Idarubicin-Loaded ONCOZENE Drug-Eluting Bead Chemoembolization in a Rabbit Liver Tumor Model: Investigating Safety, Therapeutic Efficacy, and Effects on Tumor Microenvironment

Abstract

Purpose: To investigate toxicity, efficacy, and microenvironmental effects of idarubicin-loaded 40-μm and 100-μm drug-eluting embolic (DEE) transarterial chemoembolization in a rabbit liver tumor model.

Materials and methods: Twelve male New Zealand White rabbits with orthotopically implanted VX2 liver tumors were assigned to DEE chemoembolization with 40-μm (n = 5) or 100-μm (n = 4) ONCOZENE microspheres or no treatment (control; n = 3). At 24-72 hours postprocedurally, multiparametric magnetic resonance (MR) imaging including dynamic contrast-enhanced (DCE), diffusion-weighted imaging (DWI), and biosensor imaging of redundant deviation in shifts (BIRDS) was performed to assess extracellular pH (pHe), followed by immediate euthanasia. Laboratory parameters and histopathologic ex vivo analysis included fluorescence confocal microscopy and immunohistochemistry.

Results: DCE MR imaging demonstrated a similar degree of devascularization of embolized tumors for both microsphere sizes (mean arterial enhancement, 8% ± 12 vs 36% ± 51 in controls; P = .07). Similarly, DWI showed postprocedural increases in diffusion across the entire lesion (apparent diffusion coefficient, 1.89 × 10-3 mm2/s ± 0.18 vs 2.34 × 10-3 mm2/s ± 0.18 in liver; P = .002). BIRDS demonstrated profound tumor acidosis at baseline (mean pHe, 6.79 ± 0.08 in tumor vs 7.13 ± 0.08 in liver; P = .02) and after chemoembolization (6.8 ± 0.06 in tumor vs 7.1 ± 0.04 in liver; P = .007). Laboratory and ex vivo analyses showed central tumor core penetration and greater increase in liver enzymes for 40-μm vs 100-μm microspheres. Inhibition of cell proliferation, intratumoral hypoxia, and limited idarubicin elution were equally observed with both sphere sizes.

Conclusions: Noninvasive multiparametric MR imaging visualized chemoembolic effects in tumor and tumor microenvironment following DEE chemoembolization. Devascularization, increased hypoxia, coagulative necrosis, tumor acidosis, and limited idarubicin elution suggest ischemia as the predominant therapeutic mechanism. Substantial size-dependent differences indicate greater toxicity with the smaller microsphere diameter.

Conflict of interest statement

Conflict of Interest:

This research was funded by a research grant from Boston Scientific Corporation, MA, USA as well as by the NIH/NCI R01 CA206180 research grant.

T.B., J.M.M.vB., F.L-G., M.M., I.R., L.A., J.J.W., S.H., A.S., T.S., B.G. have no conflicts of interest to declare.

J.F.G. is an employee of Prescience Labs.

L.J.S. has received grants from the Leopoldina Foundation, Society of Interventional Oncology, and Rolf. W. Guenther Stiftung.

J.S.D. has received grants from Phillips Research North America and the National Institutes of Health (NIH/NCI R01 CA206180).

D.C.P has received grants from the National Institutes of Health, Philips Healthcare.

F.H. has received grants from Phillips Research North America and the National Institutes of Health (NIH/NCI R01 CA206180).

D.C. has received grants from Phillips Research North America and the National Institutes of Health (NIH/NCI R01 CA206180).

J.C. has received grants from the National Institutes of Health (NIH/NCI R01 CA206180), the Society of Interventional Oncology, the German-Israeli Foundation for Scientific Research and Development, Guerbet Healthcare, Boston Scientific and Philips Healthcare.

Copyright © 2020 SIR. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1:. Experimental study design

12 rabbits were implanted in the left liver lobe with tumors harvested from donor rabbits. Tumors were grown for 14 days until multidetector computed tomography (MD-CT), transarterial chemoembolization (TACE) and cone-beam computed tomography (CBCT) were performed. Blood samples were obtained at specific time points. Multi-parametric MRI (mpMRI) was performed 24–72 h post-procedurally, after which animals were immediately euthanized and necropsied for radiological-histopathologic evaluation.

Figure 2:. Pre- and post-procedural CT imaging

White arrows indicate the tumor in pre- and post-procedural contrast and non-contrast CT imaging in both treatment groups. Post-TACE cone-beam non-contrast CT demarcates successful tumor embolization by visualization of trapped contrast medium within tumor tissue.

Figure 3:. Laboratory analyses of liver transaminases

Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) serum levels (U/l) pre-procedural and 5 min, 24 h and 48 h after chemoembolization in both treatment groups (40μm blue and 100μm red) and controls (black). A higher increase of both ALT and AST was observed in 40μm DEEs compared to 100μm DEEs.

Figure 4:. Multiparametric MRI sequences in representative control and embolized animals

White arrows indicate the tumor. T1-weighted, non-contrast MRI in axial plane indicate tumor location. Contrast-enhanced T1-weighted images in arterial phase show a tumor rim enhancement in controls and which disappears in embolized animals. Dynamic contrast-enhanced MRI (DCE-MRI) demarcates the tumor rim as a hyperenhanced outline that becomes undistinguishable in embolized animals suggestive of tumor devascularization. The apparent diffusion coefficient (ADC) map demarcates an increase in tumor diffusion in embolized animals. The last row represents pHe mapping with BIRDS. The tumor (red outline) is more acidotic (blue voxel color) than corresponding liver parenchyma (blue outline).

Figure 5:. Time-enhancement-curves on dynamic contrast-enhanced MRI

The tumor is depicted in black, arterial enhancement in red and liver enhancement suggestive for the portal venous phase is visible in green. Overall, arterial and liver enhancement are subject to individual fluctuations but do not differ in all three animals. In controls, the tumor shows a strong enhancement whereas in both treatment groups the tumor remains completely unenhanced suggesting a successful embolization and thus devascularization of the tumor.

Figure 6:. Histopathological correlation of imaging findings

Histology included H&E, PNCA, TUNEL, HIF-1a and pimonidazole staining (vertical rows) in treatment groups (horizontal rows). Control tumors demonstrated a viable, proliferating tumor rim with high expression levels of PCNA and fragments of necrosis and corresponding hypoxic regions confined to the tumor core. In embolized tumors, no viable tumor cells were identifiable confirmed by a disappearance of PCNA and upregulation of TUNEL expression. Expressions of hypoxic markers HIF-1α and pimonidazole exceeded tumor margins and were detectable in a radius of >50% of the tumor diameter in 40μm and >10% in 100μm microspheres.

Figure 7:. Immunofluorescence and histological correlation

Immunofluorescence (IF) and H&E in a representative animal of 40μm and 100μm. A, D: H&E staining of embolized vessels containing idarubicin-loaded spheres. B, E: IF staining of tumor (blue outline) with background DAPI staining and idarubicin auto-fluorescence (green). Idarubicin drug signal is detectable peri-vascular in both DEEs. C, F: IF staining with segmented histogram heat map for quantification of idarubicin drug concentration.