Increased absorbed liver dose in Selective Internal Radiation Therapy (SIRT) correlates with increased sphere-cluster frequency and absorbed dose inhomogeneity

BACKGROUND The higher tolerated mean absorbed dose for selective internal radiation therapy (SIRT) with intra-arterially infused (90)Y microspheres compared to external beam therapy is speculated to be caused by absorbed dose inhomogeneity, which allows for liver regeneration. However, the complex liver microanatomy and rheology makes modelling less valuable if the tolerance doses are not based on the actual microsphere distribution. The present study demonstrates the sphere distribution and small-scale absorbed dose inhomogeneity and its correlation with the mean absorbed dose in liver tissue resected after SIRT. METHODS A patient with marginally resectable cholangiocarcinoma underwent SIRT 9 days prior to resection including adjacent normal liver tissue. The resected specimen was formalin-fixed and sliced into 1 to 2-mm sections. Forty-one normal liver biopsies 6-8 mm in diameter were punched from these sections and the radioactivity measured. Sixteen biopsies were further processed for detailed analyses by consecutive serial sectioning of 15 30-μm sections per biopsy, mounted and stained with haematoxylin-eosin. All sections were scrutinised for isolated or conglomerate spheres. Small-scale dose distributions were obtained by applying a (90)Y-dose point kernel to the microsphere distributions. RESULTS A total of 3888 spheres were found in the 240 sections. Clusters were frequently found as strings in the arterioles and as conglomerates in small arteries, with the largest cluster comprising 453 spheres. An increased mean absorbed dose in the punch biopsies correlated with large clusters and a greater coefficient of variation. In simulations the absorbed dose was 5-1240 Gy; 90% were 10-97 Gy and 45% were <30 Gy, the assumed tolerance in external beam therapy. CONCLUSIONS Sphere clusters were located in both arterioles and small arteries and increased in size with increasing sphere concentration, resulting in increased absorbed dose inhomogeneity, which contradicts earlier modelling studies.