Over the past decade, PET has established its diagnostic credentials in clinical oncology, both for pinpointing disease (and recurrent disease) and for monitoring the efficacy or otherwise of drug, radiation and combination therapies. Elsewhere, PET has carved out a clinical niche in brain imaging, where it can help to diagnose Parkinson's disease and distinguish various types of dementia.

For the oncologist, however, PET scanning with the radioisotope 2-fluoro-2-deoxyglucose (18F-FDG) is valued for its ability to highlight cancers, particularly in the colon and lungs. The sugar-based 18F-FDG tracer is metabolized far more rapidly by cancer cells than by normal cells, and it is drawn towards areas of malignancy in high concentrations. This metabolic overactivity is reflected as "hot spots" on PET images, enabling doctors to confirm the status of suspect tumours and assess whether they have spread.

Similarly, CT also assists in cancer imaging by providing a detailed anatomical framework in which to locate tumours. Merging PET and CT imaging information therefore provides the best of both worlds, enabling the accurate localization of metabolic activity in an anatomical framework. For the medical equipment makers, this has been a win-win scenario since the first combination PET/CT devices hit the market in 2001. Six years on and nearly all-commercial PET scanners are sold as part of combined PET/CT systems.

"The development of combined PET/CT scanners has allowed major expansion of molecular imaging," wrote Michael Welch, co-director of the Division of Radiological Sciences at the Washington University School of Medicine (St Louis), in the same special issue of Journal of Nuclear Medicine. As always, though, there's a price to be paid. Specifically, the addition of CT to PET means an increased radiation dose for the patient. "In animal imaging," Welch continued, "where research has been carried out on combined PET/CT scanners and combined SPECT/CT, the radiation dose is significant. [This means that] the development of a truly integrated MR/PET scanner would be a tremendous advantage."

The rationale is clear enough: while the PET element of the scan would deliver a modest dose of ionizing radiation, the MRI part would be radiation-free. Researchers are already on the case, even if the fusion of PET and MRI hardware is complicated by some essential incompatibilities (see PET/MRI: from theory into practice on medicalphysicsweb). For starters, the photomultiplier tubes used to detect and quantify light produced by PET scintillator elements simply will not work in a magnetic field. Undaunted, multidisciplinary teams of physicists, engineers and clinical scientists are coming up with innovative workarounds to this design challenge, with those same projects yielding the first early-stage studies using combined PET/MRI.

Technology issues notwithstanding, an even more fundamental consideration is the clinical potential (or otherwise) of PET/MRI. Put another way: will anybody outside the research lab want the scanners? It's a racing certainty that a combined PET/MRI product will come with a heavyweight price tag. As such, the clinical community will need to be doubly convinced that the combined modalities will make a significant, positive difference to patient management.

Assuming that PET/MRI one day ticks the boxes on price:performance, clinical benefit and all the rest – and that's a big assumption – it will then fall to the equipment makers and their marketeers to sell the idea to the community. Make no mistake, that's going to require at least as much creativity as the technology innovation and product development that goes before it.