Tumor Solid Stress: Assessment with MR Elastography under Compression of Patient-Derived Hepatocellular Carcinomas and Cholangiocarcinomas Xenografted in Mice

SIMPLE SUMMARY: Tumor biomechanical properties, including high viscoelasticity and tumor pressure (solid stress and interstitial fluid pressure), are linked to tumor severity. While tumor viscoelasticity can be quantified with MR elastography, a non-invasive method to quantify tumor pressure remains elusive. In patient-derived hepatocellular carcinomas and cholangiocarcinomas xenografted in mice, we observed that basal elasticity determined during MR elastography under compression had high diagnostic performance in assessing tumor fibrosis content and was independently influenced by interstitial fluid pressure. In contrast, compression stiffening rate had high diagnostic performance in assessing solid stress. Assessment of compression stiffening with MR elastography may provide a non-invasive biomarker of tumor solid stress. ABSTRACT: Malignant tumors have abnormal biomechanical characteristics, including high viscoelasticity, solid stress, and interstitial fluid pressure. Magnetic resonance (MR) elastography is increasingly used to non-invasively assess tissue viscoelasticity. However, solid stress and interstitial fluid pressure measurements are performed with invasive methods. We studied the feasibility and potential role of MR elastography at basal state and under controlled compression in assessing altered biomechanical features of malignant liver tumors. MR elastography was performed in mice with patient-derived, subcutaneously xenografted hepatocellular carcinomas or cholangiocarcinomas to measure the basal viscoelasticity and the compression stiffening rate, which corresponds to the slope of elasticity versus applied compression. MR elastography measurements were correlated with invasive pressure measurements and digital histological readings. Significant differences in MR elastography parameters, pressure, and histological measurements were observed between tumor models. In multivariate analysis, collagen content and interstitial fluid pressure were determinants of basal viscoelasticity, whereas solid stress, in addition to collagen content, cellularity, and tumor type, was an independent determinant of compression stiffening rate. Compression stiffening rate had high AUC (0.87 ± 0.08) for determining elevated solid stress, whereas basal elasticity had high AUC for tumor collagen content (AUC: 0.86 ± 0.08). Our results suggest that MR elastography compression stiffening rate, in contrast to basal viscoelasticity, is a potential marker of solid stress in malignant liver tumors.