Routine clinical imaging in oncology is largely based on imaging of tumor morphology using techniques such as Computed Tomography and Magnetic Resonance Imaging; these are used for the anatomical localization of the primary tumor and its metastases. Increasingly, functional imaging is being employed in addition to anatomical imaging, to detect processes such as tumor blood flow and diffusion of extracellular water. Newer techniques are enabling specific molecules and molecular interactions to be imaged at the tissue and cellular levels; dissolution Dynamic Nuclear Polarization (DNP) or hyperpolarized carbon-13 MRI is an example of a new molecular imaging technique that is been applied in this context within oncology [1].

Nearly a century ago it was observed that many cancers demonstrate high glucose consumption and lactate production compared to normal tissue, even in the presence of oxygen. This phenomenon of aerobic glycolysis was an early example of how tumor metabolism differs from normal tissue. Recent research has revealed more about the role of metabolism in cancer, showing that some rare cancers are directly linked to specific mutations in metabolic enzymes, oncogenes and tumor suppressors can directly increase nutrient uptake and alter metabolism, and tumor growth can be modified by altering metabolic activity. Positron Emission Tomography (PET) has been widely used for many years to image altered metabolism in patients with cancer, and the most frequently used PET tracer is a radiolabelled glucose analog, 18F-labelled fluorodeoxyglucose (FDG); although PET is very sensitive as a molecular imaging technique, it cannot discriminate the individual molecules within a metabolic pathway and exposes the patient to a significant radiation dose. In contrast, hyperpolarized carbon-13 MRI allows the in vivo detection of injected hyperpolarized 13C-labelled molecules and the products formed from them, therefore allowing metabolism to be probed non-invasively in real-time [2]. Furthermore, although MRI is several orders of magnitude less sensitive than PET (even when used in conjunction with DNP), it is free of ionizing radiation and therefore hyperpolarized MRI may have a role in patients where exposure to radiation poses a significant risk e.g. in children and women of reproductive age.

To date, a number of 13C-labelled probes have been used in conjunction with DNP to detect tumor metabolism. [1-13C]Pyruvate has several chemical, biological and physical properties which make it suitable for hyperpolarization such as its rapid metabolism into lactate; pyruvate has become the leading hyperpolarized molecule both pre-clinically and clinically. Changes in the exchange of hyperpolarized [1-13C]pyruvate into [1-13C]lactate have been used to detect and grade tumors pre-clinically [3,4]. A reduction in the measured hyperpolarized lactate following the injection of pyruvate has been used as a very early marker of successful response to chemotherapy [5]; metabolic responses to therapy are likely to occur more rapidly than the changes in tumor size which are traditionally used to measure response to treatment. Furthermore, targeted therapies are increasingly being used in oncology, many of which do not demonstrate significant changes in tumor size despite a clinical response to therapy. Therefore imaging methods to detect changes in metabolism could be clinically important and this is an area where hyperpolarized carbon-13 MRI could play a major role in the future. For example, inhibition of the phosphatidylinositol 3-kinase pathway has been shown to correlate with a drop in hyperpolarized [1-13C]lactate levels in a number of tumor models [6].

Hyperpolarized molecules other than pyruvate also offer promise for imaging in oncology; examples include hyperpolarized [1,4-13C2]fumarate for detecting necrosis (where there is an increase in the measuring malate signal seen following successful treatment and induction of necrosis), 13C-labelled bicarbonate for imaging tumor pH, [1-13C]lactate as an alternative method for measuring lactate dehydrogenase, and hyperpolarized amino acids and sugars to detect changes in uptake and metabolism that occur in cancer [1,7]. Hyperpolarized carbon-13 MR probes may also be used to study important aspects of tumor biology, such as the relationship between carbonic anhydrase activity and pH [8].

Hyperpolarized carbon-13 MRI may also play a significant role in cardiac imaging. Alterations in cardiac metabolism are now considered one of the causes of cardiac disease, rather than a consequence, and changes in pyruvate metabolism have been demonstrated in diabetes, ischemic heart disease, cardiac hypertrophy and heart failure [9]. The technique may also have a role in neurological diseases where an elevation in tissue lactate is a feature of underlying pathology such as in stroke, inflammation and brain tumors.

The first human trials of DNP has been undertaken in prostate cancer using hyperpolarized [1-13C]pyruvate [10]. This demonstrated not only the feasibility of the technique, but also the possibility that metabolic changes within a tumor may occur despite normal conventional imaging appearances. A number of sites around the world have commenced human studies of hyperpolarized carbon-13 MRI using a clinical hyperpolarizer. The majority of these early studies will probe the use of the method in oncology and cardiology. There is currently debate on the best approaches to both image the metabolism of hyperpolarized pyruvate, as well as the most appropriate methods to analyze the data produced [11].

Although the translation of hyperpolarized carbon-13 MRI to the clinic faces many challenges, it has the potential to aid diagnosis, identify disease heterogeneity, predict disease outcome, help target biopsies and determine treatment response non-invasively. Combining DNP with existing techniques, such as diffusion-weighted imaging and Positron Emission Tomography, could provide new insights into tumor biology.