Lung Dosimetry, Radiation Pneumonitis, and FDG Uptake in Patients with Stage II-III Non-Small Cell Lung Cancer Treated with Passively Scattered Proton Therapy or Intensity Modulated Pho
J. Yue,M. D. Jeter,T. Xu,J. Huo,T. Pan,Q. N. Nguyen,R. U. Komaki,D. R. Gomez,Z. Liao
DOI: https://doi.org/10.1016/j.ijrobp.2016.06.1774
2016-01-01
Abstract:Uptake of fluorodeoxyglucose (FDG) by normal lung tissue could reflect radiation dose-dependent inflammatory changes related to radiation pneumonitis (RP). To test this hypothesis, we evaluated lung dosimetric variables after passively scattered proton therapy (PSPT) or intensity-modulated photon therapy (IMRT), both with concurrent chemotherapy, for stage II-II non-small cell lung cancer (NSCLC) and assessed their potential relationships with FDG uptake and grade ≥2 RP. Subjects were 102 patients with stage II-IIIB inoperable NSCLC from a randomized clinical trial comparing PSPT and IMRT. RP was scored with the Common Terminology Criteria for Adverse Events v3. Positron emission tomography (PET) scans were obtained at 1-6 months after therapy as part of disease re-staging. Mean standardized uptake values (SUVmean) were calculated from the FDG-PET images of total bilateral lungs, including treatment-related parenchymal changes but excluding the tumor. Lung dose-volume variables (V5, V10, V20, V30, V40, V50, V60, V70, and mean lung dose) were compared between PSPT and IMRT and then examined, with RP grade and SUVmean in all patients. We further explored the ability of other variables (age, sex, tumor volume, histology, disease stage, smoking history, receipt of induction or adjuvant chemotherapy, radiation technique, and total delivered dose) to add predictive power. Potential associations between dosimetric variables, RP grade ≥2, and SUVmean were assessed using generalized linear models. No differences were found in mean lung dose, V10 or V20, RP grade, or SUVmean for patients treated with IMRT vs. PSPT. Compared with IMRT patients, PSPT patients had significantly lower V5 but had higher V30-70. In both univariate and multivariate analyses, V≥70 correlated with RP, and V≥30 correlated significantly with SUVmean. V5 did not correlate with either RP grade or SUVmean. No difference was found in normal lung dosimetric variables, RP, or FDG uptake between patients treated with PSPT versus IMRT. Even though PSPT produced lower V5 than IMRT, this advantage may be lost if high doses to smaller volumes of normal lung are not constrained because the volume of the lung received high dose radiation seems to play a more important role in the development of inflammation measured by FDG uptake in normal lung and the symptomatic RP.