Evaluation of Amide Proton Transfer-Weighted Imaging for Risk Factors in Stage I Endometrial Cancer: A Comparison With Diffusion-Weighted Imaging and Diffusion Kurtosis Imaging

Abstract

BackgroundEndometrial cancer (EC) is one of the most common gynecologic malignancies in clinical practice. This study aimed to compare the value of diffusion-weighted imaging (DWI), diffusion kurtosis imaging (DKI), and amide proton transfer-weighted imaging (APTWI) in the assessment of risk stratification factors for stage I EC including histological subtype, grade, stage, and lymphovascular space invasion (LVSI).MethodsA total of 72 patients with stage I EC underwent pelvic MRI. The apparent diffusion coefficient (ADC), mean diffusivity (MD), mean kurtosis (MK), and magnetization transfer ratio asymmetry (MTRasym at 3.5 ppm) were calculated and compared in risk groups with the Mann–Whitney U test or independent samples t-test. Spearman’s rank correlation was applied to depict the correlation of each parameter with risk stratification. The diagnostic efficacy was evaluated with receiver operating characteristic (ROC) curve analysis and compared using the DeLong test. A multivariate logistic regression was conducted to explore the optimal model for risk prediction.ResultsThere were significantly greater MTRasym (3.5 ppm) and MK and significantly lower ADC and MD in the non-adenocarcinoma, stage IB, LVSI-positive, high-grade, and non-low-risk groups (all p < 0.05). The MK and MTRasym (3.5 ppm) were moderately positively correlated with risk stratification as assessed by the European Society for Medical Oncology (EMSO) clinical practice guidelines (r = 0.640 and 0.502, respectively), while ADC and MD were mildly negatively correlated with risk stratification (r = −0.358 and −0.438, respectively). MTRasym (3.5 ppm), MD, and MK were identified as independent risk predictors in stage I EC, and optimal predictive performance was obtained with their combinations (AUC = 0.906, sensitivity = 70.97%, specificity = 92.68%). The results of the validation model were consistent with the above results, and the calibration curve showed good accuracy and consistency.ConclusionsAlthough similar performance was obtained with each individual parameter of APTWI, DWI, and DKI for the noninvasive assessment of aggressive behavior in stage I EC, the combination of MD, MK, and MTRasym (3.5 ppm) provided improved predictive power for non-low-risk stage I EC and may serve as a superior imaging marker.

Diagnostic accuracy of MRI for assessing lymphovascular space invasion in endometrial carcinoma: a meta-analysis

“…We also performed a sensitivity analysis, which showed that the results were more stable after excluding study 6 (23) and study 9 (19). Forest plots after removing study 6 (23) are shown in Fig.…”

“…For example, the sensitivity and specificity of the short axis of the tumor for distinguishing LVSI are 56.5% and 83.5%, respectively, and tumor volume ratio (TVR) allows LVSI to be predicted with a sensitivity of 77% and specificity of 83%. Diffusion-weighted imaging (DWI)-derived parameters, functional imaging features, and other new techniques are also used to predict LVSI status (16)(17)(18)(19). Recently, the application value of radiomics analysis based on MRI for assessing the status of LVSI in EC was explored, with sensitivity in the range of 66.7%-92.8% and specificity in the range of 59%-78.6% in different models (20)(21)(22)(23)(24)(25).…”

Whole‐tumor apparent diffusion coefficient histogram analysis for preoperative risk stratification in endometrial endometrioid adenocarcinoma

“…When combined with tumor volume, the diagnostic performance significantly improved, with an AUC of 0.780. By using multi-b-value DWI (intravoxel incoherent motion and diffusion kurtosis imaging), Meng et al 13 and Jin et al 25 found that mean diffusion (MD), mean kurtosis (MK), and true diffusion coefficient (D t ) outperformed ADC in identifying the low-risk group in early-stage EC, while Zhang et al 14 found the opposite, with ADC yielding a larger AUC than D t and perfusion fraction (f). Nevertheless, all of these studies reported improvement with a combined diagnosis.…”

“…and the combination of amide proton transfer-weighted imaging (MTRasym [3.5 ppm]), MK, and D t or MD achieved AUCs of 0.906 and 0.958, respectively,13,25 Chen et al 23developed a combined model incorporating radiomic features and tumor size, achieving an AUC of 0.955 in the training cohort and 0.889 in the validation cohort. Unfortunately, most of these studies were based on the three-category risk stratification with a relatively small sample size (n ≤ 102), including non-EEAs, and limited to the stage I ECs.…”

The value of multimodal functional magnetic resonance imaging in differentiating p53abn from p53wt endometrial carcinoma

“…Previous data suggested that, due to over-expressed mobile protein and peptide concentrations, APTw can assess the pathological grade of malignant tissue. For example, a previous study found that histologically type II, high-grade, high International Federation of Gynecology and Obstetrics (FIGO) stage and lymphovascular space invasion (LVSI)-positive EC have higher APT values (14,15). In addition, Li et al (16) showed that the APT value was moderately positively correlated with the expression index of the nuclear antigen Ki-67 associated with EC cell proliferation.…”