Prediction of HIF-1α expression in endometrial carcinoma by enhanced T weighted angiography and dynamic contrast-enhanced magnetic resonance imaging.

PURPOSE

To explore the value of quantitative imaging parameters by enhanced T weighted angiography (ESWAN) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for evaluating the expression of Hypoxia-inducible factor-1α (HIF-1α) in endometrial carcinoma (EC).

METHODS

Data from 122 patients with EC confirmed by clinical pathology were retrospectively analyzed. According to the number of positive cells stained with HIF-1α by immunohistochemistry, patients were divided into two groups: 65 cases with high expression of HIF-1α and 57 cases with low expression of HIF-1α. Clinical data included age, FIGO stage, menopausal status, abnormal uterine bleeding, and pathological type. All patients underwent preoperative 1.5T MRI scans, including ESWAN and DCE-MRI. The amplitude, phase, and R values derived from ESWAN and the volume transfer constant (K), rate constant (K), and extravascular volume fraction (V) values derived from DCE-MRI were measured by two observers, respectively. The intra-class correlation coefficient (ICC) was used to assess the measurement of reproducibility across observers, and the differences in imaging parameters between the two groups were compared using the independent sample t-test or Mann-Whitney U-test. Binary logistic regression analysis was used to find independent risk factors for HIF-1α expression. The efficacy of selected imaging parameters for predicting HIF-1α expression was assessed using receiver operating characteristic (ROC) curves, and the Delong test was used to compare the area under ROC curves (AUC).

RESULTS

The consistency between the two observers was good (ICC>0.75). The R , K, and K values of the HIF-1α high expression group were higher than those of the HIF-1α low expression group (14.59 ± 4.06 . 11.99 ± 2.84 Hz, 0.45 ± 0.18 . 0.36 ± 0.14/min, and 2.17 ± 1.10 . 1.54 ± 0.80/min) (< 0.001, = 0.011, and =0.001). Binary logistic regression analysis revealed that R and K values were independent risk factors for HIF-1α expression. The AUC values of R , K, and their combination for prediction of HIF-1α expression were 0.697, 0.677, and 0.781, respectively. The diagnostic efficacy was significantly improved with combination of R and K.

CONCLUSIONS

Quantitative parameters by ESWAN and DCE-MRI showed significant differences between EC patients with low and high expression of HIF-1α, and the combination of ESWAN and DCE-MRI improves the efficacy in prediction of HIF-1α expression in EC, which has an excellent clinical application prospect.