Differentiation of uterine low-grade endometrial stromal sarcoma from rare leiomyoma variants by magnetic resonance imaging

• Published in: Scientific reports Vol. 11; no. 1; p. 19124
• Main Authors: Himoto, Yuki, Kido, Aki, Sakata, Akihiko, Moribata, Yusaku, Kurata, Yasuhisa, Suzuki, Ayako, Matsumura, Noriomi, Shitano, Fuki, Kawahara, Seiya, Kubo, Shigeto, Umeoka, Shigeaki, Minamiguchi, Sachiko, Mandai, Masaki
• Format: Journal Article
• Language: English
• Published: England Nature Publishing Group 27.09.2021; Nature Publishing Group UK; Nature Portfolio
• Subjects: Adult; Aged; Diagnosis, Differential; Diffusion coefficient; Diffusion Magnetic Resonance Imaging; Endometrial Neoplasms - diagnosis; Endometrial Neoplasms - pathology; Endometrium; Endometrium - diagnostic imaging; Endometrium - pathology; Feasibility Studies; Female; Fibroids; Humans; Leiomyoma - diagnosis; Leiomyoma - pathology; Magnetic resonance imaging; Middle Aged; Retrospective Studies; ROC Curve; Sarcoma; Sarcoma, Endometrial Stromal - diagnosis; Sarcoma, Endometrial Stromal - pathology; Uterine cancer; Uterus; Young Adult
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-021-98473-z

The purpose of this study is to evaluate utility of MRI in differentiation of uterine low-grade endometrial stromal sarcoma (LGESS) from rare leiomyoma variants. This multi-center retrospective study included consecutive 25 patients with uterine LGESS and 42 patients with rare leiomyoma variants who had pretreatment MRI. Two radiologists (R1/R2) independently evaluated MRI features, which were analyzed statistically using Fisher's exact test or Student's t-test. Subsequently, using a five-point Likert scale, the two radiologists evaluated the diagnostic performance of a pre-defined MRI system using features reported as characteristics of LGESS in previous case series: uterine tumor with high signal intensity (SI) on diffusion-weighted images and with either worm-like nodular extension, intra-tumoral low SI bands, or low SI rim on T2-weighted images. Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of the two readers' Likert scales were analyzed. Intra-tumoral low SI bands (p < 0.001), cystic/necrotic change (p ≤ 0.02), absence of speckled appearance (p < 0.001) on T2-weighted images, and a low apparent diffusion coefficient value (p ≤ 0.02) were significantly associated with LGESS. The pre-defined MRI system showed very good diagnostic performance: AUC 0.86/0.89, sensitivity 0.95/0.95, and specificity 0.67/0.69 for R1/R2. MRI can be useful to differentiate uterine LGESS from rare leiomyoma variants.