Molecular MRI of the Immuno-Metabolic Interplay in a Rabbit Liver Tumor Model: A Biomarker for Resistance Mechanisms in Tumor-targeted Therapy?
Lynn Jeanette Savic,Isabel Theresa Schobert,Ruth Rebecca Montgomery,Nikhil Joshi,John James Walsh,Jessica Santana,Vasily Pekurovsky,Xuchen Zhang,MingDe Lin,Lucas Adam,Annemarie Boustani,James Duncan,Lin Leng,Richard John Bucala,Fahmeed Hyder,Daniel Coman,Julius Chapiro,Luzie A. Doemel,S. Nahum Goldberg
DOI: https://doi.org/10.1148/radiol.2020200373
IF: 19.7
2020-09-01
Radiology
Abstract:Background The immuno-metabolic interplay has gained interest for determining and targeting immunosuppressive tumor micro-environments that remain a barrier to current immuno-oncologic therapies in hepatocellular carcinoma. Purpose To develop molecular MRI tools to reveal resistance mechanisms to immuno-oncologic therapies caused by the immuno-metabolic interplay in a translational liver cancer model. Materials and Methods A total of 21 VX2 liver tumor-bearing New Zealand white rabbits were used between October 2018 and February 2020. Rabbits were divided into three groups. Group A ( = 3) underwent intra-arterial infusion of gadolinium 160 (Gd)-labeled anti-human leukocyte antigen-DR isotope (HLA-DR) antibodies to detect antigen-presenting immune cells. Group B ( = 3) received rhodamine-conjugated superparamagnetic iron oxide nanoparticles (SPIONs) intravenously to detect macrophages. These six rabbits underwent 3-T MRI, including T1- and T2-weighted imaging, before and 24 hours after contrast material administration. Group C ( = 15) underwent extracellular pH mapping with use of MR spectroscopy. Of those 15 rabbits, six underwent conventional transarterial chemoembolization (TACE), four underwent conventional TACE with extracellular pH-buffering bicarbonate, and five served as untreated controls. MRI signal intensity distribution was validated by using immunohistochemistry staining of HLA-DR and CD11b, Prussian blue iron staining, fluorescence microscopy of rhodamine, and imaging mass cytometry (IMC) of gadolinium. Statistical analysis included Mann-Whitney and Kruskal-Wallis tests. Results T1-weighted MRI with Gd-labeled antibodies revealed localized peritumoral ring enhancement, which corresponded to gadolinium distribution detected with IMC. T2-weighted MRI with SPIONs showed curvilinear signal intensity representing selective peritumoral deposition in macrophages. Extracellular pH-specific MR spectroscopy of untreated liver tumors showed acidosis (mean extracellular pH, 6.78 ± 0.09) compared with liver parenchyma (mean extracellular pH, 7.18 ± 0.03) ( = .008) and peritumoral immune cell exclusion. Normalization of tumor extracellular pH (mean, 6.96 ± 0.05; = .02) using bicarbonate during TACE increased peri- and intratumoral immune cell infiltration ( = .002). Conclusion MRI in a rabbit liver tumor model was used to visualize resistance mechanisms mediated by the immuno-metabolic interplay that inform susceptibility and response to immuno-oncologic therapies, providing a therapeutic strategy to restore immune permissiveness in liver cancer. © RSNA, 2020
radiology, nuclear medicine & medical imaging