New signs of life as an era of private production dawns for SPECT’s workhorse isotope

An estimated 80 percent of the world’s nuclear medicine – or approximately 70,000 SPECT scans per day – are contingent upon access to an isotope traditionally generated in nuclear reactors. For the Western Hemisphere, the National Research Universal (NRU) reactor at Chalk River in Ontario, Canada, has been the primary supplier of molybdenum-99 (Mo-99), the parent isotope of technetium (Tc-99m), since 1957.

In a couple of years, all that will change.

Six years ago, when the NRU reactor went out of operation at the same time as another major reactor – the High Flux Reactor (HFR) in Petten, Netherlands, the potential for supply disruptions of Tc-99m became frighteningly clear. The NRU reactor is retiring in 2018, meanwhile the HFR, and another reactor in France called Osiris, are timelined to go permanently offline as well, with replacement nuclear reactors in various stages of development.

From the time of those shutdowns, Rob Atcher, chair of the Medical Isotope Taskforce for the SNMMI, says the price of Tc-99m has gone up 200 to 300 percent. In North America government funded initiatives were put in motion to find alternative methods of generating those isotopes. Different approaches to producing the isotopes are attractive in part because reactor-generated Mo-99 requires highly-enriched uranium, which is a weapons-grade substance.

Approximately half of the Tc-99m used for SPECT imaging measures blood flow to the heart on patients with chest pain or other symptoms indicating a block in the cardiac arteries. The remaining procedures use Tc-99m to detect bone metastases from the spread of breast, prostate, and lung cancer.

As access to Tc-99m remains uncertain, some experts are turning to innovations with PET imaging to absorb some of the diagnostic burden. Sodium fluoride has been shown effective for detecting bone metastases, but is it still too soon to realistically imagine a future where molecular imaging doesn’t depend on Tc-99m at all?

Linac vs. cyclotron
Australia recently poured $157 million into the processing facility of its Open Pool Australian Lightwater (OPAL) reactor, which already produced 550,000 doses of Mo-99 per year going into the upgrade; enough to meet the country’s own needs. Slated to open in 2016, the improved facility is anticipated to triple that output and establish Australia as a major international supplier of the isotope with an output comparable to Canada’s NRU.

Unfortunately for countries in the Western Hemisphere, Mo-99 has a half-life of about 66 hours, and Tc-99m has a half-life of only six hours – rendering Australia’s supply of diminished use to the Americas. According to Atcher, roughly 25 percent of the isotope would be lost during that 24-hour transit.

Leon Zebrick, director of molecular imaging at Ochsner Health System in the U.S., says he and his colleagues have long understood the perils of depending on third parties for radiopharmaceuticals. Previous shortages of other isotopes informed their decision to invest in their own cyclotron. With it, they will not only be able to produce a variety of PET isotopes, they could potentially generate Tc-99m as well.

“The U.S. government and the researchers they’re funding have hung their hat on solution A, but the Canadian government has hung its hat on solution B,” says Zebrick, referring to linear accelerator based methods versus cyclotron methods, respectively.

In Canada, a program called the Non-reactor- based Isotope Supply Contribution Program (NISP) was created to fund and encourage participation in the market, and several companies are trying to fill the void that will be left by Chalk River. Advanced Cyclotron Systems, the company that will manufacture the cyclotron at Ochsner, is one of them.

“On a cyclotron you bombard Mo-100, not Mo-99, with proton beams and synthesize Tc-99m from that,” says Zebrick, who contrasts that process with the linear accelerator one, which still aims to produce the parent isotope, Mo-99.

Each method has its advantages. Although manufacturing Mo-99 means a longer half life and more opportunities to take the isotope to market, “we didn’t go that route because linacs are not so good at producing workhorse PET isotopes,” says Zebrick. “We need a cyclotron, we knew that, and if we can make Tc-99m on it – all the better.”

Like NISP in Canada, the U.S. government introduced a bill in 2009 called the American Medical Isotopes Production Act with the goal of building interest in isotope production. In support of that legislation, the Department of Energy’s National Nuclear Security Administration (DOE/NNSA) works with commercial entities to support development of new moly-99 sources.

The DOE/NNSA website states that under their cooperative agreements they split costs fifty-fifty to a maximum of $25 million per partnership. The organization also cites two companies, NorthStar Medical Radioisotopes and SHINE Medical Technologies, as currently making developmental headway with linear accelerators.

NorthStar is developing two technologies for generating moly-99 out of the stable isotope Mo-98; a “neutron capture” process utilizing the University of Missouri Research Reactor (MURR) in Columbia, Missouri, and an electron accelerator process. In May, the company issued a statement announcing the first successful testing and shipment of the MURR Mo-99 to their facility in Wisconsin, calling it a “milestone” toward their goal of being the first commercial U.S. producer of the crucial SPECT isotope in over 25 years.

Last August, NorthStar signed a nonexclusive letter of intent with GE Healthcare to provide the company with Mo-99. SHINE, which utilizes linear accelerators and low enriched uranium to create Mo-99 isotopes as a fission product, has entered supply agreements with GE and Lantheus Medical. Both SHINE and NorthStar have to become fully operational and market certified before any isotopes can be distributed – objectives SHINE hopes to satisfy by early 2018.

Greg Piefer, founder and CEO of SHINE, said in a statement that the partnership, “signals the beginning of a new era for the production of radioisotopes in North America, in which a private producer can thrive.”

Unlike when using a cyclotron to directly produce Tc-99m, the longer half-life of Mo- 99 means producers of the parent isotope could theoretically engage more actively in the marketplace.

While Zebrick points out that neither of these strategies have been approved for commercial use by either country, he and some colleagues have met with the FDA, and he says the administration was receptive to the cyclotron solution. “Canadians are close to submitting their cyclotron strategy to Canada Health,” he says, “and the FDA has agreed to receive a copy of that submission and let it serve as reference for them.”

Sodium fluoride as alternative to Tc-99m?
Sodium fluoride for PET imaging has emerged as a valuable diagnostic tool for bone imaging – one of the indications Tc-99m SPECT imaging is frequently used for. Some experts speculate that the comparative availability of sodium fluoride to hospitals could make it a viable imaging alternative to Mo-99.

Cancers most likely to metastasize to the bone, such as breast, prostate, lung, thyroid and kidney, “that’s where we’re seeing increased utilization of sodium fluoride,” says Edgar Alvarez, senior marketing manager of Siemens’ PETNET Solutions, a company offering a large national network of sodium fluoride.

According to Atcher, imaging with sodium fluoride is nothing new. “In the old days of nuclear medicine before we started using these fairly elegant cameras to generate images, we actually did use it to do bone scans, but with a fairly crude imaging system,“ he says.

In addition to doing similar things to SPECT, in some ways PET imaging with sodium fluoride is an improvement. “[It has] two times the uptake in the bone and the blood clearance is faster,” says Alvarez, “and because of the advancements of PET imaging technology, you get a higher resolution image.”

CMS currently covers sodium fluoride scans through participation in the National Oncologic PET Registry (NOPR), a requirement under the agency’s “coverage with evidence development” approach to evaluating products. Alvarez says he is optimistic that the clinical value of sodium fluoride PET will encourage CMS to offer national coverage for these studies, adding that the availability of sodium fluoride throughout the U.S. could provide an added incentive.

“Do I see it replacing SPECT bone imaging completely? I don’t think so. But it is realistic to see its utilization for the evaluation of metastatic disease for certain cancers, such as prostate cancer, being higher than SPECT bone imaging,” says Alvarez.

Zebrick agrees that sodium fluoride for PET has tremendous potential, but stresses the expense of PET, an emerging technology, compared to the relative low cost – and significantly larger install base – of SPECT.

For Atcher, the high level government nature of conversations about Tc-99m deal too heavily in the back-end of the conversation. “It’s always frustrating for the people actually doing these procedures – the physicians and technologists – to realize they are incredibly dependent on decision making that leaves them out of the process.”


EDIT: An earlier version of this story indicated SHINE had entered an agreement with Lantheus Medical that was "similar" to the non-exclusive letter of intent that NorthStar had signed with GE. The revised version clarifies that SHINE has entered supply agreements with Lantheus Medical and GE.