VIENNA, Austria — If doctors liked PET-CT, they are going to love PET-MRI.

And despite its high cost, doctors like PET-CT a lot, valuing the new insights into the biology of specific tumors offered by this technology that fuses metabolic processes shown by PET images layered over the anatomical references provided by with CT.

Clinical acceptance for PET-CT in cardiac imaging and oncology is running far ahead of the published scientific support and evidence that traditionally is required ahead of widespread adoption of a novel technology.

Now Philips Healthcare (Best, the Netherlands) is pushing on a fast-track for development of an even more innovative technology that promises to combine in routine practice these dynamic images of molecular activity rendered by PET with the anatomical imaging of MRI that is, beyond a clinical doubt, far superior to CT for depicting the details of the soft tissue environment where tumors grow, and rendering sharp images of the tumor itself.

At the European Congress of Radiology here, Philips was promoting this work-in-progress announcing that this month it will bring on-line the second of two beta-site installations for PET-MR that put the company on the leading edge of development in a race that includes both Siemens (Erlangen, Germany) and GE Healthcare (Little Chalfont, UK).

The objective for the Philips beta sites for the newly constructed Gemini True Flight PET/MR is to develop the scientific evidence ahead of engineering of the final commercial model to make the case for this novel technique and demonstrate clinical relevance.

The first site that has already imaged cardiac cases for four patients is at the Mount Sinai Medical Center (New York City) while the second site at the University Hospital of Geneva (Switzerland) certified the operations for the MRI scanner and expects to come online with the certification of the PET scanner by April, 2009.

Fast-forwarding three years when cases will be published, PET-MRI will begin a significant challenge to the leading role of PET-CT, potentially unseating this early champion of molecular imaging in short time due to the benefits of MRI.

The marginal quality of CT images is the first weakness.

More significant is that CT exposes a patient to radiation during every examination, and legal limits on exposure can limit the number of times a patient's progress with chemotherapy or cardiovascular medications can be followed up.

The issue becomes sharper when patients are participating in clinical trials where they are expected to undergo serial examinations.

MRI removes both these problems presenting superior, even the best-in-class images of soft tissue, and uses magnetic fields rather than radiation to acquire images.

There are two major problems with MRI to be solved by Philips engineers, however, before PET-MRI is ready for prime time in the clinic.

First, the powerful magnetic fields create a hostile environment for the vacuum photomultiplier tubes used in the PET gamma-ray detectors that reveal the metabolic processes of cells carrying radioactive tracer molecules.

The second issue is fusing the two images, to specifically locate the metabolic events on top of the right anatomical structure, which demands correction coding in software so complex it is worthy of its own branch of scientific papers.

In October, 2009 Philips reported a break-through at the Nuclear Science Symposium and Medical Imaging Conference in Orlando, Florida with a proof-of concept demonstration for a novel solid-state, scalable and compact digital detector technology.

This new digital silicon photomultiplier technology should enable Philips to replace the hundreds of tube-based gamma-ray detectors inside the powerful magnetic field of the MRI scanner to allow simultaneous acquisition of the two images for a near-instantaneous registration of the metabolic event and the precise location, or structure, where it took place.

Which begs the question of how Philips is able to install two beta site scanners if this technology is not yet available?

The short answer is that Philips has jury-rigged an existing Achieva 3 Tesla MRI scanner with a conventional PET scanner, creating the TF Gemini PET/MR, which is an investigational device and not available for commercial use in either the European Union or the U.S.

The longer explanation reveals a series of engineering feats that does Philips credit for innovation, and underlines the company's commitment in money and materials, for advancing the novel modality by generating evidence of clinical relevance.

The new solid state scanner is being tested on animals for simultaneous image acquisition in a Philip's-led upstream research program funded by the European Union, called HYPER Image. Meanwhile, the company has figured out how to bring the sensitive PET scanner close-enough to the pulsing magnet of the MRI to generate sequential images for fusion that are separated in time by a few minutes.

A first step is shielding the 420 gamma ray detectors of the PET scanner against the magnetic interference of the MRI, and also adding a second shield to prevent the gamma rays from affecting the performance of the MRI scanner.

The beta machine at the University of Geneva looks like two oversized doughnuts standing on end with a table between them.

The bore rings of the MRI and the PET scanner stand just three meters apart and the table is automated for positioning so that once an MRI scan of anatomy is completed, the table withdraws from the bore ring, turns the patient 180 degrees and inserts him or her into the PET scanner at a precisely programmed position to acquire the metabolic images.

The Chief of Nuclear Medicine at the University Hospital of Geneva, Osman Ratib, MD, pulled out his iPhone during an interviewto show a video of the Philips scanners arriving at 4 o'clock in the morning, saying, “And you can bet we had champagne, even at that hour.“

Philips provided a fully constructed examination room, wrapping it in a Faraday shield as well as a shield against X-rays, installing the two heavy scanners, and shipped the structure by special convoy up the Alps to Geneva.

“It was literally plug-and-play,“ exuded Ratib, a self-confessed gadget-guy, showing photos of himself on that morning holding three massive copper cables.

“It was fitted to an access door of the hospital, we plugged in these three cables, and it was ready,“ he said.

Where Cedar-Sinai is focusing on cardiac imaging applications, Ratib's group in Geneva is exploring clinical uses of the PET-MR scanning for oncology.

The two centers will meet in the field of plaque formations in coronary and carotid arteries where they expect to compare notes and publish, Ratib said.

“The neurologists are very keen to use this, too,“ he said, “but the brain is not as challenging for fusing PET and MRI because the cranium provides a fixed landscape for anatomical reference.“

“Our strength is oncology and we will be examining head-and-neck cancers, prostate and breast cancers where after a single chemotherapy session costing $10,000 we will be able to say definitively whether the treatment had any effect and should continue, or else recommend it is stopped,“ he said.

“Studies of tumors in a soft mass are critical, and today we have difficulties following patients to determine any recurrence after chemotherapy, surgery and especially radiotherapy because their anatomy has been changed,“ Ratib said.

“These tumors can be aggressive and our attention is especially sharp on the smallest structures because these are the tumors that will metastasize,“ he said.

“Putting together the PET and the MR scans the way we do it today, we cannot be sure whether the PET is showing us a muscle that moved or a lymph node of an aggressive tumor,“ he said.

“Our investigation at this point is not a formally designed study,“ he said.

“Instead, the goal is to demonstrate that reading two studies separately does not have anywhere near the diagnostic confidence as reading them fused,“ said Ratib.

According to Troy Haven, Philips senior manager for PET marketing, “We need to demonstrate clinical relevance first, because if there is not a clinical value to drive adoption, there is no point to incurring further costs and utilizing the people and resources of hospitals.“