Risk factors and dosimetric limitations for acute radiation pneumonitis after gemcitabine induction chemotherapy for non-small cell lung cancer

Objective: To summarize the incidence of acute radiation pneumonitis (ARP) after gemcitabine induction chemotherapy for non-small cell lung cancer (NSCLC) and identify the high risk factors and dosimetric limitations of ARP after gemcitabine induction chemotherapy. Methods: We retrospectively analyzed 191 NSCLC cases who were received gemcitabine induction chemotherapy and chest radiotherapy in radiotherapy department of Zhejiang Cancer Hospital between January 2010 and December 2010. Base line data, treatment information and the incidence of ARP after treatment were collected. The risk factors of ARP were analyzed with univariate and multivariate Logistic regression method. Results: A total of 49 patients developed ≥ grade II ARP, accounting for 25.7% of all cases. Univariate analysis indicated that the probability of ARP in patients who received the cumulative dose of gemcitabine ≥ 9.0 g was 3.45 times higher than that in those treated at a dose of < 9.0 g (P = 0.015). Radiation dose ≥ 50 Gy was significantly correlated with the occurrence of ARP (P = 0.008). The risk of ARP was increased by 7.69 times if the time interval between radiotherapy and chemotherapy was within 10 weeks (P = 0.047). Among the dosimetric parameters, V_5Gy, V_20Gy, V_30Gy and mean lung dose (MLD) of bilateral lungs were 45%, 22%, 16%, and 1200 cGy respectively. All of them could effectively predict the occurrence of ARP (all P ≤ 0.001). Multivariate analysis indicated that only radiotherapy dose (P = 0.044) and V_5Gy (P = 0.02) were the independent predictors of ARP. Conclusions: For NSCLC patients who receive gemcitabine induction chemotherapy, the cumulative dose of gemcitabine, the interval time between chemotherapy and radiotherapy and the radiation dose are associated with the occurrence of ARP. We should strictly limit the total lung dosimetric parameters, such as V_5Gy, V_20Gy, V_30Gy and MLD to reduce the incidence of ARP. (authors)