Multifrequency MR Elastography for Tumor Stiffness Outperforms Conventional Imaging and Clinical Variables in Determining Lymphovascular Space Invasion in Endometrial Cancer

Take-Away Points

• ■ Major Focus: To determine the potential role of MR elastography in predicting lymphovascular space invasion (LVSI) in individuals with endometrial cancer.

• ■ Key Results: Tumor stiffness (represented by shear-wave speed) was significantly higher in the LVSI-positive group compared with the LVSI-negative group. A combined model based on volumetric imaging data, serum cancer antigen 125 (CA125) levels, and tumor stiffness attained a negative predictive value of 96% in determining LVSI status.

• ■ Impact: Reliable early identification of negative LVSI status in patients with endometrial cancer can avoid unnecessary pelvic and para-aortic lymph node dissections, reducing the risk of associated complications such as peripheral lymphedema. This imaging technology may enable better allocation of younger patients to fertility-sparing treatments for endometrial cancer.

Tumor stiffness, as measured by MR elastography, reflects the mechanical properties of the tumor microenvironment, including cellular composition and extracellular matrix density. Increased stiffness often results from extracellular matrix remodeling, such as higher collagen deposition and cross-linking, driven by interactions between tumor and stromal cells. This altered environment is associated with more aggressive tumor behavior, including enhanced proliferation, invasion, and metastasis. Clinically, higher tumor stiffness correlates with advanced disease stages and poorer prognosis. By noninvasively measuring these mechanical properties, MR elastography may aid in identifying aggressive lesions and provide insights regarding treatment planning.

Long et al conducted a prospective single-center study to assess the difference in viscoelasticity of endometrial tumors larger than 1 cm with and without LVSI determined by postoperative histopathologic studies. Tumor stiffness was significantly higher in the LVSI-positive group (median, 4.1 msec [IQR, 3.2–4.6]) compared with the LVSI-negative group (2.2 msec [IQR, 2.0–2.8]; P < .001). The interobserver intraclass correlation coefficient for tumor stiffness measurement was 0.97 (95% CI: 0.97, 0.99), and Bland-Altman analysis showed good reproducibility. Multivariable analysis identified CA125 level (odds ratio [OR], 1.02 [95% CI: 1.01, 1.04]; P = .009), tumor volume (OR, 1.01 [95% CI: 1.00, 1.01]; P = .04), and tumor stiffness (OR, 9.06 [95% CI: 3.64, 22.53]; P < .001) as independent predictors of LVSI. A clinical-radiologic model incorporating tumor volume and CA125 levels achieved an area under the receiver operating characteristic curve (AUC) of 0.77 (95% CI: 0.68, 0.85). A combined model including the clinical-radiologic variables and tumor stiffness achieved a significantly higher AUC of 0.93 (95% CI: 0.86, 0.97; P = .003) and a negative predictive value of 96%.

The promising results of this study should be viewed considering limitations such as the two-dimensional region of interest–based tumor stiffness measurements, small sample size, and exclusion of lesions smaller than 1 cm. Nonetheless, this study by Long et al adds to a growing body of evidence suggesting that MR elastography–based quantitative biomarkers have promise in characterizing the tumor microenvironment as proxy indicators of tumor invasiveness and the clinical trajectory of patients with endometrial cancer.

Highlighted Article

• Long L, Liu M, Deng X, et al. Tumor Stiffness Measurement at Multifrequency MR Elastography to Predict Lymphovascular Space Invasion in Endometrial Cancer. Radiology 2024;311(3):e232242. doi: 10.1148/radiol.232242