MRE for Assessment of Histopathological Growth Patterns in Colorectal Liver Metastases

Colorectal cancer ranks third in terms of incidence and second in terms of mortality among all cancers on the global cancer burden, based on the GLOBOCAN 2020 estimates. Among people diagnosed with colorectal cancer, 20% have metastatis, and 40% present with recurrence after previously treated localized disease. Liver is the most common site of metastasis besides lymph nodes, with a 5-year survival rate of less than 20%. Histopathological growth patterns (HGPs) appear at the interface between the tumour border and surrounding liver parenchyma which can be identified by light microscopy on tissue sections. Three primary HGPs have been identified: desmoplastic HGP (dHGP), replacement HGP and the pushing HGP. In the dHGP, the cancer cells are separated from the liver by a capsule of desmoplastic stroma in which new blood vessels are formed by sprouting angiogenesis. Recently, it has been reported that dHGP is linked to a better outcome in terms of survival and more clinical benefit achieved with preoperative anti-angiogenic therapy in patients with colorectal cancer liver metastasis (CRLM). Thus, planning of preoperative treatment might greatly benefit from HGP prediction. Preoperative prediction of HGP is challenging, however, as HGP is a histopathologic finding that can be diagnosed only with a postoperative surgical specimen. Previous studies have made many efforts to identify HGP through radiologic features (e.g., tumor margin, perilesional rim enhancement, or fused radiomics signature) by CT or MRI. However, high variability was observed in the assessment of these imaging characteristics, and they did not provide a direct measure of tumor-liver adhesion, leading to controversy in clinical practice. Recently, an MRE-based technique, termed "slip interface imaging" (SII), has been used to quantify the degree of tumor-peritumoral tissue adhesion in brain tumors and microvascular invasion in hepatocellular carcinoma noninvasively, may have the potential to characterize the mechanical characteristics of the tumor-liver interface and predict HGP of CRLM.