Evaluation of pulmonary toxicity and efficacy of a radioprotector using MRI

Abstract

18506 Background: Radiation induced pulmonary symptoms occur in 20% of patients who receive radiation treatment for cancer of the lung or breast. Presently, there is no clinical test available to predict symptomatic toxicity or to assess the extent of radiation induced lung injury early after treatment. We report the use of time-resolved magnetic resonance angiography (Tr-MRA) to assess the efficacy of a radioprotector in limiting radiation toxicity in a rabbit model of stereotactic radiation therapy (SRT) pulmonary injury. Methods: 10 rabbits were used, of these 4 received different radiation doses in 3 equal fractions of 7, 9, 11 or 13 Gy within a week and were intended to determine the minimum radiation dose capable of producing detectable injury using our MRA technique. Effects were measurable by MRA at the minimum dose of 9 Gy × 3 fractions. We irradiated the remaining rabbits with 3 × 11 Gy. Four of these animals received 50mg/Kg of the radioprotector amifostine, 20 min. before each SRT dose to the right lung delivered on a TomoTherapy scanner. Tr-MRA was performed using a 1.5T clinical MR scanner at baseline, 4, 8 weeks post SRT. A 3D FLASH sequence was used with 6 slices; resolution 1.4 × 1.1 × 12mm3; TR/TE = 3.3/1.2 ms; total time = 32 sec. A 3 cc injection of a Gd chelate was delivered after 2 sec. of the start. Perfusion changes were calculated from the subtracted MIP images and correlated over the multiple time points. Results: The mean baseline areas of perfused tissue were 12±1.6 cm2 (mean±STD) for the right lung (RL) and 9±1.1 cm2 for the left lung (LL). At 4 weeks post treatment, the three control rabbits (3 × 11 Gy and no radioprotector) had mean perfused areas of 9±2.1 cm2 for the RL and 10±0.5 cm2 for the LL. At the same time point, the animals that received the same radiation dose plus the radioprotector had mean perfused areas of 12±1.2 cm2 for RL and 10±0.3 cm2 for the LL. At 8 weeks post treatment, the control rabbits had obvious perfusion defects in the irradiated zone, with mean perfused areas of 7±2.3 cm2 (RL) and 9±0.1 cm2 (LL). At this time point the animals that received the radioprotector had mean values of 10±0.8 cm2 (RL) and 9±0.6 cm2 (LL). Conclusions: Tr-MRA detected radiation induced pulmonary injury 4 weeks post treatment and was able to detect reduced early treatment toxicity as a result of a clinically used radioprotector. No significant financial relationships to disclose.