Guidance for Y-90 Radioembolization: Lung Shunt Fraction Characteristics Linked to Cancer Type, Tumor Size, and Imaging Modality

Sunday, March 30, 2025

3:18 PM – 3:27 PM CT

Location: Music City Center, 207 CD

Presenting Author(s)

Allan Thomas, PhD (he/him/his)

Assistant Professor
Mallinckrodt Institute of Radiology, Washington University School of Medicine, United States

Author/Co-author(s)

TA

Tharun T. Alamuri, BS

Medical Student
Renaissance School of Medicine, Stony Brook University, United States
JK

John Karageorgiou, MD

Assistant Professor
Washington University School of Medicine St. Louis, United States
DG

Dan Giardina, MD

Associate Professor
Mallinckrodt Institute of Radiology, Washington University School of Medicine, United States
CM

Christopher Malone, MD

Associate Professor
Mallinckrodt Institute of Radiology, Washington University, United States

Purpose: Lung shunt fraction (LSF) estimates are directly tied to patient selection and treatment planning in ⁹⁰Y radioembolization. MAA-based LSF from planar imaging remains the standard in most clinical workflows despite biased and uncertain results {1,2}. This work establishes LSF value distributions based on clinical variables and imaging modality in >200 ⁹⁰Y cases.

Materials and Methods: All glass ⁹⁰Y cases from a 21-month period at a single institution were analyzed. Cancer type (HCC vs. non-HCC) and maximum tumor size were tracked. LSF was computed from planar imaging (planar) and three methods using MAA-SPECT/CT: full (no corrections), 2-cm (2-cm lung/liver contour corrections at boundary), and left (left lung only) {2}.

Results: A total of 217 cases had usable MAA-SPECT/CT for LSF estimates (HCC: 182 cases). Significant differences were observed in LSF from different imaging/correction methods (p< 0.001), tumor size groups (p≤0.03), and HCC vs. non-HCC (p≤0.03). Using planar imaging, 70% and 21% of cases had LSF >5% and >10%, respectively. This reduced to 6% and 1% using 2-cm corrected SPECT/CT. Non-HCC LSF distributions were comparable to HCC < 4 cm (p≥0.06), despite a larger median tumor size of 5.7 cm. In outlier cases beyond the 90^th pct, 12/22 were common between planar and 2-cm SPECT/CT, while 20/22 were common between 2-cm and left SPECT/CT. For cases beyond the 95^th pct, 7/11 and 9/11 were common between planar/2-cm and 2-cm/left, respectively.

Conclusion: LSF varies significantly when computed from different imaging and correction methods. But consistent trends in LSF magnitude with cancer type and tumor size were still uncovered. Dosimetry estimates may be impacted by the details of LSF methods. But identifying outlier cases warranting extra caution for selection and treatment planning can generally be achieved with any LSF method.