Medical Imaging

EPSRC Centre for Doctoral Training


215 - Molecular imaging tools to identify evolving cardiac injury caused by cancer therapy

1st Supervisor: Rick Southworth
2nd Supervisor: Tom Eykyn

Project overview:

Many chemotherapeutic agents are toxic to the heart, such that cancer survivors are at elevated risk of heart failure. There is often no apparent contractile dysfunction immediately post-chemotherapy, but the deterioration which eventually develops can be life-limiting and lethal. Despite this, cardiotoxicity is only monitored clinically by changes in cardiac contractility, by which time the injury has often progressed beyond meaningful intervention.

We are therefore developing radionuclide-based molecular imaging agents to target the biochemical changes caused by cardiotoxicity which precede contractile dysfunction, such as mitochondrial depolarisation, ROS generation, and hypoxia. We have an ongoing program of synthesizing and screening these agents in isolated perfused rat hearts. This PhD project represents the next step towards clinical evaluation by creating, characterising and validating clinically relevant rodent models of cardiotoxicity, and then using them to develop our prototype imaging agents into new approaches for diagnosing cardiotoxicity earlier than is currently possible.

Project Background:

Worldwide cancer detection and survival rates continue to improve. Ironically, with cancer survivors living longer, chronic cardiovascular disease arising as a direct result of cancer therapy is an increasing problem.  Cardiotoxicity can develop at any time, even decades after cancer therapy has ceased. During this latency period, there is often no evidence of contractile dysfunction, but the progressive deterioration in cardiac function which subsequently develops is life-limiting and frequently lethal. Despite this, the current techniques clinically used to evaluate cardiac injury during chemotherapy (echocardiography, MRI and MUGA scanning) evaluate changes in cardiac structure and contractility, which are likely too insensitive and too late in the pathology to provide the optimal window for intervention. There is therefore an urgent need for robust, well validated biomarkers of early cardiovascular complications arising from chemotherapy for follow-up surveillance in cancer patients.

Cardiotoxicity by anthracyclines is primarily mediated by reactive oxygen species (ROS) production and mitochondrial dysfunction, while radiotherapy and anti-angiogenic therapies like sunitinib may be cardiotoxic by inducing coronary microvascular injury and ischaemia, with likely involvement of mitochondrial dysfunction and ROS generation. In several parallel funded collaborations between an imaging biologist (Southworth), a physical chemist (Eykyn), a cardiovascular physiologist (Clark), a medical physicist (Livieratos) and several radiochemists (Yan, Arstad (UCL), Long, Blower), we are developing a platform of molecular imaging agents capable of visualising these biochemical targets with PET and SPECT (18F dihydroethidinium derivatives for ROS, 99mTc sestamibi and 68Ga lipophilic cations for mitochondrial membrane potential (m), and 64Cu bis(thiosemicarbazones) and 18F-pimonidazoles for hypoxia (Figure 1). These approaches have the potential to non-invasively identify cardiotoxicity far earlier than is currently possible clinically. Our tracer screening and validation program in isolated perfused rat hearts is well underway (Figure 1).

However, what we do not know is the timeline between chemotherapeutic treatment, expression of biomarkers of injury, and measurable changes in contractile dysfunction. In this project, we will characterise these time profiles, and determine how our novel imaging agents could be exploited to provide the earliest and most robust predictive index of evolving cardiotoxicity before contractile dysfunction manifests. It will be essential to understand what the cardiac retention of each of our imaging agents means in terms of the prevailing cardiac biochemistry, both for tracer validation, and providing biological context in terms of cardiac viability and prognosis.

In summary, this project aims to develop clinically relevant chronic rodent models of cardiotoxicity, and characterise them in terms of metabolic derangement, mitochondrial dysfunction, iron overload, ROS generation, microvascular injury and hypoxia. Cardiotoxicity is a continuum of injury; we will therefore characterise how these indices evolve over time. We will use this insight to guide and validate imaging experiments to demonstrate the capacity of our panel of novel imaging agents to predict cardiotoxicity before it manifests as contractile change, and to propose how each might be exploited in the clinic. This is an ambitious but highly achievable project, providing a student with a unique multidisciplinary education spanning cardiology, biochemistry, experimental design, radiochemistry and imaging.

Figure 1: Quantifying doxorubicin cardiotoxicity in vivo. Rats were treated with a single i.p dose of doxorubicin ranging from 0 to 10mg/kg. Progressive loss of contractile function and dilated cardiomyopathy were demonstrated by echocardiography (left); mitochondrial dysfunction was demonstrated both by loss of structural integrity and swelling (by electron microscopy) and loss of cardiac ATP content (centre); and a dose dependent loss of cardiac capacity to retain 99mTc sestamibi by dynamic planar SPECT imaging and biodistribution (right).