3462 - Is Stereotactic Body Radiation Therapy Associated with Accelerated Pulmonary Fibrosis in Patients with Interstitial Lung Disease? A Comprehensive Radiologic Analysis
Presenter(s)
S. Emam1, J. Adamson1, J. Christensen2, D. Neish1, A. Kumar1, A. L. Pomeroy1,3, J. Liu1, T. C. Mullikin1, and C. R. Kelsey4; 1Duke University Medical Center, Department of Radiation Oncology, Durham, NC, 2Duke University Medical Center, Department of Radiology, Durham, NC, 3Virginia Commonwealth University, Richmond, VA, 4Department of Radiation Oncology, Duke University Medical Center, Durham, NC
Purpose/Objective(s): Patients with interstitial lung disease (ILD), particularly idiopathic pulmonary fibrosis (IPF), are at increased risk of pulmonary toxicity following thoracic radiation therapy (RT). Compared with patients without ILD, these patients experience a markedly higher incidence of radiation pneumonitis after stereotactic body radiation therapy (SBRT), which is occasionally fatal. Acute exacerbations of pulmonary fibrosis beyond the RT field have also been reported. We sought to objectively quantify changes in fibrotic burden outside the RT field before and after SBRT in patients with ILD.
Materials/Methods: A thoracic radiologist (JC), blinded to subsequent clinical outcomes, reviewed pre-SBRT CT scans from 139 consecutive patients treated with SBRT from 2010–2017 for stage I NSCLC to identify ILD using established imaging criteria. For patients with ILD, pre- and post-SBRT CT scans obtained up to 3 years before and after treatment were deformably registered to the planning CT, with dose transfer performed using MIM Software Inc. (Cleveland, OH). Lung texture analysis was performed using Coreline Soft AView Lung Texture software (Suwanee, GA, USA), segmenting lung parenchyma into normal lung, ground-glass opacity, reticulation, honeycombing, consolidation, and emphysema. Fibrotic burden was defined as the combined proportion of reticulation and honeycombing, excluding lung volumes receiving =20 Gy. Linear mixed-effects modeling assessed pre- to post-SBRT changes, accounting for within-patient correlation. A trajectory analysis evaluated slope differences in patients with =2 scans per period. Pulmonary toxicity was graded per CTCAE v6.0.
Results: Of 139 patients, 33 (24%) had radiologic ILD. One patient was excluded due to prior lung RT and eight due to incomplete imaging, leaving 24 evaluable patients. Five had a formal ILD diagnosis (non-specific interstitial pneumonitis in 3; IPF in 2). There were two grade 3/4 pulmonary toxicities and five grade 5 pulmonary events (median 193 days post-SBRT; range, 109–257), all seemingly precipitated by SBRT. Mean fibrotic burden outside the RT field increased from 1.97% pre-SBRT to 3.89% post-SBRT (p<0.001). This increase was more pronounced in patients with a formal ILD diagnosis (9.96%) compared with those without (0.96%) (p<0.001). In patients with sufficient imaging for trajectory analyses (n=10), fibrotic burden outside the RT field was stable pre-SBRT (-0.27%/year, 95% CI -1.18% to 0.65%, p=0.570) but increased post-SBRT (1.75%/year, 95% CI 1.23%–2.27%, p<0.001).
Conclusion: This unique and objective radiologic analysis demonstrates that SBRT is associated with accelerated fibrosis outside the RT field in patients with ILD, particularly those with a formal clinical diagnosis. Further work is needed to better define SBRT-related risks across ILD subtypes and imaging phenotypes.