Contents lists available at ScienceDirect European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad Research article Prognostic value of 18 F-FDG PET/MR imaging biomarkers in oesophageal squamous cell carcinoma Chih-Wei Yu a , Xin-Jia Chen b , Yen-Heng Lin a , Yao-Hui Tseng c , Ching-Chu Lu d , Bang-Bin Chen a , Shwu-Yuan Wei a , Jang-Ming Lee e, ⁎, Tiffany Ting-Fang Shih a, ⁎ a Department of Radiology and Medical Imaging, National Taiwan University College of Medicine and Hospital, No. 7, Chung-Shan South Rd, Taipei 10016, Taiwan b Department of Medical Imaging, Sin-Lau Medical Foundation the Presbyterian Church in Taiwan, No. 57, Sec. 1, Dongmen Rd., East Dist., Tainan 70142, Taiwan c Department of Medical Imaging, Cardinal Tien Hospital, No. 362, Zhongzheng Rd., Xindian Dist., New Taipei City 23148, Taiwan d Department of Nuclear Medicine, National Taiwan University Hospital, No. 7, Chung-Shan South Rd., Taipei 10016, Taiwan e Department of Surgery, National Taiwan University College of Medicine and Hospital, No. 7, Chung-Shan South Rd., Taipei 10016, Taiwan ARTICLE INFO Keywords: Positron emission tomography Magnetic resonance imaging Oesophageal carcinoma Diffusion-weighted imaging Progression-free survival Overall survival ABSTRACT Purpose: To correlate the clinical stage and prognosis of oesophageal squamous cell carcinoma (SCC) using the imaging biomarkers from integrated positron emission tomography (PET)/magnetic resonance imaging (MRI). Methods: In total, 54 consecutive patients with oesophageal SCC who receive PET/MRI scan were recruited before treatment. The imaging biomarkers used were the mean and minimal apparent diffusion coefficients (ADC mean and ADC min ), standardized uptake value (SUV), metabolic tumour volume (MTV), and total lesion glycolysis (TLG) of tumours. The correlation between each imaging biomarker and survival was investigated using the Cox proportional hazards model. Results: ADC mean was negatively correlated with SUV max (r = -0.414, P = 0.025). ADC min was negatively correlated with SUV max (r = -0.423, P = 0.001) and SUV peak (r = -0.402, P = 0.003), and was significantly lower in M1 than in M0 tumours (829.6 vs. 1069.8, P = 0.005). MTV was significantly higher in T3 + (P < 0.001), N1 + (P = 0.014) and TNM stage III + (P < 0.001) tumours. TLG was significantly higher in T3 + (P < 0.001), N1 + (P < 0.001), M1 (P = 0.045) and TNM stage III + (P < 0.001) tumours. The MTV/ ADC min ratio exhibited the highest area under the receiver operating characteristic curve (AUROC) for predicting M1 and advanced TNM stage tumours. Multivariate analysis for progression-free survival (PFS) and overall survival (OS) showed that a larger MTV/ADC min was associated with a shorter PFS and OS (P = 0.024 and 0.046, respectively). Conclusion: The imaging biomarkers in integrated PET/MRI may predict clinical stage and survival in patients with oesophageal SCC. 1. Introduction Oesophageal cancer is the 6th leading cause of cancer-related death in the world [1]. Oesophageal cancer has two main histological types: squamous cell carcinomas and adenocarcinomas. For most of the 20th century, squamous cell carcinoma has predominated and is associated with a 5-year survival rate of 15%–25 % [2]. However, the incidence of adenocarcinoma of the oesophagus has been rising rapidly in the Western nations for the past three decades [3]. According to the 7th edition of the AJCC staging system for oesophageal cancer, published in 2010, histopathologic cell type defines staging classification [4]. Multimodality diagnosis using endoscopic ultrasound, contrast-en- hanced computed tomography (CECT) of the neck, chest, and abdomen and whole-body 18 F-fluorodeoxyglucose (FDG)-positron emission to- mography (PET)/CT, is critical for staging before treatment of oeso- phageal cancer. Although 18 F-FDG-PET/CT has limited value for re- gional tumour staging in oesophageal cancer, it is useful to detect distant lymphatic and haematogenous metastases [5,6]. The use of the 18 F-FDG tracer in PET/CT allows for a quantitative assessment of cel- lular glucose metabolism in tumours by measuring the standardized uptake value (SUV). SUV of the primary tumour may serve as a prog- nostic factor in oesophageal cancer [7]. Furthermore, volume-based parameters such as metabolic tumour volume (MTV) and total lesion glycolysis (TLG) measured in 18 F-PET/CT may be valuable as prog- nostic biomarkers in oesophageal cancer imaging [8–11]. Presently, the role of magnetic resonance imaging (MRI) in https://doi.org/10.1016/j.ejrad.2019.108671 Received 18 March 2019; Received in revised form 5 September 2019; Accepted 12 September 2019 ⁎ Corresponding authors. E-mail addresses: jmlee@ntu.edu.tw (J.-M. Lee), ttfshih@ntu.edu.tw (T.T.-F. Shih). European Journal of Radiology 120 (2019) 108671 0720-048X/ © 2019 Elsevier B.V. All rights reserved. T