Prediction of lymph node metastases using pre-treatment PET radiomics of the primary tumour in esophageal adenocarcinoma: an external validation study

• DOI: 10.1259/bjr.20201042
• Publication Date: 2020-12-02T19:46:27Z
• Authors: Chong Zhang; Zhenwei Shi; Petros Kalendralis; Phil Whybra; Craig Parkinson; Maaike Berbee; Emiliano Spezi; Ashley Roberts; Adam Christian; Wyn Lewis; Tom Crosby; Andre Dekker; Leonard Wee; Kieran G Foley
• Keywords: Esophageal adenocarcinoma

To improve clinical lymph node staging (cN-stage) in oesophageal adenocarcinoma by developing and externally validating three prediction models; one with variables only, positron emission tomography (PET) radiomics a combined model.Consecutive patients fluorodeoxyglucose (FDG) avid tumours treated neoadjuvant therapy between 2010 2016 two international centres (n = 130 n 60, respectively) were included. Four (age, gender, T-stage tumour regression grade) PET from the primary used for model development. Diagnostic accuracy, area under curve (AUC), discrimination calibration calculated each model. The prognostic significance was also assessed.The incidence of metastases 58% both cohorts. areas clinical, models 0.79, 0.69 0.82 developmental cohort, 0.65, 0.63 external validation good demonstrated. current cN-stage development cohorts 0.60 0.66, respectively. For overall survival, achieved best performance cohort (X2 6.08, df 1, p 0.01).Accurate diagnosis is crucial prognosis guiding treatment decisions. Despite finding improved predictive using derived not fully replicated an cohort.This study attempted to validate new radiomics. A combining metastases, but these results replicated.