OBJECTIVETo investigate the correlation of radiation pneumonitis (RP) with standardized uptake value (SUV) for fluorodeoxyglucose (FDG) positron emission tomography and computed tomography (PET-CT) in lung cancer patients treated with radiation therapy.

METHODSFourty patients with unresectable non-small cell lung cancer (NSCLC) received FDG PET-CT before and after radiotherapy. The average SUV of the lung tissue irradiated with a dose of < or = 5 Gy, 5.1 approximately 15 Gy, 15.1 approximately 35 Gy, 35.1 approximately 60 Gy, >60 Gy were measured. The correlation between SUV and RP was analyzed by comparing the SUV in the patients with RP and without. The SUV ratio of the irradiated lung tissue to that of the non-irradiated lung tissue (L/B) was also calculated.

RESULTSOf the 40 patients, 8 developed RP, including 6 cases of grade 2 and 2 cases of grade 3. The SUV of irradiated lung tissues with a dose of 35.1 approximately 60 Gy was significantly correlated with RP. When SUV > or =1, the RP incidence rate was 41.7% versus 20.0% in the whole group, with a statistically significant difference. (chi2 = 3.96, P < 0.05). The sensitivity and specificity of SUV in predicting RP was 62.5% and 78.1%, respectively. When the value of L/B > or = 2.5, the RP incidence rate was 40.7% in this group versus 20.0% in the whole group, with a statistical significance (chi(2) = 4.92, P < 0.05). If taking L/B > or = 2.5 as a threshold value, the sensitivity and specificity in predicting RP was 72.7% and 90.9%, respectively. No statistically significant difference was found in predicting radiation pneumonitis between SUV > or =1 and L/B > or = 2.5 (chi2 = 0.002, P > 0.05).

CONCLUSIONThe standardized uptake value (SUV) and the SUV ratio of the irradiated lung tissue to that of the non-irradiated lung tissue (L/B) for FDG PET-CT are positively correlated with radiation pneumonitis, and clinicians may use it to predict the occurrence of radiation pneumonitis.