184 - Tumor-to-normal Ratio Derived from Direct Hepatic Arterial Injection 4DCT Perfusion Imaging and Dual Energy CT During Mapping Angiography: Comparison with Tc-99m MAA SPECT/CT

Monday, April 13, 2026

3:54 PM - 4:03 PM ET

Presenting Author(s)

HB

Harika Barri, MD

Post Doctoral Fellow
University of Washington, United States

Author/Co-author(s)

CH

Charles Hua, MD

Assistant Professor
University of Washington, United States
AS

Avik Som, MD, PhD

Assistant Professor
University of Washington, United States
Grace Laidlaw, MD, MS (she/her/hers)

Assistant Professor
University of Washington, United States
JJ

James Jeffries, MD

Assistant Professor
University of Washington, United States
MA

Matthew Abad-Santos, MD

Assistant Professor
University of Washington, United States
Hugh McGregor, MD

Associate Professor
University of Washington, United States

Purpose: To compare tumor-to-normal ratio (TNR) derived from direct hepatic arterial injection 4-dimensional computed tomography perfusion imaging (4DCT) and dual energy CT (DECT) during mapping angiography with single photon emission computed tomography (SPECT)/CT after technetium-99m macroaggregated albumin (Tc-99m MAA) infusion.

Materials and Methods: Twenty consecutive mapping angiograms prior to radioembolization for hepatocellular carcinoma performed on a hybrid angiography-CT unit (Alphenix 4DCT, Canon Medical Systems, Otawara, Japan) were retrospectively reviewed. Catheters were positioned in the common hepatic artery and low dose volumetric CTs were obtained at 1.5-5 second intervals for 60 seconds after injection of 15 milliliters of contrast (Iohexol, Omnipaque 300, GE Healthcare, IL, USA) over 5 seconds. DECT was then obtained 5-7 seconds after a repeat contrast injection. Four millicuries of Tc-99m MAA were then infused into the right or left hepatic artery followed by SPECT/CTs. Tumor and background liver volumes of interest (VOI) were drawn using a semi-automated process (Vitrea Advanced Visualization, Canon Medical Systems, Otawara, Japan) on 4DCT to generate time-density curves. Areas under the curve were then measured and used to determine TNR. Regions of interest around the tumor and non-tumoral perfused liver were drawn on DECT. Iodine concentrations were recorded and used to measure TNR. VOIs were drawn around the tumor and non-tumoral perfused liver on SPECT/CT using a semi-automated process (Visage Imaging, CA, USA) and the mean standardized uptake value was recorded and used to calculate TNR. A paired t-test was used to compare TNR derived from 4DCT and DECT with SPECT/CT.

Results: Mean tumor size was 3.4 +/- 1.9 cm (range 1.5-8.8 cm). Mean tumor volume was 50.4 +/- 56.6 cm³ (range 13.3-227.2 cm³) and mean non-tumoral perfused liver volume was 427.4 +/- 256.7 cm³ (range 100.4-1263.5 cm³). Mean tumor to normal ratio was 3.7 +/- 3.6 (range 1.4-14.3) on 4DCT perfusion imaging, 4.0 +/- 2.6 (range 2-12) on DECT, and 4.2 +/- 1.9 (range 1.7-8.84) on SPECT/CT. There was no significant difference in TNR calculated from 4DCT and SPECT/CT (p=0.90) or DECT and SPECT/CT (p=0.80).

Conclusion: TNR derived from direct hepatic arterial injection 4DCT perfusion and DECT during mapping angiography is comparable to Tc-99m MAA SPECT/CT. These techniques may enable real-time TNR measurement, potentially obviating MAA SPECT/CT and guiding strategies to modulate TNR before treatment (flow diversion or vasodilator infusion).