MR volumetry in predicting the aggressiveness of endometrioid adenocarcinoma: correlation with final pathological results. (May 2020)

• Record Type: Journal Article
• Authors: Cai, Shulei; Zhang, He; Chen, Xiaojun; Wang, Tianping; Lu, Jiaqi; Liu, Xuefen; Zhang, Guofu
• Is Part Of: Acta radiologica. Volume 61:Number 5(2020)
• Page Start: 705
• Page End: 713
• Publication Date: 2020-05
• Subjects: Endometrioid adenocarcinoma -- magnetic resonance imaging -- frozen section -- myometrial invasion; Radiology, Medical -- Periodicals; Radiography, Medical -- Periodicals; Radiotherapy -- Periodicals; 616.0757
• DOI: 10.1177/0284185119877331
• Languages: English
• ISSNs: 0284-1851

Abstract

Background: Magnetic resonance (MR) has been widely used in predicting the aggressiveness of endometrioid adenocarcinoma. However, the diagnostic value of the MR volume of the lesion has been controversial. Purpose: To determine whether the whole-lesion MR volume measurement could be used as a better predictor for evaluating the aggressiveness of endometrioid adenocarcinoma. Material and Methods: In this retrospective study, we include 357 patients with pathologically demonstrated endometrioid adenocarcinoma at our institution between 1 January 2013 and 31 December 2018. Whole-lesion MR volume was calculated on sagittal T2-weighted images with ITK-SNAP software on a personal computer. Results: According to the receiver operating characteristics curve analysis, whole-lesion MR volume has the competitive advantage in evaluating deep myometrial invasion compared with the frozen results, generating area under the curve (AUC) values of 0.751 vs. 0.834 ( P = 0.0629, Z = 1.860). The AUC of tumor maximum diameter, simple tumor volume, and whole-lesion MR volume in predicting deep myometrial invasion was 63.8%, 67.6%, and 75.1%, respectively. Conclusion: Whole-lesion MR volume is a good diagnostic tool for prediction of deep myometrial invasion, lymph node metastasis, and lymphovascular invasion. MR volumetry could reflect the aggressiveness of endometrioid adenocarcinoma more accurately than traditional lesion measurements.