3D MR elastography-based stiffness as a marker for predicting tumor grade and subtype in cervical cancer
Yuanqiang Xiao,Wenying Chen,Xi Long,Mengsi Li,Lina Zhang,Chang Liu,Ying Deng,Chao Li,Bingjun He,Jun Chen,Jin Wang
DOI: https://doi.org/10.1016/j.mri.2024.03.006
IF: 3.13
2024-03-15
Magnetic Resonance Imaging
Abstract:Background Increasing evidence has indicated that high tissue stiffness (TS) may be a potential biomarker for evaluation of tumor aggressiveness. Purpose To investigate the value of magnetic resonance elastography (MRE)-based quantitative parameters preoperatively predicting the tumor grade and subtype of cervical cancer (CC). Study type Retrospective. Population Twenty-five histopathology-proven CC patients and 7 healthy participants. Field strength/sequence 3.0T, magnetic resonance imaging (MRI) (LAVA-flex) and MRE with a three-dimensional spin-echo echo-planar imaging. Assessment The regions of interest (ROIs) were manually drawn by two observers in tumors to measure mean TS, storage modulus (G′), loss modulus (G′′) and damping ratio (DR) values. Surgical specimens were evaluated for tumor grades and subtypes. Statistical tests Intraclass correlation coefficient (ICC) was expressed in terms of inter-observer agreements. t -test or Mann-Whitney nonparametric test was used to compare the complex modulus and apparent diffusion coefficient (ADC) values between different tumor groups. Area under the receiver operating characteristic curve (AUC) analysis was used to evaluate the diagnostic performance. Results The TS of endocervical adenocarcinoma (ECA) group was significantly higher than that in squamous cell carcinoma (SCC) group (5.27 kPa vs. 3.44 kPa, P = 0.042). The TS also showed significant difference between poorly and well/moderately differentiated CC (5.21 kPa vs. 3.47 kPa, P = 0.038), CC patients and healthy participants (4.18 kPa vs. 1.99 kPa, P < 0.001). The cutoff value of TS to discriminate ECA from SCC was 4.10 kPa (AUC: 0.80), while it was 4.42 kPa to discriminate poorly from well/moderately differentiated CC (AUC: 0.83), and 2.25 kPa to distinguish normal cervix from CC (AUC: 0.88), respectively. There were no significant difference in G′′, DR and ADC values between any subgroups except for comparison of healthy participants and CC patients ( P = 0.001, P = 0.004, P < 0.001, respectively). Data conclusion 3D MRE-assessed TS shows promise as a potential biomarker to preoperatively assess tumor grade and subtype of CC.
radiology, nuclear medicine & medical imaging