In cancer, there is a strong body of evidence that increased tissue stiffness contributes to tumour progression¹. The mechano-sensing mechanisms by which tumour cells adapt to microenvironmental changes leading to tumour growth are being actively pursued². These differences are also being exploited in the development of drugs targeting the mechanical properties of the matrix, as therapeutic prevention of tissue stiffening is predicted to impede cancer progression and metastasis¹. Methods to accurately quantify tissue matrix stiffening in vivo may thus provide prognostic biomarkers of tumour progression, and prove useful for monitoring treatment response.

Magnetic resonance elastography (MRE) is being increasingly exploited to directly visualise and quantify tumour mechanical properties in vivo. Our initial pre-clinical MRE investigations have demonstrated that MRE can provide acute imaging biomarkers of treatment-induced tumour necrosis³, and that intracranially-implanted tumours are significantly softer and less viscous than surrounding brain parenchyma⁴. Orthotopic and transgenic mouse models of cancer, which more faithfully emulate human tumour growth patterns and tumour-host stromal interactions, are being increasingly exploited for pre-clinical cancer research. Their effective use must be underpinned by case-specific evidence, establishing that tumour development, progression, radiology and chemoresponsiveness recapitulates the human disease.

In this study, the feasibility of using MRE to assess the viscoelastic properties of orthotopic pancreatic ductal adenocarcinoma (PDAC) xenografts, and tumours arising in a transgenic mouse model of MYCN-amplified neuroblastoma, within the mouse abdomen, was assessed.

All experiments were performed in accordance with the UK Animals (Scientific Procedures) Act 1986. Anaesthetised CD1 nu/nu mice bearing orthotopic PDAC tumours derived from PANC-1 cells (n=2), and Th-MYCN transgenic mice bearing spontaneous neuroblastomas (n=5)⁵, were imaged using a 3cm birdcage coil on a 7T Bruker MicroImaging horizontal MRI system (Bruker Instruments, Ettlingen, Germany). MRE data was acquired in the axial plane using a purpose built platform as previously described³. Maps of the total amplitude of the mechanical wave (A_tot, µm), and absolute value of elasticity G_d and viscosity G_l (both kPa), were reconstructed with an isotropic pixel size of 300 µm, and G_d and G_l determined from a region of interest covering the whole tumour (mean ± 1 s.e.m.).A_tot was larger than 0.3 µm through both tumour types, with wave attenuation along the propagation depth (Figure 1). Homogeneously appearing PDAC xenografts in T₂-weighted images were associated with elevated and discretely distributed regions of elasticity and viscosity, and which enabled clear tumour delineation from surrounding (normal) tissue (Figure 1A). Neuroblastomas appeared more heterogeneous in both T₂-weighted images and maps of G_d and G_l (Figure 1B). Interestingly, the displaced aorta commonly associated with abdominal neuroblastoma, and evident with marked hypointensity in T₂-weighted images (arrowed), was associated with relatively lower G_d and higher G_l (Figure 1B). Comparison of the quantitative viscoelastic properties between the two tumour types revealed the PDAC tumours to be markedly stiffer (G_d: PDAC 5.9 ± 1.0, neuroblastoma 4.1 ± 0.5 kPa) and viscous (G_l: PDAC 4.5 ± 1.1, neuroblastoma 2.3 ± 0.3 kPa).The feasibility of applying MRE to abdominal tumours using our preclinical platform was clearly demonstrated using orthotopic PDAC xenografts, and neuroblastomas arising in a transgenic mouse model. The markedly elevated viscosity and elasticity quantified in the PDAC tumours is wholly in agreement with the well-described stromal-rich environment associated with solid pancreatic tumours⁶. Neuroblastoma is a pathologically diverse disease, with extensive vascularisation. Mechanical cues from the extracellular matrix have been shown to modulate the proliferation, differentiation and expression of MYCN in neuroblastoma in vitro⁷, providing a strong motivation to exploit MRE-derived measurements of viscoelasticity in vivo in the Th-MYCN model shown here. Such measurements may provide additional, complementary imaging biomarkers of neuroblastoma progression and treatment response to those we have already evaluated⁸.MRE can be used as a non-invasive pre-clinical imaging tool to assess the viscoelastic properties of deep-seated tumours arising within the abdomen of mice in vivo. Pancreatic tumours are markedly stiff, whilst neuroblastomas exhibit more heterogeneity in their biomechanical properties and are relatively soft.