Rapid-Steady-State-T1 signal modeling during contrast agent extravasation: toward tumor blood volume quantification without requiring the arterial input function
Abstract
PURPOSE: This study demonstrates how to quantify the tumor blood volume fraction (BVf) using the dynamic Rapid-Steady-State-T1 (RSST1 )-MRI method despite contrast agent (CA) leakage and without arterial input function (AIF) determination. METHODS: For vasculature impermeable to CAs, the BVf is directly quantified from the RSST1 signal amplitude. In case of CA extravasation, we propose a two-compartment model to describe the dynamic RSST1 signal increase. We applied the mathematical model in a pilot-study on a RG2-glioma model to compare extravasation of two Gd-based CAs. The BVf quantification using the mathematical model in a C6-glioma model (n = 8) with the clinical CA Gd-DOTA was validated using a ΔR2 *-steady-state MRI method with an USPIO and by immunohistochemical staining of perfused vessels labeled with Hoechst-33342 dye in the same rats. RESULTS: BVf in tumor and in healthy brain tissues (0.034 ± 0.005 and 0.026 ± 0.004, respectively) derived from the dynamic RSST1 signal were confirmed by ΔR2 *-steady-state MRI (0.036 ± 0.003 and 0.027 ± 0.002, respectively, correlation coefficient rS = 0.74) and by histology (0.036 ± 0.003 and 0.025 ± 0.004 respectively, rS = 0.87). CONCLUSION: Straightforward tumor BVf quantification without AIF determination is demonstrated in presence of CA leakage. The method will facilitate angiogenesis assessment in longitudinal neuro-oncologic studies in particular when monitoring the response to antiangiogenic therapies.