A Report Exploring the Influence of Abdominal Compression on Dosimetry of Adjacent Gastro-Intestinal Critical Structures and Toxicities for Patients Treated with Non-Hepatic Abdominal SBRT

Abstract

There is a concern that use of abdominal compression (AC) for respiratory motion management in abdominal SBRT might result in inconsistencies between predicted and delivered dose for nearby critical structures (CS) especially the gastrointestinal (GI) CS. The current study aimed at evaluating the impact of AC on the dosimetry of GI-CS and toxicity for patients treated with non-hepatic abdominal SBRT. Two sets of CT scans (planning CT scans with AC and pre-treatment diagnostic CT scans without AC) were analyzed for patient cohort, treated with AC with prescribed dose to planning target volume (PTV) ≥25 Gy/5-fractions. Target volumes were delineated on both scans, and PTV was expanded by 2 cm (PTV+2) and 4 cm (PTV+4). All GI-CS (duodenum, stomach, small bowel and large bowel) were contoured on both sets of scans. They were fused to create a composite CS (GI-lumen). Rigid registration of AC and non-AC scans were performed using commercial software, which allowed the transfer of GI-CS from non-AC to AC-CT scans, resulting in required dose information. The quality of the rigid registration was evaluated using Dice similarity coefficient (DSC). We calculated the volume of GI-CS within PTV+2 and PTV+4 and dose-volume parameters including V30Gy, and D0.2cc. Independent sample t-test was used for statistical comparison, and toxicity events were gathered from prospectively collected clinical data. A total of 12 patients met the DSC set criterion (DSC≥0.85). Primary targets included pancreas (n=3), retro-peritoneal nodal mass (n=6), adrenal gland mass (n=3). No difference was seen between the AC vs. non-AC with respect to volume of stomach (22.3 cc vs. 20.3 cc; p=0.88), duodenum (17.2 cc vs. 5.8 cc; p=0.1), small bowel (30.9 cc vs. 47.5 cc; p=0.37), large bowel (28.9 cc vs. 30.1 cc; p=0.95) and GI-lumen (120.6 cc vs. 107 cc; p=0.75) within PTV+2. Similarly, there was no difference in volume of any GI-CS within PTV+4: stomach (61.1 cc vs. 69.9 cc; p=0.77), duodenum (34.24 cc vs. 14.07 cc; p=0.13), small bowel (83.91 cc vs. 129.44 cc; p=0.26), large bowel (80.32 cc vs. 86.12 cc; p=0.88) and GI-lumen (258.78 cc vs. 297.32 cc; p=0.59) between AC vs non-AC respectively. There was a significant improvement in V30Gy of GI-lumen with AC (0.107cc vs. 4.9cc) (p=0.036). There was no difference in D0.2cc of GI-lumen (26.09 Gy vs. 30.44 Gy; p=0.20) or other GI-CS between AC vs. non-AC. Three patients had acute grade-1 anorexia, one patient had acute grade-2 gastritis. There was no grade ≥3 acute or chronic toxicity. Use of AC did not confer any dose inconsistencies in our study cohort. There was a significant improvement in V30 of GI-lumen with use of AC. Patients, originally treated using AC, completed their radiation treatment with minimal toxicity.