The Impact of Radiobiologically Informed Dose Prescription on the Clinical Benefit of 90Y SIRT in Colorectal Cancer Patients.

The purpose of this study was to establish the dose-response relationship of selective internal radiation therapy (SIRT) in patients with metastatic colorectal cancer (mCRC), when informed by radiobiological sensitivity parameters derived from mCRC cell lines exposed to 90Y. Methods: Twenty-three mCRC patients with liver metastases refractory to chemotherapy were included. 90Y bremsstrahlung SPECT images were transformed into dose maps assuming the local dose deposition method. Baseline and follow-up CT scans were segmented to derive liver and tumor volumes. Mean, median, and D70 (minimum dose to 70% of tumor volume) values determined from dose maps were correlated with change in tumor volume and volumetric RECIST response using linear and logistic regression, respectively. Radiosensitivity parameters determined by clonogenic assays of mCRC cell lines HT-29 and DLD-1 after exposure to 90Y or external beam radiotherapy (EBRT; 6 MV photons) were used in biologically effective dose (BED) calculations. Results: Mean administered radioactivity was 1,469 ± 428 MBq (range, 847-2,185 MBq), achieving a mean absorbed radiation dose to tumor of 35.5 ± 9.4 Gy and mean normal liver dose of 26.4 ± 6.8 Gy. A 1.0 Gy increase in mean, median, and D70 absorbed dose was associated with a reduction in tumor volume of 1.8%, 1.8%, and 1.5%, respectively, and an increased probability of a volumetric RECIST response (odds ratio, 1.09, 1.09, and 1.10, respectively). Threshold mean, median and D70 doses for response were 48.3, 48.8, and 41.8 Gy, respectively. EBRT-equivalent BEDs for 90Y are up to 50% smaller than those calculated by applying protraction-corrected radiobiological parameters derived from EBRT alone. Conclusion: Dosimetric studies have assumed equivalence between 90Y SIRT and EBRT, leading to inflation of BED for SIRT and possible undertreatment. Radiobiological parameters for 90Y were applied to a BED model, providing a calculation method that has the potential to improve assessment of tumor control.