A Retrospective Review Of Outcomes In Locally Advanced Pancreatic Cancer Patients Treated With Dose-Escalated External Beam Radiation

Abstract

Previous work has suggested improved clinical outcomes with increasing radiation dose for patients with locally advanced pancreatic cancer (PC), but heterogeneous doses and schedules are used across institutions. To evaluate the utility of more uniform standard versus dose-escalated external beam radiation given after modern chemotherapy, we reviewed the outcomes at a single institution among locally advanced PC patients. We identified patients with inoperable or borderline resectable (BR) PC treated with conventional or hypofractionated radiation since the routine use of FOLFIRINOX and Gemcitabine/Abraxane began at our institution (2014-2020). Patients were excluded if they had metastatic disease or concurrent malignancy. Patients were stratified into high-dose (BED10 >70) and standard-dose (BED10<70) groups. Toxicity was evaluated based on CTCAE v 5.0. Overall survival (OS), freedom from local failure (FFLF), and freedom from distant failure (FFDF) were calculated using the Kaplan-Meier method, and compared using the log-rank test. A total of 42 patients, 21 in each dose group, were identified. Median age was 67, with 43% of female subjects. Twenty-nine (71%) subjects had BR disease. Median follow up was 8.5 months. The most common standard dose regimen was 50.4 Gy in 28 fractions (57%), and the dose escalated regimen was 67.5 Gy in 15 fractions (95%). Five (24%) patients in the standard dose and 15 (71%) patients in the escalated dose groups were treated using MR-guided RT. Seven (33.3%) and 6 (28.5%) patients in the standard- and high- dose groups, respectively, underwent pancreaticoduodenectomy following RT. OS was not significantly different between groups (median 8.3 vs 10.0 months, p = 0.853). FFLP (12.2 vs 12.4 months, p = 0.42) and FFDF (16.4 vs 9.0, p = 0.72) were also not different between dose groups. Regardless of dose, patients that underwent resection after radiation had significantly greater OS (7.6 vs 25.7 months, p = 0.01). Among patients who did not proceed to resection, there was a significant improvement in FFLF (median 4.3 vs. 9.1 months, p = 0.05) in the dose-escalated group, and a non-significant improvement in OS (median 6.4 vs 10.0 months, p = 0.10). Three grade 3 GI toxicities occurred (one directly attributable to RT) in the standard dose group, and no grade ≥3 toxicities occurred in the dose-escalated group. Among a cohort of patients with locally advanced PC, use of a dose-escalated radiation resulted in no improvement in overall clinical outcomes with similar toxicity. In patients that did not receive surgery, there was a clinically significant improvement in FFLF and trend towards improvement in survival suggesting truly unresectable patients may benefit most from dose escalation. Prospective trials are warranted to address the utility of dose-escalation in this setting.