Using Voxel-Based Dosimetry to Evaluate Sphere Concentration and Tumor Dose in Hepatocellular Carcinoma Treated with Yttrium-90 Radiation Segmentectomy with Glass Microspheres
Clinical question
How can we use voxel-based dosimetry to better understand sphere distribution and tumor dose in Y90 glass microsphere radiation segmentectomy; and does it validate the current literature on radiologic and pathologic outcomes following radiation segmentectomy?
Take away point
Voxel dosimetry is a crucial tool in better understanding of Y90 sphere activity/distribution and their effects on tumor treatment response and lasting oncologic outcomes. It is indispensable
Reference
Sandow, T., Gimenez, J., Nunez, K., Tramel, R., Gilbert, P., Oliver, B., Cline, M., Fowers, K., Cohen, A., & Thevenot, P. (2024). Using Voxel-Based Dosimetry to Evaluate Sphere Concentration and Tumor Dose in Hepatocellular Carcinoma Treated with Yttrium-90 Radiation Segmentectomy with Glass Microspheres. Journal of vascular and interventional radiology : JVIR, 35(11), 1602–1612.e1.
https://doi.org/10.1016/j.jvir.2024.05.020 Click here for the full article Study Design
Retrospective, observational, single institution
Funding Source
None
Setting
Academic setting, Oschner Health, New Orleans, LA.
Figure
Summary
This retrospective study looked at 56 patients with treatment-naive, solitary hepatocellular carcinoma (HCC) with Barcelona clinic Liver Cancer (BCLC) Stage 0-A tumor burden over a two-year period from January 2020 to October 2022. The included patients were treated with Y90 glass microspheres in a segmental or subsegmental delivery for ablation radiation segmentectomy. Treatment was done in two-step fashion starting with a mapping angiography session, which included calculation of a lung shunt fraction, followed by a treatment session. Initially, target radiation doses were > 200 Gy. This was increased to > 400 Gy after the publication of the LEGACY study in 2021. Immediately following treatment, patients were imaged using Bremsstrahlung SPECT/CT. For post-treatment dosimetry, pre-treatment diagnostic CT or MRI studies were incorporated with the post-treatment nuclear medicine SPECT/CT.
Using voxel dosimetry and the Mirada DBx Build 1.2.0 Simplicit90Y software, total liver volume, perfused volumes, and tumor volume were calculated. Tumor absorbed dose (TAD) and normal tissue absorbed dose were calculated using the multicompartment dosimetry model. Additional calculations included the sphere concentration per tumor volume, the number of spheres delivered to normal and tumoral tissue, minimum absorbed doses at 70% (D70), 90% (D90), and 99% (D99) of the total tumor volume, and minimum dose threshold to the entire tumor for 200 Gy (V200 100%) and 400 Gy (V400 100%). The primary endpoint for the study was progression of HCC based on mRECIST imaging criteria. In patients successfully bridged to transplant, pathologic information related to degree of necrosis was obtained. Statistical analysis included linear regression, Kaplan-Meier survival curves, Mann-Whitney test for continuous variables, and chi-square or Fisher exact test for categorical variables. Adverse events were monitored for 180 days and graded based on the Common Terminology Criteria for Adverse Events version 5.0.
Key findings:
· Smaller perfused volumes correlated with higher sphere concentrations in nontumoral liver and tumor.
· Larger tumors showed a nonlinear increase in total spheres deposited.
· Tumoral and nontumoral sphere concentrations had a direct linear relationship.
Clinical outcomes:
· Objective response rate: 96% (83% complete response [CR], 14% partial response).
· CR correlated with higher D70, D90, D99, and V400 100%.
· Histopathology (15 patients): 33% achieved complete pathologic necrosis (CPN); remaining tumors showed 80-99% necrosis.
· CPN was associated with smaller tumors and higher D99.
· All adverse events were treatment-related (≤Grade 3) and unrelated to nontumoral TAD.
Commentary
As the literature on Y90 radioembolization for grows in both quantity and nuance, sphere activity (SA) and sphere distribution has grown in importance. To understand why, it’s useful to review both the goal of radioembolization and the current techniques of treatment. The goal of Y90 treatment in HCC is to deliver a critical radiation dose to the tumor, with the ideal goal being complete pathologic necrosis (CPN). The LEGACY trial included a cohort of patients who received median doses of 400 Gy, 100% of which had CPN. Since the LEGACY trial, 400 Gy has become the benchmark dose for achieving CPN.
In an ideal treatment scenario, each part of the tumor would see at least 400 Gy and become completely necrotic. However, HCC’s intrinsic heterogeneity—driven by “solid stress” from tumor growth compressing vessels—creates cold areas with limited sphere deposition. Enough activity needs to be deposited into all parts of the tumor, even the cold areas, so the surrounding tumor cells get adequate absorbed dose to achieve complete pathologic necrosis. Balancing sufficient SA to penetrate these regions while avoiding uneven distribution remains a challenge.
Utilizing voxel dosimetry, the article by Sandow et al. reproduced and further delineated many of the findings from the Y90 literature. Voxel dosimetry metrics of D70, D90, D99, and V400 100% correlated strongly with radiologic and pathologic response. Smaller tumors, likely due to homogeneous hypervascularity, achieved higher sphere concentrations and CPN rates. Sphere concentration linearly predicted dose coverage (D70/ D99), emphasizing the need for both activity and distribution optimization.
The study validates voxel dosimetry as a robust analytic tool for future research in personalized Y90 radioembolization for the treatment of HCC and other types of malignancies, offering granular insights into radioembolization dose thresholds and tumor biology.
Post Author
Sean Roger, MD
Interventional Radiology Fellow, PGY-6
University of Massachusetts
