3385 - Influence of Pancreatic Ductal Adenocarcinoma Anatomic Tumor Subregion on Oncologic Outcomes Following MRI-Guided Stereotactic Body Radiotherapy
Presenter(s)
R. Ariani, P. T. Courtney, T. C. Wu, J. Deng, A. Raldow, and X. Qi; Department of Radiation Oncology, University of California, Los Angeles, Los Angeles, CA
Purpose/Objective(s): Pancreatic ductal adenocarcinoma (PDAC) remains an aggressive cancer with poor survival. While the clinical efficacy of MRI-guided stereotactic body radiotherapy (MRgSBRT) for PDAC has been described, oncologic outcomes by tumor subregion following MRgSBRT are underexplored.
Materials/Methods: We conducted a retrospective, single-institution study of patients with PDAC treated with MRgSBRT (2015–2024). Demographics, tumor & clinical characteristics, and treatment details were collected. Specific MRgSBRT dosimetric details collected included total dose fractionation, biological effective dose (BED10), and planning target volume (PTV) – a uniform 3 mm expansion from contoured gross tumor volume. Patients were categorized by anatomic tumor subsite and grouped together as head and neck (head/neck) versus body and tail (body/tail) for analysis. Recurrences were categorized as local (LR) or distant (DR), with time to recurrence, overall (OS) and pancreatic cancer-specific survival (PCSS) measured from MRgSBRT completion. Kaplan-Meier and univariate/multivariate (UVA/MVA) Cox regression analyses were performed to evaluate survival outcomes using statistical software, with a two-sided alpha 0.05.
Results: A total of 96 patients were identified. Median age at time of diagnosis was 68 years (IQR 59–76). Anatomic tumor subsites included head (56, 58%), body (20, 21%), tail (12, 13%), and neck (8, 8%). Of cases involving head/neck tumors, most were unresectable (35, 54%). Of those that were borderline (17, 27%) and resectable (12, 19%), 14 underwent surgery. Of cases involving body/tail tumors, most were unresectable (19, 59%). Of those that were borderline (8, 25%) and resectable (5, 16%), 10 underwent surgery. Median PTV was 71.1 cc (IQR 43.7–85.8), and dose fractionation was 50 Gy in 5 fractions (IQR 40-50 Gy), achieving BED10 of 100 Gy (IQR 72–100). Median time to LR was 24.9 months for head/neck tumors (95% CI 3.3-46.4) vs. for body/tail tumors was 17.0 months (95% CI 0.0-45.7) (p=0.39). Median time to DR was 15.5 months for head/neck tumors (95% CI 11.1-19.9) vs. 3.5 months for body/tail tumors (95% CI 0.0-7.7) (p=0.05). On UVA, body/tail tumor location approached significance with shorter time to DR (p=0.05; HR 1.7, ref group head/neck); larger tumor dimension at diagnosis was significantly associated with shorter time to DR (p=0.01, HR 1.2). On MVA, only body/tail tumor location remained significant (p=0.02, HR 2.2). Median OS for head/neck tumors was 23.4 months (95% CI 14.0-32.67) vs. for body/tail tumors was 20.6 months (95% CI 5.5-35.7) (p=0.47). Median PCSS for head/neck tumors was 31.6 months (95% CI 16.9-46.3) vs. for body/tail tumors was 20.6 months (95% CI 5.5-35.7) (p=0.50).
Conclusion: Following MRgSBRT, body/tail tumors were found to be associated with shorter time to DR and worse OS and PCSS. These differences highlight the need for further investigation into the biologic and clinical factors influencing tumor behavior and treatment response in this subgroup.