Preoperative imaging markers and PDZ-binding kinase tissue expression predict low-risk disease in endometrial hyperplasias and low grade cancers

// Anna Berg 1, 2 , Ankush Gulati 3 , Sigmund Ytre-Hauge 3, 4 , Kristine E. Fasmer 3 , Karen K. Mauland 1, 2 , Erling A. Hoivik 1, 2 , Jenny A. Husby 3, 4 , Ingvild L. Tangen 1, 2 , Jone Trovik 1, 2 , Mari K. Halle 1, 2 , Ingunn Stefansson 5, 6 , Lars A. Akslen 5, 6 , Kathrine Woie 2 , Line Bjorge 1, 2 , Helga B. Salvesen 1, 2 , Oyvind O. Salvesen 7 , Henrica M.J. Werner 1, 2 , Ingfrid S. Haldorsen 3, 4, * and Camilla Krakstad 1, 2, * 1 Centre for Cancer Biomarkers, Department of Clinical Science, University of Bergen, Bergen, Norway 2 Department of Gynecology and Obstetrics, Haukeland University Hospital, Bergen, Norway 3 Department of Radiology, Haukeland University Hospital, Bergen, Norway 4 Section of Radiology, Department of Clinical Medicine, University of Bergen, Norway 5 Department of Pathology, Haukeland University Hospital, Bergen, Norway 6 Department of Clinical Medicine, Centre for Cancer Biomarkers, Bergen, Norway 7 Unit for Applied Clinical Research, Department of Cancer Research and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway * These authors have contributed equally to this work Correspondence to: Camilla Krakstad, email: camilla.krakstad@uib.no Keywords: endometrial carcinoma, endometrial hyperplasia, PBK, MRI, FDG-PET/CT Received: April 27, 2017 Accepted: June 19, 2017 Published: July 31, 2017 ABSTRACT Purpose: Distinguishing complex atypical hyperplasia (CAH) from grade 1 endometrioid endometrial cancer (EECG1) preoperatively may be valuable in order to prevent surgical overtreatment, particularly in patients wishing preserved fertility or in patients carrying increased risk of perioperative complications. Material and methods: Preoperative histological diagnosis and radiological findings were compared to final histological diagnosis in patients diagnosed with CAH and EECG1. Imaging characteristics at preoperative magnetic resonance imaging (MRI) and fluorodeoxyglucose positron emission tomography/computer tomography (FDG-PET/CT) were compared with tumor DNA oligonucleotide microarray data, immunohistochemistry findings and clinicopathological annotations. Results: MRI assessed tumor volume was higher in EECG1 than in CAH (p=0.004) whereas tumor apparent diffusion coefficient value was lower in EECG1 (p=0.005). EECG1 exhibited increased metabolism with higher maximum and mean standard uptake values (SUV) than CAH (p≤0.002). Unsupervised clustering of EECG1 and CAH revealed differentially expressed genes within the clusters, and identified PDZ-binding kinase (PBK) as a potential marker for selecting endometrial lesions with less aggressive biological behavior. Conclusion: Both PBK expression and preoperative imaging yield promising biomarkers that may aid in the differentiation between EECG1 and CAH preoperatively, and these markers should be further explored in larger patient series.