Background: This study aimed to propose a machine learning model to predict the local response of resectable locally advanced esophageal squamous cell carcinoma (LA-ESCC) treated by neoadjuvant chemoradiotherapy (NCRT) using pretreatment 18-fluorodeoxyglucose positron emission tomography (FDG PET) images. Methods: The local responses of 98 patients were categorized into two groups (complete response and noncomplete response). We performed a radiomics analysis using five segmentations created on FDG PET images, resulting in 4250 features per patient. To construct a machine learning model, we used the least absolute shrinkage and selection operator (LASSO) regression to extract radiomics features optimal for the prediction. Then, a prediction model was constructed by using a neural network classifier. The training model was evaluated with 5-fold cross-validation. Results: By the LASSO analysis of the training data, 22 radiomics features were extracted. In the testing data, the average accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve score of the five prediction models were 89.6%, 92.7%, 89.5%, and 0.95, respectively. Conclusions: The proposed machine learning model using radiomics showed promising predictive accuracy of the local response of LA-ESCC treated by NCRT.

No abstract

Introduction:To compare the efficacy and safety of stereotactic body radiation therapy with or without transcatheter arterial chemoembolization for patients with small hepatocellular carcinoma who were ineligible for resection or ablation therapies.Methods:A total of 150 patients with 185 hepatocellular carcinoma (≤3 nodules, Child-Turcotte-Pugh class A or B, and no vascular or extrahepatic metastases) were treated with stereotactic body radiation therapy. In principle, transcatheter arterial chemoembolization was combined before stereotactic body radiation therapy (combination group), but some patients were treated with stereotactic body radiation therapy alone. The prescribed dose of stereotactic body radiation therapy was 48 Gy in 4 fractions at the isocenter and 40 Gy in 4 or 5 fractions at the dose covering 95% of the planning target volume. The overall survival, progression-free survival, local progression free survival, and complication rates were retrospectively compared between the groups. Local progression was defined as irradiated tumor growth in dynamic computed tomography follow-up. Tumor responses were assessed according to the Modified Response Evaluation Criteria in Solid Tumors. Treatment-related toxicities were evaluated according to the Common Terminology Criteria for Adverse Events version 4.0.Results:Twenty-eight and 122 patients were enrolled in the stereotactic body radiation therapy alone and combination groups, respectively. The median follow-up periods were 16 and 29 months, respectively. The 2-year overall, progression-free, and local progression-free survival times in stereotactic body radiation therapy alone and combination groups were 78.6% and 80.3% (P = .6583), 49.0% and 42.9% (P = .188), and 71.4% and 80.8% (P = .9661), respectively. The incidence of ≥grade 3 toxicities was 17.9% in stereotactic body radiation therapy alone group and 18.9% in combination group (P = .903).Conclusions:Stereotactic body radiation therapy alone may be a good treatment option for patients with small hepatocellular carcinoma who were ineligible for resection or ablation therapies.

The purpose of this study was to evaluate the efficacy and safety of volumetric modulated arc therapy (VMAT) after extrapleural pneumonectomy (EPP) in patients with malignant pleural mesothelioma (MPM). A total of 15 patients who received VMAT after EPP were enrolled. All patients were males, and the median age was 67 years (Stage IB in two, II in six, and III in seven patients). The clinical target volume (CTV) included the entire preoperative ipsilateral hemithorax and involved nodal stations. The CTV was generally expanded by 10–15 mm beyond the planning target volume (PTV). The dose prescription was designed to cover 95% of the PTV with 54 Gy in 30 fractions. The median follow-up period was 11 months. Treatment-related toxicities were evaluated by Common Terminology Criteria for Adverse Events (CTCAE) ver. 4. One-year local control, disease-free survival, and overall survival rates were 55.7% [95% confidence interval (CI): 25.6–85.8%], 29.3% (95% CI: 5.3–53.3%), and 43.1% (95% CI: 17.1–69.0%), respectively. According to the histological analysis, the one-year LC rate was significantly worse in patients with non-epithelial type (biphasic and sarcomatoid types) than in patients with epithelial type [epithelial type: 83.3% (95% CI, 53.5–100%), non-epithelial type: 0% (95% CI, 0%), P = 0.0011]. Grade 3 pneumonitis after VMAT was observed in three patients (20.0%); however, no patients died of pulmonary toxicity. VMAT appears to be relatively safe for patients with MPM after EPP because of the low pulmonary dose.

up and only a single patient died of liver toxicity. Among pts treated with definitive intent, only 10% experienced a change in CP score of 2. Neither liver mean dose nor PTV size predicted for CP score change 2. Following SBRT, 2 pts developed new grade III hypoalbuminemia, 11 pts developed new grade III hyperbilirubinemia, and 6 pts developed new grade III transaminitis. Liver mean dose, PTV size, and prescription dose did not predict for development of new grade III hepatotoxicity. However, in patients who did not have disease progression in the liver, RT dose 40 Gy predicted for grade III hepatotoxicity (pZ0.003). Conclusion: SBRT can be successfully employed as local therapy for wellselected patients with CP B and C HCC with low rates of clinically significant hepatotoxicity. SBRT is also feasible in such patients as a part of bridging or downstaging prior to liver transplant.