Dosimetric Correlations With Toxicity In The First Experience Validating Adjuvant Fractionated Stereotactic Body Radiation Therapy For Resectable Pancreatic Cancer

We previously generated a systematic map of local failures (LF) in patients with resectable pancreatic cancer (R-PCA) following pancreaticoduodenectomy (PD) with reference to the superior mesentery and celiac arteries (SMA and CA). Using this plot outlining high-risk areas of LF we proposed a radiation treatment field based on expansions of the SMA and CA that has since been incorporated in an ongoing clinical trial for R-PCA. We report preliminary outcomes of this the first experience validating adjuvant stereotactic body radiation therapy (SBRT) and provide dosimetric correlations of observed toxicities. Patients with R-PCA with R0 or R1 PD were enrolled in a clinical trial to estimate the safety of a whole cell vaccine (GVAX) with low-dose cytoxan (Cy) followed by SBRT and 6 cycles of FOLFIRINOX (FFX). The first 3 patients received SBRT and FFX only. Due to toxicity, FFX was modified to drop the 5-FU bolus (mFFX) and the next 4 patients received SBRT +mFFX. The final 13 patients received Cy/GVAX + SBRT + mFFX + Cy/GVAX (4 doses, 1 month apart). SBRT targeted the tumor bed and expansions of the CA and SMA contoured 1 cm and 3 cm respectively from their origins at the aorta. Patients received a total of 33Gy in 6.6 Gy to a smaller planning target volume (PTV33) contained in a larger planning volume receiving 25Gy in 5Gy fractions (PTV25) to at least 90% of the PTVs with one exception detailed below. Of 19 patients, 15 (78%) had close/positive margin resections and 15 (78%) had nodal involvement. All completed SBRT within 12 months of PD except 1 patient with surgical site dehiscence resulting in 1 month delay of SBRT. Median PTV33 was 117 cm (range 57-195) with median 91.3% coverage. Median PTV25 was 166 cm (range 101-246) with median 95.9% coverage. The patient with the largest PTV33 required dose reduction to 30Gy in 6Gy fractions to adhere to predetermined dose constraints. During SBRT, one patient reported grade 2 fatigue and pain. 95.5% of the PTV33 of this patient received 33 Gy. The remaining patients reported grade 1/2 or no toxicity with SBRT. There was one unplanned break during SBRT due to machine malfunction. One patient treated with 95.8% coverage of PTV33 experienced grade 3 pancreatitis 8 weeks post-SBRT. One patient with 98.3% coverage of PTV25 experienced grade 3 partial small bowel obstruction. Both patients improved without surgical intervention. Of the 11 patients with 12-month follow-up, 3 (27%) had local recurrences (all positive or close margins). Preliminary data is growing to support adjuvant SBRT. Future PTV dosimetric constraints may be tailored based on this experience.