Anatomically Restricted Dose Escalated SBRT to the Pancreas Effects High Local Control and Rapid and Substantial Pain Relief

Abstract

Pancreatic cancer (PC) remains a recalcitrant cancer with a dismal probability of long-term survival. While most patients (pts) ultimately develop distant metastases (DM), local progression nonetheless profoundly affects quality of life. Despite this, local treatments like radiotherapy have been de-emphasized due to the perceived primacy of systemic control. SBRT is a convenient, effective, and safe treatment strategy for most pts. However, its use remains limited due to proximity of the tumor to sensitive organs at risk (OAR). We used an anatomically restricted dose painting and escalation strategy in attempts to effect improved local control (LC) while minimizing toxicity in OAR. All pts treated with SBRT for PC from 2010 to 2017 were reviewed. A vast majority were treated with IMRT using a simultaneous integrated boost (SIB) dose painting technique. PTVs included a SIB PTV1 dose within a standard PTV2. The former was created by subtraction of the OARs (duodenum, stomach, or bowel) +5 mm from the PTV2. Dose per fraction was individualized on the basis of normal tissue constraints. Treatment plans were assessed for prescription isodose line (IDL), plus GTV, PTV1, and PTV2 mean doses. The BED of the GTV mean doses were calculated. Acute/late toxicity was documented by CTCAE v4.0 criteria. Overall survival (OS) and local progression free survival (LPFS) was assessed by Kaplan-Meier analysis. Abdominal pain was scored from 0-10 and documented at initial consult and 6 weeks and 3 months post SBRT. Change in narcotic usage was inventoried as well. Of the 60 pts treated on the study, 7 had local recurrence only after surgery, 31 were locally advanced, 10 medically inoperable, and 12 had DM. Median follow up was 7.33 months. Pain was reduced at 6 weeks (p<0.001) and 3 months (p<0.0001), with significant reductions in narcotic usage. Average pain score was 8.3 prior to SBRT and 0.97 at 3 months post SBRT. The median prescription dose to PTV2 was 30 Gy in 5 fx (25-40 Gy). The mean prescription IDL was 79.7%. The median GTV, PTV1, and PTV2 mean dose was 39 Gy, 39 Gy, and 37 Gy, respectively. The median BED to the GTV was 68 Gy [38-105 Gy], with 32% having a BED of >74 Gy. One year LPFS was 84.8%, with no failures in pts receiving >74 Gy BED. 45% and 53% received chemotherapy (CT) before and after SBRT, respectively. LPFS was not affected by receipt of CT after SBRT (p=0.37). One year OS following SBRT was 46.8%. CT after SBRT was associated with better OS (p=0.05). One acute toxicity >=grade 3 and 5 >= grade 3 (4 grade 3; 1 grade 5) late toxicities occurred. We implemented an anatomically restricted dose painting and escalation strategy that effects rapid and effective pain relief, high LPFS rates and acceptable toxicity. Our results suggest that inhomogeneous dosing with SIB may confer an advantage over conventional homogeneity across a singular PTV in pancreas SBRT.