3003 - The Radiation Dose-Volume Data of Thymus Contribution to Lymphopenia of Post-Operative Radiotherapy in Early-Stage Breast Cancer Patients
Presenter(s)
Y. Tian; Department of Radiotherapy & Oncology, The Second Affiliated Hospital of Soochow University, Suzhou, China
Purpose/Objective(s): The radiation-induced lymphopenia (RIL) is a common side effect of radiotherapy (RT). Some studies have shown it is associated with deteriorating the treatment outcomes in different cancers. For breast cancer (BC) patients with postoperative RT, RIL is frequent in BC RT patients also, and several factors have shown to contribute to the occurrence and severity of RIL. Owing to its anatomical proximity within the thoracic area, the thymus is inevitably irradiated during RT for BC patient. However, the relationship between thymic irradiation and the RIL remains unclear. This study aims to analyze the correlation between thymus dose-volume parameters and RIL in patients receiving adjuvant RT.
Materials/Methods: We collected the data from ? - ? stage BC patients retrospectively. Absolute lymphocyte counts (ALC) (×109/L) were collected about before RT, during and after RT weekly. RIL was graded based on the lowest ALC (Nadir ALC) according to the CTCAE version 5.0 and calculate Nadir/baseline ALC. The thymus wad delineated according to anatomical boundaries in the RT planning system (TPS) by CT mediastinum window. Dosimetric parameters of each organs at risk (OARs) were obtained from TPS by dose-volume histograms (DVH). Stepwise multiple linear regression analysis was used to explore the factors associated with Nadir/baseline ALC. Using the Receiver Operating Characteristic (ROC) curve to derive the cut-off values for predicting = Grade 3 RIL (< 0.5×109/L).
Results: A total of 72 BC patients met the eligibility criteria, their median age was 46 years. 72.3% patients received adjuvant chemotherapy before RT, the most common chemotherapy regimens were Anthracycline + Axane (near 60%), and the median cycle number was 5. The proportion of patients with Grade 3 RIL was 46%. The thymus median volume (cm3), mean radiation-dose (Gy), V5Gy (%), and V10Gy (%) were 14.4 cm3, 5.1 Gy, 35.8%, and 6.4% respectively. Stepwise multiple linear regression analysis revealed that thymus V5 (P = 0.010) and baseline ALC (P = 0.000) was independently associated with Nadir/baseline ALC. ROC analysis revealed that the cut-off value for thymus V5Gy predicting the Grade 3 RIL is 36.1%. The combined model established using baseline ALC, RT techniques (3D-CRT, IMRT), DVH parameters for the heart and the bilateral lungs, and thymus V5 had the highest AUC of 0.871.
Conclusion: Except for the DVH of known OARs such as the heart and the bilateral lung, and the baseline ALC of patient. The study revealed RIL in early-stage BC patients with adjuvant RT is significantly correlated with higher dosimetric parameters of the thymus, among them thymus V5Gy (%) may serve as the optimal dosimetric indicator. Optimizing treatment plans through considering the thymus as a routine OAR for dose limitation may reduce the risk of severe lymphopenia. Our results may have implications for clinical guidance, but require validation by future large-size prospective clinical trials.