.001). For the whole cohort, Cox regression modelling showed that the mean IMN dose was not significantly associated with RR (HR 0.87, 95% CI 0.74-1.02, p=0.08) or BCSS (HR 0.94, 95% CI 0.82-1.09, p=0.4). Similar results were shown for an N2/N3 subset, RR (HR 0.80, 95% CI 0.62-1.03, p=0.08) and BCSS (HR 0.86, 95% CI 0.70-1.07, p=0.2), and for a subset with inner/central tumors, RR (HR 0.87, 95% CI 0.65-1.17, p=0.4) and BCSS (HR 0.99, 95% CI 0.78-1.25, p=1.0). <h3>Conclusion</h3> Patients with IMN relapse received significantly lower mean IMN doses than those without IMN relapse, however, our sample size did not show a difference in clinical outcomes. The impact on IMN relapse from incidental RT dose should be considered when planning future studies in this field.