Consideration of clinicopathologic features improves patient stratification for multimodal treatment of gastric cancer

Abstract

// In Cho 1,2 , In Gyu Kwon 1,3 , Ali Guner 4 , Taeil Son 5,6 , Hyoung-Il Kim 5,6 , Dae Ryong Kang 7 , Sung Hoon Noh 5,6 , Joon Seok Lim 8 and Woo Jin Hyung 1,5,6,7,9 1 Department of Surgery, Graduate School, Yonsei University College of Medicine, Seoul, Republic of Korea 2 Department of Surgery, Soonchunhyang University College of Medicine, Bucheon, Republic of Korea 3 Department of Surgery, Keimyung University School of Medicine, Daegu, Republic of Korea 4 Department of General Surgery, Karadeniz Technical University, Trabzon, Turkey 5 Department of Surgery, Yonsei University College of Medicine, Seoul, Republic of Korea 6 Gastric Cancer Center, Yonsei Cancer Center, Yonsei University Health System, Seoul, Republic of Korea 7 Department of Humanities and Social Medicine, Ajou University School of Medicine, Suwon, Republic of Korea 8 Department of Diagnostic Radiology, Yonsei University College of Medicine, Seoul, Republic of Korea 9 Robot and MIS Center, Severance Hospital, Yonsei University Health System, Seoul, Republic of Korea Correspondence to: Woo Jin Hyung, email: // Keywords : gastric cancer, preoperative staging, diagnostic accuracy, patient stratification, multimodal treatment Received : December 05, 2016 Accepted : May 12, 2017 Published : June 22, 2017 Abstract Preoperative staging of gastric cancer with computed tomography alone exhibits poor diagnostic accuracy, which may lead to improper treatment decisions. We developed novel patient stratification criteria to select appropriate treatments for gastric cancer patients based on preoperative staging and clinicopathologic features. A total of 5352 consecutive patients who underwent gastrectomy for gastric cancer were evaluated. Preoperative stages were determined according to depth of invasion and nodal involvement on computed tomography. Logistic regression analysis was used to identify clinicopathological factors associated with the likelihood of proper patient stratification. The diagnostic accuracies of computed tomography scans for depth of invasion and nodal involvement were 67.1% and 74.1%, respectively. Among clinicopathologic factors, differentiated tumor histology, tumors smaller than 5 cm, and gross appearance of early gastric cancer on endoscopy were shown to be related to a more advanced stage of disease on preoperative computed tomography imaging than actual pathological stage. Additional consideration of undifferentiated histology, tumors larger than 5 cm, and grossly advanced gastric cancer on endoscopy increased the probability of selecting appropriate treatment from 75.5% to 94.4%. The addition of histology, tumor size, and endoscopic findings to preoperative staging improves patient stratification for more appropriate treatment of gastric cancer.