249 - Voxel-based versus Single Compartment Ablative Radioembolization Dosimetry Thresholds to Predict Outcomes for Moderate to Large Hepatocellular Carcinomas

Tuesday, April 14, 2026

3:27 PM - 3:36 PM ET

Author/Co-author(s)

Kurt Pianka, BS (he/him/his)

Medical Student
University of California San Diego, United States

Presenting Author(s)

Breanna Collins, BS (she/her/hers)

Medical Student
University of California, San Diego School of Medicine, United States

Author/Co-author(s)

Sarah Placek, DO (she/her/hers)

IR Resident
University of Arizona, United States
SR

Sean Robinson, MD

Integrated Interventional Radiology Resident
Ochsner Health System, United States
Juan Gimenez, MD

Interventional Radiologist
Oschner Health, United States
JM

Jeet Minocha, MD (he/him/his)

Professor, Division of Interventional Radiology
University of California San Diego, United States
TS

Tyler Sandow, MD

Interventional Radiologist
Ochsner Health, United States
Shamar Young, MD

Professor
University of Arizona, United States
ZB

Zachary Berman, MD

Associate Professor of Clinical Radiology
University of California, San Diego, United States

Purpose: Single compartment and personalized tumor dose thresholds have been demonstrated to improve tumor responses and survival for patients with hepatocellular carcinoma (HCC) undergoing radioembolization. However, for larger tumors, which may not be amenable to thermal ablation, optimal ablative dosimetry methods and dose thresholds remain unknown.

Materials and Methods: Multi-center retrospective analysis of consecutive patients at three centers who underwent radioembolization for untreated HCC ≥4 cm between June 2018 – June 2024 was performed. Imaging response was assessed by modified Response Evaluation Criteria in Solid Tumors (mRECIST). Post-TARE SPECT/CT were analyzed using voxel-based dosimetry software to estimate perfused volume and tumor-volume doses. The associations between imaging response at three months post-TARE with estimated dose and overall survival (OS) were evaluated.

Results: 109 patients met inclusion criteria for this study: 55 complete-response (CR), 54 non-CR at three-months post-TARE. Single compartment dose, minimum dose to 95% of tumor volume (D95), and D50 were significantly higher in patients with CR: 601 versus 316 Gy perfused dose (p < 0.001), 500 versus 213 Gy D95 (p < 0.001), and 832 versus 401 Gy D50 (p < 0.001). CR tumors tended to have smaller estimated pre-TARE size (4.8 versus 5.8 cm, p < 0.001) and were treated within a smaller perfused volume (235 versus 441, p < 0.001). CR was associated with a median OS of 3.99 versus 2.23 years for non-CR, with a hazard ratio (HR) of 0.59 (95% confidence interval (CI) 0.37 – 0.93, p = 0.02). The associations between dose thresholds and OS are summarized in Table 1.

Conclusion: Complete imaging response to radioembolization at three months was correlated with overall survival for HCC ≥4 cm. Multiple dose thresholds were correlated with complete response and survival, with D95 consistently having the numerically lowest HR at each dose threshold.