Endometrial Carcinoma: MR Imaging–based Texture Model for Preoperative Risk Stratification—A Preliminary Analysis

Abstract

Purpose To evaluate the associations among mathematical modeling with the use of magnetic resonance (MR) imaging-based texture features and deep myometrial invasion (DMI), lymphovascular space invasion (LVSI), and histologic high-grade endometrial carcinoma. Materials and Methods Institutional review board approval was obtained for this retrospective study. This study included 137 women with endometrial carcinomas measuring greater than 1 cm in maximal diameter who underwent 1.5-T MR imaging before hysterectomy between January 2011 and December 2015. Texture analysis was performed with commercial research software with manual delineation of a region of interest around the tumor on MR images (T2-weighted, diffusion-weighted, and dynamic contrast material-enhanced images and apparent diffusion coefficient maps). Areas under the receiver operating characteristic curve and diagnostic performance of random forest models determined by using a subset of the most relevant texture features were estimated and compared with those of independent and blinded visual assessments by three subspecialty radiologists. Results A total of 180 texture features were extracted and ultimately limited to 11 features for DMI, 12 for LVSI, and 16 for high-grade tumor for random forest modeling. With random forest models, areas under the receiver operating characteristic curve, sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were estimated at 0.84, 79.3%, 82.3%, 81.0%, 76.7%, and 84.4% for DMI; 0.80, 80.9%, 72.5%, 76.6%, 74.3%, and 79.4% for LVSI; and 0.83, 81.0%, 76.8%, 78.1%, 60.7%, and 90.1% for high-grade tumor, respectively. Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of visual assessment for DMI were 84.5%, 82.3%, 83.2%, 77.7%, and 87.8% (reader 3). Conclusion The mathematical models that incorporated MR imaging-based texture features were associated with the presence of DMI, LVSI, and high-grade tumor and achieved equivalent accuracy to that of subspecialty radiologists for assessment of DMI in endometrial cancers larger than 1 cm. However, these preliminary results must be interpreted with caution until they are validated with an independent data set, because the small sample size relative to the number of features extracted may have resulted in overfitting of the models. RSNA, 2017 Online supplemental material is available for this article.

Preoperative tumor texture analysis on MRI predicts high‐risk disease and reduced survival in endometrial cancer

“…We did not see a clear pattern in which certain filter levels had a better predictive or prognostic performance than others. This observation complies with previous TexRad‐based publications, including the previous endometrial cancer study, in which no specific filter level has proven general superiority . Our findings also illustrate some of the diversity and complexity of radiomics; characterizing cancers of different origin, type, subtype, and grade, using images originating from different modalities and protocols and employing image analyses being inherently software‐ and operator‐dependent.…”

“…Similarly, in our study the top‐ranked single parameter, ADC_Entropy6, predicted DMI with an AU‐ROC of 0.81. For prediction of high‐grade tumor, a subset of 16 texture parameters jointly yielded an AU‐ROC of 0.83, whereas the top‐ranked single parameter in our study, T 1 c_MPP4, yielded an AU‐ROC of 0.66. However, in spite of being based on somewhat different approaches, both studies suggest a very promising role of MRI texture analysis for better preoperative risk assessment in endometrial cancer.…”

“…Interestingly, a recent preliminary MRI study using TexRAD in endometrial cancer found that a subset of 11 texture parameters jointly predicted DMI with an AU‐ROC of 0.84. Similarly, in our study the top‐ranked single parameter, ADC_Entropy6, predicted DMI with an AU‐ROC of 0.81.…”

“…Magnetic resonance imaging (MRI)‐ and computed tomography (CT)‐derived texture parameters have been proposed as tools for accurate diagnosis, preoperative risk stratification, or assessment of treatment response in several cancer types, eg, in the brain, lung, breast, and prostate . A recent preliminary study also reported MR texture analysis (MRTA) as a useful tool for preoperative risk stratification in endometrial cancer …”

“…9 A recent preliminary study also reported MR texture analysis (MRTA) as a useful tool for preoperative risk stratification in endometrial cancer. 10 Endometrial cancer is the most common gynecologic malignancy in industrialized countries, with an increasing incidence. 11 Treatment strategy and prognosis are traditionally based on the International Federation of Gynecology and Obstetrics (FIGO) stage and histological subtype and grade determined from the hysterectomy specimen.…”