Abstract
51
Objectives: Cardiotoxicity is a common complication of doxorubicin-based chemotherapy, limiting the use of this otherwise effective chemotherapeutic.1 Current approaches to monitoring patients for signs of doxorubicin induced cardiotoxicity (DIC) primarily rely on assessment of left ventricular function, whose decline usually represents late and often irreversible stages of the disease.2 Therefore, new imaging approaches for early detection of DIC are direly needed. While the mechanisms of DIC are incompletely understood, a key role has been attributed to mitochondrial damage and dysfunction.3 Our group has recently introduced 18F-triphenylphosphonium (18F-TPP+) PET imaging for non-invasive measurement of total cardiac membrane potential (ΔΨT), a proxy of mitochondrial membrane potential (ΔΨM).4 ΔΨM represents a comprehensive index of mitochondrial function and ΔΨT mapping could therefore serve as an early biomarker of DIC. This study aimed at assessing whether an acute cardiotoxic effect of doxorubicin could be detected by ΔΨT mapping in a swine model.
Methods: In total, three male Yucatan pigs (49 to 56 kg) were studied. Prior to PET imaging, a 24G catheter was surgically placed into the left anterior descending artery (LAD) without impeding blood flow. After animal preparation, 18F-TPP+ was administered via the auricular vein using a bolus-plus-infusion schedule with Kbol=600 min, consisting of a ~670 MBq bolus followed by ~205 MBq infused over 180 minutes. Dynamic PET imaging over 180 minutes was performed using a hybrid PET/CT system (GE Discovery MI). After reaching secular equilibrium in tissue (60 - 90 minutes after bolus and infusion start), 0.5 ml/kg of saline (control) were infused over 10 minutes into the LAD, followed by two doxorubicin infusions (2 mg/kg dissolved in 0.5 ml/kg saline) started 10 and 50 minutes after the end of the control saline infusion and lasting 10 and 5 minutes, respectively. Arterial blood samples were drawn from the femoral artery every 5 minutes starting from the intracoronary infusions. At the end of the scan, the animals were euthanized. Dynamic PET images were transformed into short axis orientation and time activity curves (TACs) of tracer concentration were extracted in all 17 segments of the left ventricle. The volume of distribution (VT) was calculated as the ratio of 18F-TPP+ concentration in tissue over concentration in plasma and related to ΔΨT through the Nernst equation, taking the extracellular space and mitochondrial volume fraction into account.4
Results: For all animals, a local, acute drop of VT was observed during or following the doxorubicin infusions, corresponding to a depolarization of ΔΨT of approximately 8 mV in the LAD territory, i.e., the areas distal to the catheter, whereas no changes were measured for the segments not exposed to the doxorubicin infusions nor for any regions for the control saline infusion. Parametric images of VT and ΔΨT and TACs for four segments of one example case are shown in Fig. 1 and 2.
Conclusions: Intra-coronary infusion of doxorubicin was associated with an acute, local depolarization of ΔΨT quantified using 18F-TPP+ PET. Future studies will aim at assessing the potential of ΔΨT mappingas a biomarker for early detection of DIC in a chronic setting. Research Support: P41EB022544, R01HL137230, S10OD018035 Acknowledgments: We thank Marina T. Macdonald-Soccorso, Helen Deng, Eric J. McDonald, Brittan A. Morris, and Julia-Ann Scotton for their essential contribution to this project.