Predictive Partition Dosimetry and Outcomes after Yttrium-90 Resin Microsphere Radioembolization of Colorectal Cancer Metastatic to the Liver: A Retrospective Analysis

Abstract

Purpose: To characterize estimated absorbed tumor dose (AD_T), objective response (OR), and estimated target dose of liver metastatic colorectal cancer (mCRC) after resin microsphere yttrium-90 (⁹⁰Y) radioembolization using partition dosimetry.

Materials and methods: In this retrospective, single-center study, multicompartment dosimetry of index tumors undergoing ⁹⁰Y radioembolization from October 2013 to July 2022 was performed using MIM SurePlan and pretreatment technetium-99m macroaggregated albumin infusion data. Thirty-eight patients with mCRC underwent treatments for 59 index tumors. Patients were imaged every 2-3 months after treatment and then every 3-6 months after disease control to determine the best response per Response Evaluation Criteria in Solid Tumors 1.1. Responses were categorized as OR or nonresponse (NR). A Cox proportional hazards model evaluated the probability of tumor OR and local progression-free survival (LPFS) based on AD_T.

Results: Patients had a median follow-up of 116 days (interquartile range [IQR], 69-231 days). The AD_T was higher for OR patients than for NR patients (median, 130.8 [IQR, 85.6-239.0] vs 40.6 [IQR, 26.0-66.3] Gy; P < .001). A greater percentage of OR than NR patients were treated with activities calculated by partition modeling (54% vs 12%; P = .005). Only AD_T predicted response (P = .032). At 6 months, an AD_T of 120 Gy predicted a 55% (95% CI, 0.0%-89%) probability of OR. Only AD_T (P = .010) and female sex (P = .014) predicted LPFS. At 1 year, an AD_T of 120 Gy predicted a 70% (95% CI, 35%-100%) probability of LPFS.

Conclusions: Tumor dose was the strongest predictor of OR for mCRC. Administration of an estimated 120 Gy to mCRC predicted 55% OR with ⁹⁰Y resin microspheres at 6 months.