UVA Department of Radiation Oncology Charlottesville, VA
K. Walker1, S. DSilva1, M. Conaway2, Y. Chavis1, D. F. Leach1, G. L. Andrade de Sousa3, C. Nguyen4, D. Liyanage5, T. C. Jones1, M. S. Peach1, E. M. M. Janowski1, and K. Wijesooriya6; 1University of Virginia, Department of Radiation Oncology, Charlottesville, VA, 2University of Virginia, Department of Public Health Sciences, Division of Translational Research and Applied Statistics, Charlottesville, VA, 3Department of Physics, University of Virginia, Charlottesville, VA, 4University of Virginia, Department of Physics, Charlottesville, VA, 5Albermarle High School, Charlottesville, VA, 6University of Virginia, Department of Radiation Oncology, Department of Physics, Charlottesville, VA
Purpose/Objective(s): Radiation Therapy (RT) induced immune suppression (RIIS) correlates with worsened overall survival in pancreatic cancer patients. However, there is no computational model to predict RIIS. We evaluated lymphocyte reduction from RT in pancreatic cancer as a function of dose volume data to abdominal regional nodes and critical structures, hypothesizing that there are differential degrees of RIIS among these critical structures. Materials/
Methods: We analyzed 69 patients who received radiation therapy to the pancreas, including 63 patients who received concurrent chemotherapy. The abdominal CT scans used for planning each patient’s radiation dose were obtained, and structures of interest were manually contoured, including the heart, inferior vena cava, aorta, celiac axis, superior mesenteric artery, liver, duodenum (when present), bowel bag, kidneys, portal vein, spleen, stomach, and vertebral bodies per the RTOG contouring atlas. To identify regional lymph node basins, expansions were performed on the aorta, superior mesenteric artery, celiac trunk and portal vein. Lymphocyte information was obtained from these patients prior to treatment and at 3-month intervals. For each patient and critical structure, integral dose, dose volumes: V1-V50, PTV volume, and dose fractionation were extracted, along with degree of radiation exposure. Associations between RIIS and dose administered to each structure of interest were then assessed using Spearman rank correlations. Data on overall survival were analyzed using Cox-regression, and chi-squared tests. Results: Maximum lymphocyte reduction was found at day 35 following initiation of RT, which represented a 78.4% reduction from Pre-Tx ALC value, including 81.2% with = 3 lymphopenia. This reduction remained 65% of pretreatment values at day 185. Number of fractions of RT correlated with RIIS. In addition, significant correlations were found for spleen (mean-dose, integral-dose, V1-V10), liver (V1-V10), R Kidney (integral dose, V1-V15), duodenum (V15-V40), lymph-node-stations (V1-V30), external-PTV (V1-V10), and PTV volume. Acute-long-term-RIIS correlated with right kidney (V20-V30), and duodenum (integral-dose, V1-V40). On multivariate cox regression analysis, minimum post treatment lymphocyte count < 0.2 (G4 lymphopenia) correlated with OS, though it did not reach statistical significance (p=0.15). Conclusion: Our data indicate that acute RIIS correlates with dose to spleen, liver, kidneys, and duodenum and prolonged RIIS correlates with dose to right kidney and duodenum. Optimizing treatment planning to limit doses to these structures may improve outcomes in pancreatic cancer patients. Future studies are indicated to investigate the role of this hypothesis generating work.
