Premium
Tumor apparent diffusion coefficient as an imaging biomarker to predict tumor aggressiveness in patients with estrogen‐receptor‐positive breast cancer
Author(s) -
Shin Hee Jung,
Kim So Hee,
Lee Hee Jin,
Gong Gyungyub,
Baek Seunghee,
Chae Eun Young,
Choi Woo Jung,
Cha Joo Hee,
Kim Hak Hee
Publication year - 2016
Publication title -
nmr in biomedicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.278
H-Index - 114
eISSN - 1099-1492
pISSN - 0952-3480
DOI - 10.1002/nbm.3571
Subject(s) - medicine , effective diffusion coefficient , breast cancer , imaging biomarker , estrogen receptor , biomarker , nuclear medicine , cancer , pathology , oncology , magnetic resonance imaging , radiology , chemistry , biochemistry
The purpose of this retrospective study was to evaluate whether tumor apparent diffusion coefficient (ADC) was correlated with pathologic biomarkers such as tumor cellularity, Ki67, tumor‐infiltrating lymphocytes (TILs), and peritumoral lymphocytic infiltrate (PLI) in patients with estrogen receptor (ER)‐positive breast cancer. The study was approved by the institutional review board and informed consent was waived. From July 2014 to December 2014, we reviewed 140 ER‐positive tumors in 138 consecutive patients (range, 28–77 years; mean, 52 years) who underwent preoperative breast MRI and definitive surgery. All patients underwent diffusion‐weighted imaging with a 3T scanner. Two radiologists drew the region of interest of the entire tumor and obtained the mean and pixel‐based histogram of ADC. On pathology, two pathologists reviewed tumor cellularity, Ki67, TILs, and PLI. Multiple linear regression analysis was used to determine pathologic variables independently associated with ADC. Tumors with high tumor cellularity and high Ki67 had significantly lower ADCs than those with low tumor cellularity and low Ki67 ( P < 0.05 for all). Tumors without PLI had significantly higher standard deviation than those with PLI (0.23 ± 0.08 versus 0.18 ± 0.05; P < 0.001). Median ADC was negatively correlated with tumor cellularity ( r = −0.441), and Ki67 ( r = −0.382). The standard deviation of ADC was also negatively correlated with the degree of PLI ( r = −0.319). On multivariate linear regression analysis, tumor cellularity and Ki67 were independently associated with tumor ADC. Tumor ADC would be an MRI biomarker for the prediction of tumor aggressiveness indicators such as Ki67, tumor cellularity, and PLI in ER‐positive breast cancer. Copyright © 2016 John Wiley & Sons, Ltd.