SPIO-labeled 90 Y microspheres permit accurate quantification of macroscopic intrahepatic biodistribution
Weiguo Li,Zhuoli Zhang,Yang Guo,Jodi Nicolai,Reed A. Omary,Andrew C. Larson
2012-01-01
Abstract:Target audience Interventional radiologist for liver cancer therapy. Purpose Radioembolization with Yttrium-90 (Y) microspheres is a promising form of intra-arterial therapy for liver tumors [1]. Visualization of Y microsphere biodistribution using conventional radiologic modalities is challenging [2]. Labeling Y microspheres with SPIOs offers the potential to use MRI for visualization. Feasibility studies demonstrated that SPIO-labeled radioembolization microspheres permit qualitative visualization of microsphere biodistribution following transcatheter delivery to the liver [3]. However, optimization of the amount of SPIO material included within these glass microspheres may be critical. The objective of this study was to characterize the relationship between Y microsphere SPIO content and associated MR imaging properties and to validate that quantitative R2* measurements permit in vivo quantification of SPIO-labeled microsphere biodistribution. Methods Phantom: We constructed 32 agarose gel phantoms for each of four different microsphere compositions (2%, 5%, 10%, and 20% SPIO (%by-mass)). Agarose gel was placed within 15 mm ID NMR tubes along with 0, 0.25, 0.5, 1.0, 2.0, 4.0, 8.0, or 16.0 mg/ml samples of SPIO-labeled microspheres for each composition of Y microsphere to form a homogeneous distribution of the microspheres. Rat model: All the experiments are approved by the Institutional Animal Care and Use Committee. Male Sprague-Dawley rats weighing 200–350 g were used. The rats were anaesthetized by inhalation of 2% vaporized isoflurane. Each rat was catheterized through hepatic portal vein with a rat portal vein catheter (Charles River, Wilmington, MA). After catheterization, MR imaging was performed before and after administration of microspheres. 5 mg samples of microspheres containing 2 % SPIO were administered with two injections and two saline flushes for each injection through the catheter. After MRI, rats were euthanized, and livers were harvested and fixed using 10% buffered formalin for histological analysis (hematoxylin and eosin (H&E) staining). MRI: All studies were performed using a 7.0 T 30 cm bore Bruker ClinScan MRI scanner (Bruker Biospin, Ettlingen, Germany) with a) 75mm Quad Transceiver rat coil, b) isoflurane anesthesia system, body temperature control and monitoring system for vital signs, and c) MRI-compatible small animal gating system (SA Instruments, NY) to permit free-breathing acquisitions for preand post-procedural MRI measurements. The phantoms are positioned at the center of magnet. Careful manual shimming was performed before R2* measurement. Quantitative R2* measurements were obtained by using a multiple-gradient-echo (mGRE) sequence with TR/TEs = 200/2.6, 5.7, 8.8, 11.9 ms; 30° flip angle, 1-mm section thickness, 65-mm field of view, 192 × 192 matrix, readout bandwidth of 360 Hz/pixel. For rat imaging, all the scans were synchronized with the respiratory cycle to minimize motion artifacts. Quantitative R2* measurements were performed using mGRE sequence with TR/TEs = 200/2.6, 5.7, 8.8, 11.9 ms; 30° flip angle, FOV = 65 × 60 mm; matrix = 192 × 192; slice thickness = 0.7 mm, readout bandwidth of 360 Hz/pixel. Data Analysis: Image post-processing was performed offline by using Matlab software (The Math Works, Natick, MA). R2* maps were calculated using the nonlinear Levenberg-Marquardt algorithm to fit the monoexponential decay component: STEi = S0 · exp(−R2* · TEi), where STEi is the MR signal intensity at echo time TEi, S0 is the MR signal intensity at echo time 0, and TEi is the echo time. Regions of interest (ROI) that encompassed each concentration of microspheres were drawn on the R2* map obtained in each microsphere composition phantom. Color-coded R2* maps of the rat liver was overlaid on the T2* weighted images to highlight the voxel-wise changes following the intra-hepatic microsphere delivery. Results T2* weighted images of microsphere phantom showed a decreasing signal intensity with increase in microsphere concentration (Fig.1a). For microspheres with 5, 10, and 20% iron compositions, the measured R2*s dropped when concentrations produced R2* values beyond an upper limit threshold (about 750 sec -1, Fig. 1b). A linear relationship was found across all seven concentrations of 2% Y microspheres (blue line Fig.1b). Histological analysis results (Fig.2) verified microsphere deposits in liver vessels. Coronal T2* weighted images with color-coded overlays depicting voxel-wise estimates of intra-hepatic microsphere concentration are shown in Fig. 3. Representative examples are shown preand post-infusion (Figs. 3a, and b) of the microspheres. Discussion and Conclusion The clinical ability to quantify intra-hepatic Y microsphere deposition would provide enormous benefits permitting dose optimization to maximize tumor kill while limiting toxic effects on normal liver tissues. SPIO-labeled Y microspheres offer the potential to use MRI to detect in vivo biodistribution; however, the strong magnetic susceptibility effects of these SPIO-labeled microspheres can rapidly reduce signal below the noise floor complicating rigorous quantification. In current study, given the measured approximately 500 sec R2* value at the concentration of 16 mg/ml for 2% microspheres, higher concentrations of the microsphere could also be quantitatively detected for Y microspheres with 2% of SPIO. Y microspheres doped with 2% SPIO have the potential to be used for in vivo to quantification of macroscopic intra-hepatic biodistribution. SPIO-labeled Y microspheres offer the potential to use MRI R2* maps to quantitatively depict in vivo biodistribution. With the current study we have demonstrated the potential to optimize SPIO content for future studies intended to quantify microsphere concentrations delivered to liver tumors in vivo. Reference [1] Salem R et al. Int J Radiat Oncol Biol Phys. 2006. [2] Salem R et al.Tech Vasc Interv Radiol. [3] Gupta T et al. Radiology. 2008. Acknowledgements: The authors wish to acknowledge the support of NIH grant RO1 CA134719, and SIR Foundation Pilot Grant. Fig.3. Coronal MGRE images with color-coded overlays depicting voxel-wise estimates of intra-hepatic microsphere concentration. A) pre-infusion; B) postinfusion. Colorbar unit: sec Fig.2 H&E staining of liver tissue with microspheres. Fig.1 a) T2* weighted images of 5% Y microsphere phantoms (TR/TE =200/5.7 ms ). b) Transverse relaxation rates R2* versus Y concentration for 4 different compositions