Multi-Parametric Chest Magnetic Resonance Imaging for Assessment of Local Recurrence of Non-Small Cell Lung Cancer after Stereotactic Body Radiation Therapy

Abstract

Radiation induced lung fibrosis after lung SBRT include dense consolidation that may be dynamic over time and thus difficult to discern from local recurrence. We intended to evaluate the use of multi-parametric MRI to distinguish stereotactic Body Radiation Therapy (SBRT)- induced pulmonary fibrosis from local recurrence (LR). In this prospective study we included SBRT treated non-small cell lung cancer (NSCLC) patients, with suspected of local recurrence by conventional imaging. The patients were consented, and underwent MR imaging: T2 weighted, diffusion weighted imaging (DWI) dynamic contrast enhancement (DCE) and 5-minute delayed T1 weighted imaging after contrast injection. On the date of the MRI the LR status was unknown. LR status was defined as recurrence within high dose-radiation field, and confirmed by biopsy and/or subsequent growth on follow-up imaging following the next 12 months. The study included 18 patients with 20 irradiated lesions. MRI was conducted at a median of 23.5 months (range 6-55 months) from SBRT. Six lesions had MRI features suspicious for local recurrence that were confirmed by either biopsy (N = 3) or by follow-up imaging (N = 3) in the 12 months following the MRI. None of the lesions thought to be benign on MRI were shown to be malignant by biopsy (N = 1) or follow-up imaging (N = 13). All confirmed recurrences showed heterogeneous enhancement and had washout curves in the suspected focus which differed from the washout of the surrounding fibrosis. Enhancement of benign lesions was either homogenous or heterogeneous, but washout curves were similar in all regions. Apparent diffusion coefficient (ADC) mean values within the recurrence lesions was 1.21 10-3mm/s2 (range 0.98-1.7) which overlapped with the ADC range of fibrotic changes (mean 1.73 10-3mm/s2, range 0.98-2.6). In this pilot prospective study of NSCLS patients treated with lung SBRT, multi-parametric chest MRI was able to correctly support suspicion of LR and correctly minimize suspicion in the benign changes. Confirmed recurrences showed heterogeneous enhancement and had washout curves in the suspected focus which differed from the washout of the surrounding fibrosis. Still, we could not identify one parameter that would be diagnostic by itself. Although lower apparent diffusion coefficient values were seen in the recurrent lesions, no absolute cut-off value could differentiate recurrent tumor from fibrosis.