3490 - Study of Relationship between Proton Beam Therapy to Liver Cancers and Persistent Pleural Effusion
Presenter(s)
S. Matsumoto1, H. Tatebe1, S. Asahi1, Y. Sato1, H. Tamamura1, K. Yamamoto1, K. Matsushita1, Y. Tameshige1, Y. Maeda1, M. Sasaki1, N. Fuwa2, A. Okubo3, S. Takamatsu3, and S. Kobayashi3; 1Fukui Prefectural Hospital Proton Therapy Center, Fukui, Japan, 2Proton therapy center, Central Japan International Medical Center, Gifu, Japan, 3Department of Radiotherapy, Kanazawa University Hospital, Kanazawa, Japan
Purpose/Objective(s): Persistent pleural effusions (PE) are often observed after proton beam therapy (PBT) for liver masses. PE after PBT is not clinically critical in most cases, but may become serious in a few cases. We hypothesized that the amount of irradiation dose to the diaphragm or pleura would be associated with the risk of persistent PE.
Materials/Methods: We studied 176 patients who underwent PBT irradiation to diaphragm and pleura for primary liver tumors at our institution with radiological follow-up exceeding one year. Patients with lung disorders such as pneumonia or metastases and those with massive ascites were excluded in order to select the patients suspected as PBT-induced PE. We analyzed patient characteristics and determined dose-volume histograms (DVHs) for the irradiated diaphragm and pleura, then estimated relationships between the risk of PE and various dose-volume parameters. The analyses were conducted using absolute dose and EQD2-equivalent dose (a/ß=3), respectively.
Results: Twenty-three patients (13.1%) developed PBT-induced persistent PE, With twenty-one patients of CTCAE grade 1 or 2, and only two (1.1%) classified as grade 3 or higher. Logistic regression analysis of predictors of PE based on patient background factors identified a significant correlation only with tumor size (p=0.007). In dose analysis, nonparametric tests revealed significant relationship in several parameters. In particular, irradiation dose to diaphragm showed primary relationship to the risk of PE in comparing with that to pleura. The results of receiver operating characteristic (ROC) curve analysis showed that the largest area under the curve value was the EQD2-equivalent diaphragm mean dose (0.67), with a cutoff value of 31.0 Cobalt-Gray-equivalents (GyE). Using the normal tissue complication probability (NTCP) models, the TD50 was calculated to be 75.6 ± 14.2 GyE. The volume of EQD2-equivalent 40 GyE irradiated in the diaphragm was also considered useful parameter to estimate PE risk.
Conclusion: Proton beam irradiation dose to diaphragm was considered to primarily relate to the risk of persistent PE. We calculated absolute values of cutoff value of EQD2 in ROC curve analysis and TD50 in NTCP models. To our knowledge, this is the first report linking persistent PE with diaphragmatic dose.