DOI: 10.3390/livers6050100 ISSN: 2673-4389

Association of Multiparametric Magnetic Resonance Imaging-Derived Lipid Content in the Liver and Tumorous Liver Tissue with Hepatic Metastatic Characteristics in Uveal Melanoma

Sebastian Fuckel, Christina Dix, Tobias Hepp, Claudius Sebastian Mathy, Katharina Tkotz, Michael Uder, Carola Berking, Michael Erdmann, Tobias Bäuerle

Background: Uveal melanoma (UM) typically metastasizes to the liver for yet unclear reasons. Hepatic steatosis influences metastasis in several cancer entities, and increased lipid synthesis and metabolism in the cancer tissue promotes tumor progression. The aim of this study was to investigate the association of MRI-derived lipid content in non-tumorous liver tissue and liver metastases with the biological behavior and metastatic characteristics of UM liver metastases, as well as with survival in patients with metastatic UM. Methods: The lipid content in the liver and hepatic metastases was quantified by multiparametric magnetic resonance imaging in 46 patients with metastasized UM prior to therapy. The perfusion characteristics of liver metastases were assessed using contrast-enhanced MRI-derived arterial enhancement ratios (metastasis-to-aorta signal intensity ratio), and diffusion-weighted imaging was used to evaluate tissue cellularity in metastatic tissues. Results: We found a significant correlation between body mass index (BMI) and hepatic lipid content (Spearman’s r = 0.47; unadjusted p = 0.0027; adjusted p = 0.03; 95% CI: 0.17 to 0.69), between the hepatic lipid content and volume of the metastases (Spearman’s r = −0.42; unadjusted p = 0.007; adjusted p = 0.036; 95% CI: −0.65 to −0.12) after correction for multiple testing. Higher hepatic lipid content (≥5%) was associated with improved survival with stage IV (p = 0.00043), whereas metastatic lipid content showed no significant association in Kaplan–Meier analysis. In Cox proportional hazards analysis, higher hepatic lipid content was associated with reduced hazard ratio of death independent of tebentafusp treatment (without tebentafusp: p = 0.0034; HR = 0.497; with tebentafusp: p = 0.0016; HR = 0.318) while no consistent association was observed for metastatic lipid content. Conclusions: Metabolic changes associated with a high BMI may contribute to lipid-induced changes in the liver microenvironment that may play a role in the tumor biology of hepatic metastases in patients with UM.