Jian Zheng1, Jayasree Chakraborty1, William C Chapman2, Scott Gerst3, Mithat Gonen4, Linda M Pak1, William R Jarnagin1, Ronald P DeMatteo1, Richard K G Do3, Amber L Simpson5. 1. Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY. 2. Department of Surgery, Washington University School of Medicine, St Louis, MO. 3. Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY. 4. Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY. 5. Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY. Electronic address: simpsonl@mskcc.org.
Abstract
BACKGROUND: Microvascular invasion (MVI) is a significant risk factor for early recurrence after resection or transplantation for hepatocellular carcinoma (HCC). Knowledge of MVI status would help guide treatment recommendations, but is generally identified after operation. This study aims to predict MVI preoperatively using quantitative image analysis. STUDY DESIGN: One hundred and twenty patients from 2 institutions underwent resection of HCC from 2003 to 2015 were included. The largest tumor from preoperative CT was subjected to quantitative image analysis, which uses an automated computer algorithm to capture regional variation in CT enhancement patterns. Quantitative imaging features by automatic analysis, qualitative radiographic descriptors by 2 radiologists, and preoperative clinical variables were included in multivariate analysis to predict histologic MVI. RESULTS: Histologic MVI was identified in 19 (37%) patients with tumors ≤5 cm and 34 (49%) patients with tumors >5 cm. Among patients with tumors ≤5 cm, none of the clinical findings or radiographic descriptors were associated with MVI; however, quantitative features based on angle co-occurrence matrix predicted MVI with an area under curve of 0.80, positive predictive value of 63%, and negative predictive value of 85%. In patients with tumors >5 cm, higher α-fetoprotein level, larger tumor size, and viral hepatitis history were associated with MVI, and radiographic descriptors were not. However, a multivariate model combining α-fetoprotein, tumor size, hepatitis status, and quantitative feature based on local binary pattern predicted MVI with area under curve of 0.88, positive predictive value of 72%, and negative predictive value of 96%. CONCLUSIONS: This study reveals the potential importance of quantitative image analysis as a predictor of MVI.
BACKGROUND: Microvascular invasion (MVI) is a significant risk factor for early recurrence after resection or transplantation for hepatocellular carcinoma (HCC). Knowledge of MVI status would help guide treatment recommendations, but is generally identified after operation. This study aims to predict MVI preoperatively using quantitative image analysis. STUDY DESIGN: One hundred and twenty patients from 2 institutions underwent resection of HCC from 2003 to 2015 were included. The largest tumor from preoperative CT was subjected to quantitative image analysis, which uses an automated computer algorithm to capture regional variation in CT enhancement patterns. Quantitative imaging features by automatic analysis, qualitative radiographic descriptors by 2 radiologists, and preoperative clinical variables were included in multivariate analysis to predict histologic MVI. RESULTS: Histologic MVI was identified in 19 (37%) patients with tumors ≤5 cm and 34 (49%) patients with tumors >5 cm. Among patients with tumors ≤5 cm, none of the clinical findings or radiographic descriptors were associated with MVI; however, quantitative features based on angle co-occurrence matrix predicted MVI with an area under curve of 0.80, positive predictive value of 63%, and negative predictive value of 85%. In patients with tumors >5 cm, higher α-fetoprotein level, larger tumor size, and viral hepatitis history were associated with MVI, and radiographic descriptors were not. However, a multivariate model combining α-fetoprotein, tumor size, hepatitis status, and quantitative feature based on local binary pattern predicted MVI with area under curve of 0.88, positive predictive value of 72%, and negative predictive value of 96%. CONCLUSIONS: This study reveals the potential importance of quantitative image analysis as a predictor of MVI.
Authors: William C Chapman; Goran Klintmalm; Alan Hemming; Neeta Vachharajani; Maria B Majella Doyle; Ron DeMatteo; Victor Zaydfudim; Haniee Chung; Keith Cavaness; Robert Goldstein; Ivan Zendajas; Laleh G Melstrom; David Nagorney; William Jarnagin Journal: J Am Coll Surg Date: 2015-01-06 Impact factor: 6.113
Authors: Su Yeon Ahn; Jeong Min Lee; Ijin Joo; Eun Sun Lee; Soo Jin Lee; Gi Jeong Cheon; Joon Koo Han; Byung Ihn Choi Journal: Abdom Imaging Date: 2015-04
Authors: Timothy M Pawlik; Ana L Gleisner; Robert A Anders; Lia Assumpcao; Warren Maley; Michael A Choti Journal: Ann Surg Date: 2007-03 Impact factor: 12.969
Authors: F van Zijl; M Mair; A Csiszar; D Schneller; G Zulehner; H Huber; R Eferl; H Beug; H Dolznig; W Mikulits Journal: Oncogene Date: 2009-08-31 Impact factor: 9.867
Authors: S M Schlichtemeier; T C Pang; N E Williams; A J Gill; R C Smith; J S Samra; V W T Lam; M Hollands; A J Richardson; H C Pleass; S Nozawa; M Albania; T J Hugh Journal: Eur J Surg Oncol Date: 2016-06-11 Impact factor: 4.424
Authors: Sudeep Banerjee; David S Wang; Hyun J Kim; Claude B Sirlin; Michael G Chan; Ronald L Korn; Aaron M Rutman; Surachate Siripongsakun; David Lu; Galym Imanbayev; Michael D Kuo Journal: Hepatology Date: 2015-07-01 Impact factor: 17.425
Authors: Jennifer S Golia Pernicka; Johan Gagniere; Jayasree Chakraborty; Rikiya Yamashita; Lorenzo Nardo; John M Creasy; Iva Petkovska; Richard R K Do; David D B Bates; Viktoriya Paroder; Mithat Gonen; Martin R Weiser; Amber L Simpson; Marc J Gollub Journal: Abdom Radiol (NY) Date: 2019-11
Authors: John M Creasy; Kristen M Cunanan; Jayasree Chakraborty; John C McAuliffe; Joanne Chou; Mithat Gonen; Victoria S Kingham; Martin R Weiser; Vinod P Balachandran; Jeffrey A Drebin; T Peter Kingham; William R Jarnagin; Michael I D'Angelica; Richard K G Do; Amber L Simpson Journal: Ann Surg Oncol Date: 2020-09-20 Impact factor: 5.344