Preoperative risk stratification in women with endometrial cancer: A comparison of contrast-enhanced MR imaging and diffusion-weighted MR imaging

Preoperative risk stratification of early-stage endometrial cancer assessed by multimodal magnetic resonance functional imaging

Endometrial cancer is a common disease in women. Stratifying the risk of early-stage endometrial cancer can aid in personalized treatment for patients. Risk stratification is primarily based on tumor grade, histological type, lymph node metastasis, and depth of myometrial invasion. Multimodal magnetic resonance functional imaging (including DCE-MRI, DWI, IVIM, DTI, DKI) has significant value in assessing the extent of myometrial and cervical infiltration, extrauterine involvement range, determining lymph node metastasis and tumor size. This article provides a brief overview of these techniques.

Molecular classification improves preoperative risk assessment of endometrial cancer

OBJECTIVE

We aimed to evaluate the performance of endometrial cancer (EC) molecular classification in predicting extrauterine disease after primary surgery alone and in combination with other clinical data available in preoperative setting.

METHODS

Retrospective single-center observational study including patients with endometrial adenocarcinoma treated with primary surgery between December 1994 and May 2022. Molecular profiling was performed using immunohistochemistry of p53, MLH1, PMS2, MSH2 and MSH6; and KASP genotyping of the 6 most common mutations of POLE gene. Clinical, pathological and imaging information was reviewed. Logistic regression, regression trees and random forest classification techniques (CART) were performed.

RESULTS

We enrolled 658 patients, 47 with POLEmut (7.1%), 234 with MMRd (35.6%), 95 with p53abn (14.4%) and 282 with NSMP (42.8%) tumors. Advanced stage after primary surgery (III-IV FIGO 2009) was diagnosed in 11.7% of patients, p53abn tumors showed increased extrauterine spread (34.1%) and nodal involvement (30.1%) (p < .001). In multivariate analysis, only p53abn subgroup (aOR = 16.0, CI95% = 1.5-165.1) and radiological suspicion of extrauterine disease (aOR = 24.2, CI95% = 12.2-48.2) independently predicted the finding of extrauterine disease after primary surgery. In patients with preoperative uterine-confined disease, deep myometrial and cervical involvement in radiological assessment and p53abn molecular subtype were the best variables to identify patients at-risk of occult extrauterine disease after the staging surgery.

CONCLUSION

EC molecular classification is more accurate than histotype or grade in preoperative biopsy to predict advanced disease, and together with imaging tests are the most reliable preoperative information. This work provides an initial framework for using molecular information preoperatively to tailor surgical treatment.

Evaluating the image quality and local tumor invasion of uterine cancer by MUSE DWI with RPG

Diffusion-weighted imaging (DWI) is widely utilized for evaluating uterine diseases. However, the prevalent technique, single-shot echo planar imaging (ssEPI), is hindered by notable image distortion and low spatial resolution. Therefore, optimizing uterine DWI sequences is vital for improving image quality. To investigate the efficacy of multiplexed sensitivity encoding (MUSE) combined with reverse polarity gradient (RPG) in enhancing uterine DWI quality and assessing local invasion in endometrial and cervical cancer, we included 149 patients. Each patient underwent DWI of the uterus using ssEPI, MUSE, and RPG-MUSE techniques. We compared these three sequences regarding image quality, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), geometric distortion rate (GDR), ADC values, accuracy in determining the extent of cancer invasion, and the Area Under the Curve (AUC) for identifying endometrial cancer and benign endometrial lesions using ADC values. The results indicated that RPG-MUSE DWI had less artifacts than MUSE and ssEPI (P < 0.05). Lesions were more apparent in MUSE and RPG-MUSE sequences compared to ssEPI (P < 0.05), with RPG-MUSE providing clearer lesion edges (P < 0.05). Additionally, RPG-MUSE DWI demonstrated higher SNR and CNR than ssEPI and MUSE (P < 0.05), along with a lower GDR (P < 0.05). The ADC values did not show significant differences among the three sequences (P > 0.05). Furthermore, the AUC of the ROC for detecting endometrial cancer and benign endometrial lesions using ADC values showed no significant differences across the sequences (P = 0.7609, 0.7186, and 0.8706, respectively). When combining each DWI sequence with T2WI for FIGO staging, RPG-MUSE and MUSE exhibited better alignment with pathology findings compared to ssEPI (P < 0.05). Overall, RPG-MUSE DWI showed fewer artifacts, higher SNR and CNR, reduced geometric distortion, and clearer lesion visualization compared to ssEPI and MUSE, leading to a more precise assessment of endometrial and cervical cancer invasion extent.

A prediction model based on deep learning and radiomics features of DWI for the assessment of microsatellite instability in endometrial cancer

BACKGROUND

To explore the efficacy of a prediction model based on diffusion-weighted imaging (DWI) features extracted from deep learning (DL) and radiomics combined with clinical parameters and apparent diffusion coefficient (ADC) values to identify microsatellite instability (MSI) in endometrial cancer (EC).

METHODS

This study included a cohort of 116 patients with EC, who were subsequently divided into training (n = 81) and test (n = 35) sets. From DWI, conventional radiomics features and convolutional neural network-based DL features were extracted. Random forest (RF) and logistic regression were adopted as classifiers. DL features, radiomics features, clinical variables, ADC values, and their combinations were applied to establish DL, radiomics, clinical, ADC, and combined models, respectively. The predictive performance was evaluated through the area under the receiver operating characteristic curve (AUC), total integrated discrimination index (IDI), net reclassification index (NRI), calibration curves, and decision curve analysis (DCA).

RESULTS

The optimal predictive model, based on an RF classifier, comprised four DL features, three radiomics features, two clinical variables, and an ADC value. In the training and test sets, this model exhibited AUC values of 0.989 (95% CI: 0.935-1.000) and 0.885 (95% CI: 0.731-0.967), respectively, demonstrating different degrees of improvement compared with the clinical, DL, radiomics, and ADC models (AUC-training = 0.671, 0.873, 0.833, and 0.814, AUC-test = 0.685, 0.783, 0.708, and 0.713, respectively). The NRI and IDI analyses revealed that the combined model resulted in improved risk reclassification of the MSI status compared to the clinical, radiomics, DL, and ADC models. The calibration curves and DCA indicated good consistency and clinical utility of this model, respectively.

CONCLUSIONS

The predictive model based on DWI features extracted from DL and radiomics combined with clinical parameters and ADC values could effectively assess the MSI status in EC.

Utility of the Diffusion Weighted Sequence in Gynecological Imaging: Review Article

Simple Summary

Diffusion weighted imaging (DWI) is a magnetic resonance imaging sequence with diverse clinical applications in malignant and nonmalignant gynecological conditions. It provides vital supplemental information in the diagnosis and management of various gynecological conditions. Radiologists should be aware of fundamental concepts, clinical applications and pitfalls of DWI. Additionally we briefly discuss potential scope of newer advanced techniques based on DWI including diffusion tensor imaging and diffusion-weighted whole-body imaging with background signal suppression.

Abstract

Functional imaging with diffusion-weighted imaging (DWI) is a complementary tool to conventional diagnostic magnetic resonance imaging sequences. It is being increasingly investigated to predict tumor response and assess tumor recurrence. We elucidate the specific technical modifications of DWI preferred for gynecological imaging, including the different b-values and planes for image acquisition. Additionally, we discuss the problems and potential pitfalls encountered during DWI interpretation and ways to overcome them. DWI has a wide range of clinical applications in malignant and non-malignant gynecological conditions. It provides supplemental information helpful in diagnosing and managing tubo-ovarian abscess, uterine fibroids, endometriosis, adnexal torsion, and dermoid. Similarly, DWI has diverse applications in gynecological oncology in diagnosis, staging, detection of recurrent disease, and tumor response assessment. Quantitative evaluation with apparent diffusion coefficient (ADC) measurement is being increasingly evaluated for correlation with various tumor parameters in managing gynecological malignancies aiding in preoperative treatment planning. Newer advanced DWI techniques of diffusion tensor imaging (DTI) and whole body DWI with background suppression (DWIBS) and their potential uses in pelvic nerve mapping, preoperative planning, and fertility-preserving surgeries are briefly discussed.