Last author

Tissue pH is a potent indicator of many disease states, including cancer, inflammation and infection. Yet there is no non-invasive way to image it in the clinic. Although magnetic resonance imaging is a powerful tool with which to visualize the body's soft tissue architecture, its sensitivity for imaging tissue chemistry has been low. Biochemist Kevin Brindle at Cancer Research UK's Cambridge Research Institute and his colleagues show on page 940 that they can markedly increase that sensitivity — and document the lower tissue pH associated with tumours in mice.

How does this technique image pH?

We knew tissue chemistry could be imaged if we could increase sensitivity. Normally, when atoms with a magnetic moment — such as carbon-13 — are put into a magnetic field, the 'spins' of their nuclei align themselves with it. But the interaction between the magnetic field and the nuclear spins is very weak, so the spins jostle each other, which causes some to fall out of line. This has a knock-on effect on sensitivity, and so image resolution. To increase sensitivity, we used a trick called hyperpolarization. We cooled carbon-13-labelled molecules together with stable radicals that had fully polarized electron spins to just above absolute zero; then transferred the polarization to the carbon-13 nuclei with microwave irradiation. We took advantage of the body's natural pH buffer, bicarbonate, and injected hyperpolarized carbon-13-labelled bicarbonate into mice. The labelled carbon dioxide produced from this in body tissues allowed us to calculate the pH.

Is hyperpolarization an established technique?

It was first described in 1953, but didn't become practical for imaging applications until 2003, when my industry-based co-authors figured out how to blast a sample with hot water to bring it from almost absolute zero to body temperature in a fraction of a second. Our collaboration led first to detecting labelled pyruvate as an indicator of treatment-induced tumour cell death. Being able to detect labelled bicarbonate may provide a generic method for imaging disease.

When might this technique make it to the clinic?

It's hard to say. If all goes well with a clinical trial GE Healthcare is planning for labelled pyruvate in 2009, one for bicarbonate could take place during the next few years.

What challenges does the technique face?

The polarization is very short lived, with a half-life of less than a minute. That means we'll have to polarize, inject and image very quickly. It's a challenge, but a solvable one.