Sep 29

QP 27 - Getting SMART About Ablative Radiation Therapy in Pancreaticobiliary Cancer and Liver Metastases

1157 - Excellent 2- and 3-Year OS after Ablative MRgRT for Localized Pancreatic Cancer without Surgery

05:20pm - 05:25pm ET

Room 259

Presenter(s)

Marsha Reyngold, MD, PhD - Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, New York

M. Reyngold¹, V. Kirichenko^1,2, E. O'Reilly³, M. Mcmillan¹, A. J. Wu¹, P. B. Romesser¹, N. Cote⁴, P. Godoy Scripes⁵, J. G. Mechalakos⁵, N. Tyagi⁵, and C. H. Crane¹; ¹Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, ²Thomas Jefferson Medical School, Philadelphia, PA, ³Department of Medical Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, ⁴Memorial Sloan Kettering Cancer Center, New York, NY, ⁵Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY

Purpose/Objective(s): Ablative doses of radiotherapy (RT) (=100Gy biologically effective dose) are associated with improved overall survival (OS) in patients with localized pancreatic adenocarcinoma (PDAC). MR-guided (MRg) adaptive technique facilitates safe delivery of high-dose RT, but long-term outcomes have not been reported.

Materials/Methods: Retrospective single institution analysis of consecutive patients with localized PDAC treated with ablative MRgRT on a 1.5 Tesla Unity MR Linac without surgery and a minimum follow-up of 12 months. Patients with localized PDAC and no or minimal contact between the tumor and luminal gastrointestinal (GI) tract were candidates for MRgRT. Treatment consisted of 50Gy to high-dose planning target volume (PTV) and 25-33Gy to microscopic-dose PTV in 5 fractions over 2 weeks. Dose constraints included D0.035cm³ < 33Gy and D5cm³ of < 27Gy for stomach/small bowel and D0.035cm³ < 35-38Gy and D5cm³ < 30Gy for large bowel. Online contouring for the adapt-to-shape (ATS) workflow was performed in MIM software on T1, T2 and balanced MR sequences. Respiratory motion was reduced below 5mm with a compression belt and continuously monitored with cine. OS and freedom from local progression (FFLP) were estimated from the time of RT with the Kaplan-Meier method. Univariate analysis was performed using Cox logistic regression.

Results: Between 5/2020 and 12/2024, 123 patients with localized PDAC were treated with ablative MRgRT without surgery (primary disease: N=106, locoregional recurrence: N=17, median age 68 [range, 41-94] years, 55.2% male, 73.9% ECOG performance status 0-1, median tumor size 3.4 [range, 1.0-7.2] cm, 54.5% in the head and median CA19-9 171 [range, 0-6465] U/mL). Of 106 patients with primary disease, 94 (88.7%) were locally advanced or borderline resectable, 3 (2.8%) were resectable and 9 (8.4%) had metastatic disease. One hundred five patients (85.4%) received induction chemotherapy (mFOLFIRINOX: N=82, gemcitabine/nab-paclitaxel: N=18, other: N=5). Median PTV_high-doseD95 was 41.5 Gy (range, 33.1Gy-51.2Gy) and gross tumor volume (GTV) V100% was 80% (48-100%). The median follow up was 24.2 (95%CI, 19.5 – 38.2) months. OS rates at 2 and 3 years from RT were 37.4% (95%CI, 27.0-47.9) and 26.3% (95%CI, 16.5-37.2), respectively. FFLP rate at 2 and 3 years from RT was 76.8% (95%CI, 52.4-89.8) for both. On univariate analysis, only GTV D80 (HR-0.9978, 95%CI, 0.9956-0.9999, p=0.0389), but not minimum dose, PTV D95/D90/D80, GTV D95/D90, GTV V100% showed a significant association with FFLP. No clinical or dosimetric factor was associated with OS.

Conclusion: Delivery of 50Gy in 5 fractions to localized PDAC using MRgRT with online ATS replanning resulted in a high probability of durable local control and unprecedented long-term survival without surgery. Better local control is achieved with higher target coverage by 40Gy. These data support prioritization of 50Gy in 5 fractionation in prospective trials of ablative RT when online adaptive technology is available.