Main Session
Sep 30
PQA 08 - Gastrointestinal Cancer, Nonmalignant Disease, Palliative Care

3413 - Association of Central Hepatobiliary Tract Radiation Dose and Hepatobiliary Outcomes after MRI-Guided Liver Stereotactic Body Radiotherapy

02:30pm - 03:45pm PT
Hall F
Screen: 13
POSTER

Presenter(s)

Patrick Courtney, MD, MAS - UCLA David Geffen School of Medicine/UCLA Medical Center, Los Angeles, CA

P. T. Courtney1, J. Deng1, M. Ghafarian2, D. Ruan1, M. Cao3, M. L. Steinberg1, A. Raldow1, and T. C. Wu1; 1Department of Radiation Oncology, University of California, Los Angeles, Los Angeles, CA, 2University of California Los Angeles, Los Angeles, CA, United States, 3Department of Radiation Oncology, University of California San Francisco, San Francisco, CA

Purpose/Objective(s): To better understand toxicity after MRI-guided SBRT (MRgSBRT) to liver tumors in relation to the central hepatobiliary tract (cHBT).

Materials/Methods: This is a single-institution retrospective study of patients with primary or secondary liver tumors who received MRgSBRT between 2016-2024. The cHBT was retrospectively contoured as a 15mm isotropic expansion upon the portal vein. Endpoints were Child-Pugh score (CPS), elevation above normal range of total bilirubin (Tbili) and alkaline phosphatase (ALP) on first and 3-month lab evaluations after MRgSBRT, and hepatobiliary events (HBEs: hepatobiliary infection; biliary or gastroduodenal stricture, ulcer, or fistula; or new biliary stent/unplanned stent exchange) after MRgSBRT. For HBEs, patients were censored upon intrahepatic progression or additional liver-directed therapy. Logistic regression was used to measure the association of endpoints with previously reported cHBT constraints for 5- and 3-fraction CT-guided SBRT (Osmundson et al, IJROBP, 2015), (V40<21cc, V37.7<24cc) and (V33.8<21cc, V32<24cc), respectively, and pre-MRgSBRT liver function. 1-year freedom-from local progression (FFLP) was calculated.

Results: We identified 44 (49% primary) liver tumors in 41 patients. Median follow-up was 10 months (IQR 4-31) from MRgSBRT. The median prescribed dose was 50 Gy in 5 fractions. Before MRgSBRT, 3 (7%) patients had a CPS of B/C, 10 (23%) had elevated Tbili, and 25 (57%) had elevated ALP. Median time to first lab evaluation after MRgSBRT was 17 days. 2 (4%) patients had a worse CPS at first posttreatment lab evaluation of which 1 returned to baseline at 3 months post-MRgSBRT. 2 (4%) patients with normal Tbili pre-MRgSBRT had Tbili elevation at first lab evaluation of which 1 normalized and 1 remained elevated to CTCAE grade 3 level at 3 months. 1 (2%) patient with normal ALP pre-MRgSBRT had ALP elevation at first evaluation which normalized at 3 months. 13 (30%) HBEs occurred at a median of 2.1 months after MRgSBRT, comprised of 7 (16%) infections, 2 (4%) unplanned stent exchanges, 2 strictures, 1 (2%) ulcer, and 1 fistula. All HBEs occurred in patients with CPS B/C or elevated Tbili or ALP pre-MRgSBRT. Of the 2 patients with persistently worse CPS or elevated Tbili at 3 months post-MRgSBRT, neither met the V40 or V37.7 constraints, and both were hospitalized. Patients exceeding the V40 constraint had significantly higher odds of a HBE (odds ratio [OR] 11, 95% CI 2-53, p<0.01), as did those exceeding the V37.7 constraint (OR 19, 95% CI 3-106, p<0.01). These associations remained significant when controlling for pre-MRgSBRT liver function. 1-year FFLP was 72% (95% CI 51-86) and not significantly different between primary (83%, 95% CI 61-96%) and secondary liver tumors (65%, 95% CI 44-85) (p=0.39).

Conclusion: Patients with pre-MRgSBRT liver dysfunction face an increased risk of HBEs following MRgSBRT. Those exceeding cHBT constraints had sustained declines in CPS or grade 3 elevations in Tbili and increased risk of HBEs after MRgSBRT.