Correlation between SWE parameters and histopathological features and immunohistochemical biomarkers in invasive breast cancer.

Abstract

Objectives: Shear wave elastography (SWE) is a novel quantitative elastography technique that can assess the hardness of different tissues. This study introduces a novel shear wave parameter-frequency of mass characteristic (f_mass)-and investigates its correlation, along with other shear wave parameters, with the histopathological features and immunohistochemical (IHC) biomarkers of invasive breast cancer (IBC). The study aims to explore whether SWE can provide useful information for IBC treatment and prognosis.

Methods: With the pathological results as the gold standard, 258 malignant breast lesions were collected, and all patients underwent conventional ultrasound and SWE examinations. The SWE parameters [maximum elastic value (E_max), minimum elastic value (E_min), mean elastic value (E_mean), standard deviation of elastic value of the whole lesion (E_sd)] and f_mass] in the transverse and longitudinal orthogonal sections were measured, and their correlations with the prognostic factors of IBC [including tumor diameters, axillary lymph node (ALN) metastasis, lymphatic vessel invasion (LVI), calcification, histological type, histological grade, and IHC biomarkers (ER, PR, HER-2, Ki-67), and molecular subtypes] were analyzed. The correlations between the SWE parameters of the transverse and longitudinal sections of the tumors with different prognostic factors and the above indicators were analyzed. At the same time, the receiver operating characteristic (ROC) curve was used to analyze the efficacy of f_mass in predicting ER and PR expression.

Results: E_mean, E_max, E_sd, and f_mass were correlated with tumor diameters; E_mean, E_max and E_sd were correlated with histological types and histological grades. E_max and E_sd were correlated with ALN metastasis, LVI and pathological types. In the IHC biomarker-labeled masses, f_mass was correlated with ER and PR (both P<0.05), and E_mean, E_max, and E_sd were correlated with HER-2 and Ki-67 (all P<0.05). E_mean, E_max, and f_mass were all correlated with breast cancer subtypes (all P<0.05), and E_mean and E_max were higher in Luminal B [HER-2(+)] breast cancer, while f_mass was lower in HER-2(+) and triple-negative breast cancer. Among the statistically significant prognostic factors, the P values of the transverse sections of the masses were all less than or equal to those of the longitudinal sections. The AUC of f_mass in the transverse sections of the masses for predicting ER and PR expression were 0.73 (95% CI 0.65 to 0.80) and 0.67 (95% CI 0.60 to 0.74), respectively, with the optimal cut-off values being 76.50 and 60.66, the sensitivities being 72.45% and 81.98%, the specificities being 66.13% and 45.35%, and the accuracies being 70.93% and 69.77%, respectively. The AUC of f_mass in the longitudinal sections of the masses for predicting ER and PR expression were 0.74 (95% CI 0.67 to 0.81) and 0.65 (95% CI 0.58 to 0.72), respectively, with the optimal cut-off values being 131.8 and 137.5, the sensitivities being 69.90% and 66.28%, the specificities being 72.58% and 60.47%, and the accuracies being 70.54% and 64.34%, respectively. The f_mass in the transverse sections of the masses was more statistically significant.

Conclusions: The poor prognosis factors of IBC are related to high E_mean, E_min, E_max, E_sd, and low f_mass. The f_mass can predict the expression of ER and PR, and the transverse cut data are more meaningful. SWE is helpful for predicting the invasiveness of IBC.