Purpose

Gadoxetic acid (Gd-EOB-DTPA) is a dual-function contrast agent for liver MRI. It acts as an extracellular contrast agent in the early phase enabling tumor vascularity evaluation and at later times enhances hepatocytes.¹ Dynamic contrast-enhanced (DCE) MRI is a useful to assess tumor vascularity, and generally uses extracellular contrast agent such as Gd-DTPA.² We hypothesized that the liver malignant tumors might show similar hemodynamic change in the early vascular phase between DCE-MRI using Gd-EOB-DTPA and extracellular contrast agent, thus Gd-EOB-DTPA could be adopted as a contrast agent for tumor vascularity evaluation in DCE-MRI. We aim to test our hypothesis in a rat orthotopic hepatoma model.

Material and Methods

In 10 Sprague-Dawley (SD) rats, rat hepatoma cells (N1-S1) were engrafted in the left medial lobe of the liver. Ten days after tumor engraftment, DCE-MRI was repeatedly performed with 24 hours interval: first scan using Gd-DTPA for 3 minutes and second scan using Gd-EOB-DTPA for 30 minutes. DCE-MRI was performed at 3.0T clinical machine (Skyra, Siemens) using CAIPIRINHA-VIBE sequence (TR/TE, 4.3/1.5 ms; flip angle, 25°; matrix size, 128x128), enabling high temporal resolution (3 seconds) and spatial resolution (1.0 x 1.0 mm). T1 map was generated with variable flip-angle technique (α=2°, 8°, 15°, 22°, 29°). Regarding contrast agent dose, we injected 0.05 mmol/kg of Gd-EOB-DTPA and 0.1 mmol/kg of Gd-DTPA, because T1 relaxivity of Gd-EOB-DTPA is double, compared with Gd-DTPA. We compared the time-intensity curve (TIC) pattern of the aorta and liver tumor between DCE-MRI using Gd-EOB-DTPA and Gd-DTPA. Using a two-compartment model (Toft model), we calculated kinetic DCE-MRI parameters: blood volume (Vp), extravascular-extracellular space (EES) volume (Ve) and transvascular permeability (Ktrans). We evaluated the agreement of these parameters between Gd-EOB-DTPA and Gd-DTPA using intra-class correlation coefficient. After completion of DCE-MRI, the liver tumors were extracted and immunohistochemistry using anti-OATP-C was performed.

Results

Between DCE-MRI using Gd-EOB-DTPA and Gd-DTPA, TIC patterns were quite different in liver tumor, but were similar in the aorta. When we use Gd-DTPA, the tumor enhancement reached to peak level at 0.81 ± 0.17 minutes after contrast injection then decreased rapidly (Fig. 1a). When we use Gd-EOB-DTPA, the tumor enhancement slowly reached to peak level at 3.66 ± 3.21 minutes and then decreased slowly over 20 minutes (Fig. 1b). These findings imply that Gd-DTPA distributes between blood vessels and EES, working as a extracellular contrast agent, whereas Gd-EOB-DTPA distribute blood vessels, EES, and intracellular space in the hepatoma, working as a hepatobiliary contrast agent. In addition, the Ktrans, Vp, and Ve values were totally different between DCE-MRI using Gd-DTPA and Gd-EOB-DTPA (r=0.103, intraclass correlation coefficient). This result suggests that the two-compartment model does not suit for DCE-MRI using Gd-EOB-DTPA in subjects with hepatoma. Immunohistochemistry showed weak positivity for OATP1, but inconsistent across the tumors.

Discussion and conclusion

Gd-EOB-DTPA has unique mechanism, because its uptake in hepatocytes and/or hepatoma cells occurs via the OATP receptors, and its biliary excretion via the multidrug resistance-associated proteins (MRP). Therefore, the conventional two-compartment model does not fit DCE-MRI using Gd-EOB-DTPA in subjects with hepatoma which may express OATP receptors, warranting further researches.