Patients with similar tumour types frequently show different responses to the same therapy. The development of new treatments would benefit, therefore, from imaging methods that allow an early assessment of treatment response in individual patients, allowing rapid selection of the most effective treatment. We have been using (13)C MRSI (magnetic resonance spectroscopic imaging) of tumour cell metabolism, using hyperpolarized (13)C-labelled cellular metabolites, to detect treatment response. Nuclear spin hyperpolarization can increase sensitivity in the magnetic resonance experiment >10,000 times, allowing us to image labelled cell substrates in vivo and their subsequent metabolism. We showed that exchange of hyperpolarized (13)C label between lactate and pyruvate, catalysed by lactate dehydrogenase, was decreased in treated tumours undergoing drug-induced cell death, and that tissue pH could be imaged from the ratio of the signal intensities of hyperpolarized H(13)CO(3)(-) and (13)CO(2) following intravenous injection of hyperpolarized H(13)CO(3). Tumour cell glutaminase activity, a potential measure of cell proliferation, can be determined using hyperpolarized [5-(13)C]glutamine, and treatment-induced tumour cell necrosis can be imaged in vivo from measurements of the conversion of hyperpolarized [1,4-(13)C(2)]fumarate into malate. Since these substrates are endogenous and, in some cases, have already been safely infused into patients, these techniques have the potential to translate to the clinic.