Clinical evaluation of a deep learning model for segmentation of target volumes in breast cancer radiotherapy.

PURPOSE/OBJECTIVE(S): Precise segmentation of clinical target volumes (CTV) in breast cancer is indispensable for state-of-the art radiotherapy. Despite international guidelines, significant intra- and interobserver variability exists, negatively impacting treatment outcomes. The aim of this study is to evaluate the performance and efficiency of segmentation of CTVs in planning CT images of breast cancer patients using a 3D convolutional neural network (CNN) compared to the manual process. MATERIALS/METHODS: An expert radiation oncologist (RO) segmented all CTVs separately according to international guidelines in 150 breast cancer patients. This data was used to create, train and validate a 3D CNN. The network's performance was additionally evaluated in a test set of 20 patients. Primary endpoints are quantitative and qualitative analysis of the segmentation data generated by the CNN for each level specifically as well as for the total PTV to be irradiated. The secondary endpoint is the evaluation of time efficiency. RESULTS: In the test set, segmentation performance was best for the contralateral breast and the breast CTV and worst for Rotter's space and the internal mammary nodal (IMN) level. Analysis of impact on PTV resulted in non-significant over-segmentation of the primary PTV and significant under-segmentation of the nodal PTV, resulting in slight variations of overlap with OARs. Guideline consistency improved from 77.14% to 90.71% in favor of CNN segmentation while saving on average 24 minutes per patient with a median time of 35 minutes for pure manual segmentation. CONCLUSION: 3D CNN based delineation for breast cancer radiotherapy is feasible and performant, as scored by quantitative and qualitative metrics.

Omitting radiation therapy in women with triple-negative breast cancer leads to worse breast cancer-specific survival.

PURPOSE: To examine locoregional recurrence (LRR) and breast cancer-specific survival (BCSS) after breast-conserving therapy (BCT) or mastectomy (ME) with or without radiation therapy (RT) in triple-negative breast cancer (TNBC). MATERIAL & METHODS: We identified non-metastatic TNBC cases from a single institution database. BCT, ME with RT (ME + RT) and ME only were compared with respect to LRR and BCSS. Cox regression models were used to analyze the association between prognostic factors and outcome. RESULTS: 439 patients fulfilled the inclusion criteria. Median follow-up was 10.2 years (interquartile range 7.9; 12.4 years). Patients in the BCT (n = 239), ME + RT (n = 116) and ME only (n = 84) group differed with respect to age, pT, pN, lymphovascular invasion, lymph node dissection and chemotherapy administration. Ten-year LRR rates were seven percent, three percent and eight percent for the BCT, ME + RT and ME only group, respectively. pN was associated with LRR. In multivariable analysis LRR were significantly lower in the ME + RT group compared to the BCT and the ME only group (p 0.037 and 0.020, respectively). Ten year BCSS was 87%, 84% and 75% for the BCT, ME + RT and ME only group, respectively. pT, pN, lymph node dissection, lymphovascular invasion and the administration of chemotherapy were associated with BCSS. In multivariable analysis BCSS was significantly lower in the ME only group compared to the BCT group and the ME + RT group (p 0.047 and 0.003, respectively). CONCLUSION: TNBC patients treated with ME without adjuvant RT showed significant lower BCSS compared to patients treated with BCT or ME + RT and significant more LRR compared to ME + RT when corrected for known clinicopathological prognostic factors.