Multi-Institutional Comparison of Ablative Radiation Therapy in 5 Versus 15-25 Fractions for Locally Advanced Pancreatic Cancer

Abstract

<h3>Purpose/Objective(s)</h3> Ablative RT doses ∼100Gy BED₁₀ are associated with improved overall survival (OS) in patients with locally advanced unresectable pancreatic cancer (LAPC), but the optimal fractionation scheme remains unknown. <h3>Materials/Methods</h3> We performed a retrospective analysis of consecutive LAPC patients treated at 2 high-volume institutions with either 50 Gy in 5 fractions using a 0.35T MR Linac and online adaptive replanning between 2018-2021 or 67.5-75 Gy in 15-25 fractions using CBCT guidance and selective offline adaptive replanning between 2016-2019. All were treated with breath hold or gating. Elective nodal coverage was used in 94%. Clinical characteristics between the groups were compared using Fisher's exact test. Freedom from local progression (FFLP) by RECIST and OS were estimated from time of RT with the Kaplan-Meier method. Log-rank test and Cox proportional hazards regression model were used for univariate (UVA) and multivariate (MVA) analyses. <h3>Results</h3> 183 patients were evaluated including 62 (34%) treated with 50 Gy in 5 fractions, 23 (123%) treated with 67.5 Gy in 15 fractions, and 98 (54%) treated with 75 Gy in 25 fractions. Median age was 67 years (range 35 to 91), 53% were male, 72% had a head tumor location, 81% were T3/4 with a median tumor size of 3.8 cm (range 1.4-7.4 cm) and 38% were lymph node positive. Patients receiving 5 fractions were more likely to be N0 than patients receiving 15-25 fractions (69.3% vs 44.6%, p=0.0219), and there were no other clinical differences between the groups. 98% received induction chemotherapy for a median of 3.9 months (FOLFIRINOX 61%, Gemcitabine/Nab-paclitaxel 29%, other 10%). Median follow-up was 38 months. Median FFLP and OS were 31 and 18 months, with 2-yr FFLP and 2-yr OS of 62.9% (95%CI 47.9 – 69.3%) and 36.2% (26.0 – 40.9%), respectively. There were no significant differences in FFLP (HR= 1.493, 95%CI 0.8030 – 2.775, p=0.1598) or OS (HR=1.321, 95%CI 0.8835 – 1.974, p=0.1580) between patients receiving 5 vs 15-25 fractions, respectively, on UVA. Likewise, after adjusting for T stage, nodal status, tumor size, chemotherapy type and duration, there were no significant differences in OS between the groups on MVA (HR=1.234, 95%CI 0.8151 – 1.833, p=0.3070). The rate of grade 3+ GI bleeding was 4.8% and 8% (p=0.5476) in patients receiving 5 vs 15-25 fractions, respectively. <h3>Conclusion</h3> In appropriately selected LAPC patients, ablative RT (BED₁₀ ∼100 Gy) in 5 fractions delivered with MR guidance and online adaptive replanning or 15-25 fractions delivered with CT guidance and selective offline adaptive replanning both appear to achieve excellent efficacy and minimal severe toxicity. Choice of fractionation scheme should be based on anatomical considerations and available technology.