Image-Based Analysis of Dose-Dependent Radiation Effects on the Myocardium following Cardiac Radioablation Therapy

PROJECT SUMMARY/ABSTRACT The ability to treat ventricular tachycardia (VT) completely non-invasively would be a major paradigm shift in cardiovascular medicine. Sudden cardiac death due to VT is a significant public health problem and one of the leading causes of cardiovascular mortality. Implantable cardioverter defibrillators are effective in terminating VT, but do not prevent its occurrence. Although catheter ablation is recognized as a standard treatment for VT, the procedure is invasive, with risk of adverse events. Additionally, there are certain anatomical locations that are difficult to access with a catheter and it can be challenging to target mid-myocardial regions. In recent years, a novel non-invasive approach has emerged for the treatment of VT patients. Cardiac radioablation therapy has the potential to treat patients in an outpatient setting without the risks of an invasive procedure. Other advantages are that any anatomic location can be targeted and radiation can be delivered across the full thickness of the myocardium to create a transmural lesion. Our long-term goal is to treat VT patients completely non-invasively using radiation therapy safely and effectively. While initial pre-clinical and clinical studies have demonstrated promising results, there are still outstanding questions regarding this novel treatment approach. Importantly, the precise dose necessary to create transmural, homogenous lesions in myocardial tissue is not yet fully understood. Pre-clinical animal studies have demonstrated radiation induces fibrosis leading to lesions in myocardial tissue; however, studies have lacked the tools to precisely correlate in vivo radiation dose maps with ex vivo myocardial tissue location. The primary aim of the proposed work is to quantitatively correlate radiation dose with myocardial tissue fibrosis characterized by high-resolution ex vivo MR imaging and gross pathology tissue sections in an animal heart model. At Mayo Clinic, there is a rich database of irradiated swine that includes dose planning maps, in vivo imaging, high-resolution ex vivo late gadolinium enhanced MR (LGE-MR) imaging, and stained fixed heart specimens. The proposed study will utilize this existing database to accomplish the following two specific aims. In the first specific aim, a deformable image registration pipeline will be optimized and quantitatively validated to align radiation treatment dose maps to high-resolution ex vivo MR image volumes and gross pathology tissue sections. In the second specific aim, a detailed analysis of the relationship between radiation dose and myocardial lesion fibrosis will be conducted. Imaging metrics will be computed in the ex vivo high-resolution LGE-MRI including lesion volume, lesion transmurality, and lesion heterogeneity and correlated with incremental dose levels. Lesions in gross pathology will be manually traced in the digitized volume of fixed heart slices and correlated with dose levels. Results from this work will provide a critical next step in determining optimal radiation dose for treatment of VT patient with cardiac radioablation therapy.