183 - Comparative Voxel-Based Dosimetry of Underdosed Tumor Volume and Tumor Absorbed Dose Between Tc-MAA SPECT/CT and 90Y PET/CT in HCC Patients Treated with Radiation Segmentectomy

Monday, April 13, 2026

3:45 PM - 3:54 PM ET

Presenting Author(s)

Mahmoud Odeh, MD

Postdoctoral research fellow
Beth Israel Deaconess Medical Center, United States

Author/Co-author(s)

DD

Diana C. Dinh, MD

Associate Professor of Radiology
Beth Israel Deaconess Medical Center, United States
JW

Jeffrey L. Weinstein, MD, FSIR

Associate Professor of Radiology
Harvard Medical School, United States
MA

Muneeb Ahmed, MD, FSIR

Professor of Radiology
Beth Israel Deaconess Medical Center/Harvard, United States
Ammar Sarwar, MD, FSIR (he/him/his)

Associate Professor of Radiology
Beth Israel Deaconess Medical Center, United States

Purpose: To evaluate the agreement between Technetium Macroaggregated Albumin (Tc-MAA) voxel-based dosimetry and post-Yttrium-90 (90Y) voxel-based dosimetry in hepatocellular carcinoma (HCC) patients treated with radiation segmentectomy.

Materials and Methods: This single-center, retrospective study included adult patients with early-stage solitary HCC treated with radiation segmentectomy using resin microspheres between July 2014 and June 2025, with available Tc-MAA SPECT/CT and post-90Y PET/CT. Voxel-based dosimetry was carried out on both images using the local deposition method with known activity. Evaluation included traditional dosimetry measures, including the mean tumor absorbed dose (TAD), the minimum dose absorbed by 95% of the tumor tissue (D95), and tumor to normal ratio (TNR). Underdosed tumor volume, defined as the percentage of tumor volume receiving less than 100 Gy (V0-100%), was determined. Linear regression and Bland Altman analysis were performed, and Pearson correlation coefficient (r) was calculated.

Results: A total of 40 patients (median age, 71 years [IQR, 66-76], 75% males) were included in the analysis. Median tumor size was 2.7 cm [IQR, 2.1-4.4]. Mean TAD measured on Tc-MAA SPECT/CT was 331 Gy (IQR, 181-572), and mean TAD measured on post-90Y PET/CT was 356 Gy (IQR, 206-521). Both Tc-MAA SPECT/CT and 90Y PET/CT based mean TAD were strongly correlated (r = 0.77). On average, Tc-MAA based mean TAD was higher than 90Y-based mean TAD by 8%. Similarly, Tc-MAA based D95 showed strong correlation with the D95 measured on post-90Y PET/CT (Tc-based D95 = 105 [IQR, 51-216] Gy vs 90Y-based D95 = 134 [IQR, 63-239], r = 0.64), with a mean difference of 19% (95%CI: -5,42). TNR calculated on post-Tc-MAA SPECT/CT was significantly lower than the actual TNR calculated from post-90Y PET/CT (1.88 vs 2.26, p=0.007). On the other hand, there was a weak correlation between the underdosed tumor volume measured on post-Tc-MAA SPECT/CT and Post-90Y PET/CT (V0-100%: 3.6% vs 0.8%, r = 0.24). On average, V-100% predicted by Tc-MAA was 1.89 times (95%CI: 1.5,2.28). higher than the actual underdosed tumor volume determined by post-90Y PET/CT.

Conclusion: Tc-MAA voxel-based dosimetry showed strong correlation and low bias in predicting tumor absorbed doses. However, the correlation for underdosed tumor volume was weak, with Tc-MAA tending to overestimate the underdosed fraction compared with post-Y-90 PET/CT.