Abstract

Colorectal cancer (CRC) is a common cause of cancer-related deaths (12.2% overall mortality, GLOBOCAN 2012 v1.0). Approximately half of CRC patients develop or present with metastasis (mCRC) to the liver. Selective internal radiation therapy (SIRT) involves injection of yttrium-90 (90Y) radiolabelled microspheres into the arterial blood supply of tumours in the liver. Radioactivity prescription for 90Y SIRT is currently based on a body surface area (BSA) formula. The purpose of this study was to conduct a lesion-by-lesion analysis to establish a dose-response relationship. A retrospective review of 24 patients (99 lesions) treated with resin 90Y microspheres was conducted. All patients had inoperable liver metastases refractory to standard chemotherapies. Single photon emission computed tomography (SPECT) images were transformed into dose maps using an in-house MATLAB script and the direct deposition dosimetry method. The 90Y S-value was generated by Monte Carlo-simulation using the EGSnrc/EGS++ code. Baseline and follow-up CT scans were segmented to derive liver and lesion volumes. Mean, median, and D70 (minimum dose to 70% of lesion volume) values derived from dose maps were correlated to change in lesion volume and RECIST 1.1 outcome using linear and logistic regression, respectively. The median administered radioactivity was 1513 MBq (range 847-2185) for lobar and bilobar administrations. The median uniform normal liver dose was 26.3 Gy (range 15.4-41.3, IQR 22.3-30.6) and median lesion dose was 32.8 Gy (range 2.3-90.0, IQR 23.4-48.2). Stable disease or partial response was observed in 17 (71%) patients. Median survival from the date of administration was 313 days. Radiation absorbed dose correlated with lesion response (i.e. a 1 Gy increase in mean, median, and D70 dose resulted in reduction in lesion volume by 2.7%, 2.7%, and 2.3%, respectively, p < 0.02). Higher doses were associated with a decrease in total (combined) lesion volume (by 2.3% and 1.9% for a 1 Gy increase in mean and D70 doses respectively, p < 0.04). Higher lesion doses were associated with a higher probability of RECIST response (odds ratio of 1.05, 1.05, and 1.04 for a 1 Gy increase in mean, median, and D70 doses, respectively, p < 0.2). Higher mean, median, and D70 doses are associated with a decrease in lesion volume and an increased probability of RECIST response. The type of analysis presented here will allow suboptimally treated liver lesions to be managed by a multi-modality approach.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call