CT Radiomics based on Machine Learning Predicting Pathologic Complete Response after Neoadjuvant Chemoradiotherapy for Esophageal Cancer

Abstract Neoadjuvant chemoradiotherapy (nCRT) has become the standard treatment for patients with locally advanced squamous cell esophageal cancer (LA-ESCC). Patients with the complete pathologic response (pCR) have significantly improved long-term survival. All efforts should be to improve the accuracy of predicting pCR. In this study, we investigate the use of radiomics based on machine learning to identify the pathologic complete response of patients with esophageal squamous cell carcinoma (ESCC) based on Computed Tomography (CT). Methods The study included 155 patients with pathologically confirmed LA-ESCC. All 155 patients underwent simulation CT before nCRT, Quantitative radiomics features were extracted from CT images of each patient. To explore the relationship between radiomics features and the pCR, we used five-fold cross validation to classify the training and the testing cohorts. The Least Absolute Shrinkage and Selectionator operator (Lasso) were used to select features useful for the grading of pCR in the training cohort. Different models were measured in the training cohort using accuracy, sensitivity, specificity, and area under the curve of the receiver operating characteristic curve (AUC). Results There were 155 patients. The pretreatment clinical stage was II in 16 patients (10.3%), III in 132 (85.2%), and IV in 7 (4.5%). Pathologic response was complete in 69 patients (44.5%), near-partial complete in 86 (55.5%). A total of 2193 radiomics features were extracted in the training set. After the use of statistical dimensionality reduction, five radiomics features were selected by Lasso to build radiomics signature. Prediction models for pCR were developed, and the model was able to predict pCR well in the training set(AUC = 0.902). In the testing cohorts, the model had a good performance in predicting pCR (AUC = 0.78). Conclusion This study showed that CT-based radiomics features could be used as biomarkers to predict the complete pathological response of esophageal cancer underwent Neoadjuvant chemoradiotherapy.

Download Full-text

Prognostic impact of inflammation-based scores in esophageal cancer patients achieving pathologic complete response after neoadjuvant chemoradiotherapy followed by surgery

Annals of Oncology ◽

10.1093/annonc/mdy282.018 ◽

2018 ◽

Vol 29 ◽

pp. viii213

Author(s):

J. Kang ◽

H.-S. Im ◽

Y.-H. Kim ◽

H.R. Kim ◽

J.H. Kim ◽

...

Keyword(s):

Esophageal Cancer ◽

Cancer Patients ◽

Pathologic Complete Response ◽

Neoadjuvant Chemoradiotherapy ◽

Complete Response ◽

Prognostic Impact

Download Full-text

Effect of Extending the Original CROSS Criteria on Tumor Response to Neoadjuvant Chemoradiotherapy in Esophageal Cancer Patients: A National Multicenter Cohort Analysis

Annals of Surgical Oncology ◽

10.1245/s10434-020-09372-y ◽

2020 ◽

Cited By ~ 1

Author(s):

Helena Hong Wang ◽

◽

Ellen C. de Heer ◽

Jan Binne Hulshoff ◽

Gursah Kats-Ugurlu ◽

...

Keyword(s):

Esophageal Cancer ◽

Cohort Analysis ◽

Locally Advanced ◽

Pathologic Complete Response ◽

Neoadjuvant Chemoradiotherapy ◽

Postoperative Mortality ◽

Complete Response ◽

Mortality And Morbidity ◽

The Cross ◽

The Impact

Abstract Background Extending the original criteria of the Chemoradiotherapy for Oesophageal Cancer followed by Surgery Study (CROSS) in daily practice may increase the treatment outcome of esophageal cancer (EC) patients. This retrospective national cohort study assessed the impact on the pathologic complete response (pCR) rate and surgical outcome. Patients and Methods Data from EC patients treated between 2009 and 2017 were collected from the national Dutch Upper Gastrointestinal Cancer Audit database. Patients had locally advanced EC (cT1/N+ or cT2-4a/N0-3/M0) and were treated according to the CROSS regimen. CROSS (n = 1942) and the extended CROSS (e-CROSS; n = 1359) represent patients fulfilling the original or extended CROSS criteria, respectively. The primary outcome was total pCR (ypT0N0), while secondary outcomes were local esophageal pCR (ypT0), surgical radicality, and postoperative morbidity and mortality. Results Overall, CROSS and e-CROSS did not differ in total or local pCR rate, although a trend was observed (23.2% vs. 20.4%, p = 0.052; and 26.7% vs. 23.8%, p = 0.061). When stratifying by histology, the pCR rate was higher in the CROSS group compared with e-CROSS in squamous cell carcinomas (48.2% vs. 33.3%, p = 0.000) but not in adenocarcinomas (16.8% vs. 16.9%, p = 0.908). Surgical radicality did not differ between groups. Postoperative mortality (3.2% vs. 4.6%, p = 0.037) and morbidity (58.3% vs. 61.8%, p = 0.048) were higher in e-CROSS. Conclusion Extending the CROSS inclusion criteria for neoadjuvant chemoradiotherapy in routine clinical practice of EC patients had no impact on the pCR rate and on radicality, but was associated with increased postoperative mortality and morbidity. Importantly, effects differed between histological subtypes. Hence, in future studies, we should carefully reconsider who will benefit most in the real-world setting.

Download Full-text