[Preliminary study of semi-quantitative and quantitative dynamic contrast-enhanced MRI in evaluating the response to concurrent chemoradiotherapy in patients with non-small cell lung cancer].

OBJECTIVE: To investigate the capability of semi-quantitative and quantitative parameters of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to predict the response to concurrent chemoradiotherapy( CCRT) in patients with non-small cell lung cancer (NSCLC). METHODS: A total of 24 patients with stage III or IIIB NSCLC, who underwent 3.0T DCE-MRI before CCRT, were enrolled in this study. Semi-quantitative and quantitative parameters were calculated by Funtool and Omnikinetics software. The relationship between these obtained parameters and tumor response was evaluated by Spearmen' s correlation analysis. The patients were classified into two groups according to the tumor regression rate after treatment, as response group (group A) and non-response group ( group B). Mann-Whitney U test was used to compare the parameters of responders and non-responders. The value of the parameters on predicting response was calculated by receiver operating characteristic curve (ROC). RESULTS: The tumor regression rate after treatment was negatively correlated with time to peak (TTP) and the extravascular-extracellular volume fraction (Ve), and was positively correlated with signal enhancement ratio (SERmax) and volume transfer constant (Ktrans) (P < 0.05 for all). Statistical significant differences were found between group A and group B both in semi-quantitative and quantitative parameters (P < 0.05). Group A had a lower TTP value [(34.66 ± 16.37) s vs. (44.09 ± 17.41) s] and Ve value [(0.19 ± 0.03) vs. (0.25 ± 0.05)] than group B, whereas group A had a higher SERmax [(166.50 ± 44.95)% vs. (113.57 ± 46.62)%] and Ktrans [(0.41 ± 0.17) min(-1) vs. (0.28 ± 0.12) min(-1)] than group B (P < 0.05 for all). The ROC analysis indicated that when setting the threshold of Ve on ≤ 0.21 for predicting response, the specificity, sensitivity and accuracy were 85.7%, 80.0% and 83.3%, respectively, with an area under curve of 0.875 (P < 0.001). CONCLUSIONS: Both the semi-quantitative and quantitative DCE-MRI parameters are helpful for predicting the response after CCRT of NSCLC. Quantitative parameters seem to be more meaningful than semi-quantitative parameters.

[Hepatic epithelioid hemangioendothelioma: analysis of the correlation of imaging characteristics with histopathological findings].

OBJECTIVE: To analyze the imaging characteristics of hepatic epithelioid hemangioendothelioma (EHE) and their correlation with histopathological findings. METHODS: CT and MRI imaging and histopathological characteristics of five patients with hepatic EHE were retrospectively reviewed and the correlation of their imaging characteristics with pathological findings was analyzed. RESULTS: A total of 92 lesions were found in the 5 patients, all presenting with multiple nodules. All the 92 lesions were located within a 2-cm zone heneath the hepatic capsule, i.e., the shortest distance from the horder of lesions to the hepatic capsule. 28 of the 92 lesions showed the "capsular retraction" sign. 36 lesions were found in three patients receiving MRI. 77.8% of the 36 lesions demonstrated the "halo" sign on a fat-suppression T2- weighted image, while 91.7% after contrast enhancement. A peripheral dark rim was found in 91.7% of the lesions on a fat-suppression T2-weighted image. In addition, 36.1% of the lesions showed slight internal or edge enhancement at the hepatohiliary phase. In the two patients receiving CT examination, 7 of 56 lesions demonstrated the "halo" sign. CONCLUSIONS: Hepatic EHE may manifest as nodular lesions with predilection of peripheral subcapsular growth and nodular confluence, together with "halo" sign and " capsular retraction". The peripheral dark rim on a fat-suppression T2-weighted image and slight enhancement at the hepatobiliary phase can help to improve the accuracy of diagnosis and differential diagnosis of this hepatic tumor. MRI is superior to CT imaging to denict their intra-lesional characteristics.