Effect of bilateral supraclavicular postoperative radiotherapy in middle and lower thoracic esophageal carcinoma

Clinical Target Volume Auto-Segmentation of Esophageal Cancer for Radiotherapy After Radical Surgery Based on Deep Learning

Radiotherapy plays an important role in controlling the local recurrence of esophageal cancer after radical surgery. Segmentation of the clinical target volume is a key step in radiotherapy treatment planning, but it is time-consuming and operator-dependent. This paper introduces a deep dilated convolutional U-network to achieve fast and accurate clinical target volume auto-segmentation of esophageal cancer after radical surgery. The deep dilated convolutional U-network, which integrates the advantages of dilated convolution and the U-network, is an end-to-end architecture that enables rapid training and testing. A dilated convolution module for extracting multiscale context features containing the original information on fine texture and boundaries is integrated into the U-network architecture to avoid information loss due to down-sampling and improve the segmentation accuracy. In addition, batch normalization is added to the deep dilated convolutional U-network for fast and stable convergence. In the present study, the training and validation loss tended to be stable after 40 training epochs. This deep dilated convolutional U-network model was able to segment the clinical target volume with an overall mean Dice similarity coefficient of 86.7% and a respective 95% Hausdorff distance of 37.4 mm, indicating reasonable volume overlap of the auto-segmented and manual contours. The mean Cohen kappa coefficient was 0.863, indicating that the deep dilated convolutional U-network was robust. Comparisons with the U-network and attention U-network showed that the overall performance of the deep dilated convolutional U-network was best for the Dice similarity coefficient, 95% Hausdorff distance, and Cohen kappa coefficient. The test time for segmentation of the clinical target volume was approximately 25 seconds per patient. This deep dilated convolutional U-network could be applied in the clinical setting to save time in delineation and improve the consistency of contouring.

Patterns of failure and clinical outcomes of definitive radiotherapy for cervical esophageal cancer

Purpose

Because of the scarcity of cervical esophageal cancer (CEC), data for this disease entity is limited. We aim to evaluate the outcomes, prognostic factors and failure patterns of CEC treated by contemporary radiotherapy (RT).

Methods

We retrospectively analyzed 86 CEC patients consecutively treated between 2007 and 2015 by definitive RT with or without concurrent chemotherapy. RT was mainly delivered with Intensity Modulated Irradiation Therapy (IMRT) or Volumetric-Modulated Arc Therapy (VMAT). Statistical analyses were performed on survival, prognostic factors and failure patterns.

Results

The median follow-up time was 19.4 months. The 3-year overall survival (OS), local regional failure free survival (LRFFS), distant metastatic free survival (DMFS), and progression free survival (PFS) were 53.6%, 57.9%, 81.5% and 41.5%, respectively. Independent predictors for poorer OS were N stage, hoarseness and recurrent laryngeal nerve lymph node (RLN) involvement, and predictors for LRFFS were N stage and EQD2 (equivalent dose in 2 Gy fraction) to gross tumor volume (GTV), with ≥ 66Gy achieving local control of 94.7%. Patients receiving elective nodal irradiation (ENI) had better nodal regional control than those receiving involved field irradiation (IFI). 31 (36%) patients had treatment failure and 15 (17.4%), 8 (9.3%) and 14 (16.2%) patients had local, regional, and distant failure, respectively. 86.7% (13/15) local failures were within GTV, and supraclavicular region (62.5%, 5/8) was the most common regional failure site. No severe toxicities were observed.

Conclusions

Our results seem to indicate that good locoregional control might be achieved for CEC with adequate radiation dose and treatment planning approaches.