FDG/PET-CT-Based Lymph Node Atlas in Breast Cancer Patients

Abstract

Purpose: The aim of this study was to localize locoregional lymph node metastases using positron emission tomography with fluorine 18-fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) data sets in a large cohort of patients and to evaluate the existing Radiation Therapy Oncology Group (RTOG) clinical target volume (CTV) and the European Society for Radiation Therapy & Oncology (ESTRO) CTV contouring guidelines.

Methods and materials: A total of 235 patients with 580 FDG/PET-CT positive locoregional lymph node metastases were included in our analysis. The patients were divided into 4 groups according to their course of disease (primary vs recurrent breast cancer) and the presence or absence of distant metastasis at the time of the FDG-PET/CT staging (distant metastasis vs no distant metastasis). All imaging data were imported into the planning system, and each lymph node was manually contoured. A patient with "standard anatomy" was chosen as a template, and all contoured structures were registered rigidly and nonrigidly to this patient. A comprehensive 3-dimensional atlas was created, including all identified lymph node metastases. The incidences of lymph node metastases were analyzed and are presented with color coding in the atlas. Lymph node levels (axillary, internal mammary, supraclavicular) were contoured according to RTOG and ESTRO guidelines and evaluated.

Results: The mean volume of the lymph nodes was 1.7 ± 2.6 cm³ with an average diameter of 1.3 ± 0.7 cm. Most lymph nodes were in level I (n = 316; 54.5%) followed by the supraclavicular region (n = 80; 13.8%), level II (n = 57; 9.8%), level III (n = 58; 10.0%), and the internal mammary region (n = 55; 9.5%). The covered lymph node volume was 69.8% ± 35.5% (69.1% ± 36.3%) for primary breast cancer and 57.6% ± 38.9% (51.1% ± 39.1%) for recurrent breast cancer using the RTOG (ESTRO) guidelines. The internal mammary region and supraclavicular region were affected more often in recurrent breast cancer compared with primary breast cancer. The occurrence of lymph node metastases outside the RTOG and ESTRO margins in patients with and without distant metastases was similar. The largest geometric deviations between RTOG/ESTRO CTV contours and lymph node occurrence were measured in the supraclavicular region, the internal mammary region, and level II.

Conclusions: The provided lymph node atlas illustrates where lymph node metastases occur in different clinical situations and presents areas at high risk (ie "hot spots" of lymph node metastases).