Genomically adjusted radiation dose to predict for survival with adjuvant radiation in resectable pancreatic cancer.

Abstract

240 Background: The median survival of patients with resectable pancreatic cancer remains low at approximately 2 years. A major limitation of treating patients with postoperative radiation therapy (RT) is the close proximity of the small bowel, which confines the RT dose. We hypothesized that a subset of patients might exist who would benefit from an escalated RT dose based on their underlying tumor radiosensitivity. Methods: Genomically-profiled patients with pancreatic cancer were identified from an IRB-approved prospective observational protocol. Patients treated with upfront surgery and postoperative RT were included. Briefly, the radiosensitivity index (RSI) is derived from the expression of 10 specific genes and a linear regression algorithm modeled on SF2 of 48 cancer cells. The RSI was combined with the delivered RT dose to derive a genomically adjusted RT dose (GAD). The GAD patient subsets were split at the middle GAD value and rounded up to the nearest integer (27). Our primary endpoint was to assess whether the GAD would predict for survival. Results: Forty patients underwent surgery and postoperative RT with GAD and clinical outcome data available. The median RT dose was 50.4 Gy (range 45-54) and the median follow-up among surviving patients was 68 months (range 42-141). Eighteen patients with a GAD > 27 had a median survival of 32.1 months, while 22 patients with a GAD < 27 had a median survival of 17.9 months (p = 0.48). On Cox multivariate analysis, both high-risk pancreatic cancer patients (positive margins, positive lymph nodes or a postoperative CA 19-9 > 90) and patients with a GAD < 27 had significantly decreased survival (Hazard ratio [HR] 5.0 [95%CI 1.8,13.6], p = 0.002 and HR 2.6 [1.1-6.0], p = 0.03, respectively). Among patients with a GAD < 27, 3 of 22 (14%) would have exceeded a GAD of 27 with an escalated RT dose of 54 Gy, and 8 of 22 (36%) patients would have exceeded a GAD of 27 with an escalated dose of 56 Gy to the pancreatic bed. Conclusions: GAD is predictive of survival among patients with resectable pancreatic cancer when combined with known prognostic factors. The GAD provides a tool to determine who might benefit from dose-escalated RT based on the underlying tumor radiosensitivity.