Prediction of histological grade of endometrial cancer by means of MRI

Abstract

Objectives: To evaluate the ability of MRI in predicting histological grade of endometrial cancer (EC).

Methods: IRB-approved retrospective study; requirement for informed consent was waived. 90 patients with histologically proven EC who underwent preoperative MRI and surgery at our Institution between Sept2011 and Nov2016 were included. Myometrial invasion (</>50%) was assessed. Neoplasm and uterus volumes were estimated according to the ellipsoid formula; neoplasm/uterus volume ratio (N/U) was calculated. ADC maps were generated and histogram analysis was performed using commercially available software. MRI parameters were compared with the definitive histological grade (G1 = 28 patients, G2 = 29, G3 = 33) using ANOVA and Tukey-Kramer tests.

Results: Deep myometrial invasion was significantly more frequent in G2-G3 lesions than in G1 ones (p < 0,005). N/U ratio was significantly higher for high-grade neoplasms (mean 0,08 for G1, 0,16 for G2 and 0,21 in G3; P = 0,002 for G1 vs. G2-G3); a cut off value of 0,13 enabled to distinguish G1 from G2-G3 lesions with 50% sensibility and 89% specificity. ADC values didn't show any statistically significant correlation with tumour grade.

Conclusions: N/U ratio >0.13 and deep myometrial invasion are significantly correlated with high grade EC, whereas ADC values are not useful for predicting EC grade.

Keywords: Diffusion magnetic resonance imaging; Endometrial neoplasms; Magnetic resonance imaging; Neoplasm grading; Uterus.