Cyclotron vs. Reactor

In the six months that the two facilities were offline concurrently, the global imaging community lost 65 percent of its regular production of medical isotopes.

“What happens to supply lines when these materials are traveling farther and farther from the U.S. market?” - Dr. Robert Atcher

I was introduced to Al Swank about six years ago by a radio story on NPR’s “All Things Considered.” Although the term had not come into vogue at the time, it had all the makings of a viral media event: a frontier setting, futuristic and dangerous-sounding technology and community uproar.

Swank’s plan to relocate a retired cyclotron from Johns Hopkins University in Baltimore, Md., to a family property in Anchorage, Alaska, had gained national traction after a town commissioner objected to his proposal and petitioned the state for emergency revocation of his permits. Depending on which story you followed, Swank was either a real-life incarnation of eccentric tinkerer Doc Brown from “Back to the Future” or a dangerous crank scheming to plunge his hometown into nuclear winter with an illegal doomsday device.

When I reached out to Swank, he understandably had serious media fatigue. He was leery of inquiries from reporters whom he knew would inevitably get the science of his proposal wrong, or who would frame their articles in terms of how long it would take an “amateur” cyclotron technologist to turn his basement into Three Mile Island.

Swank, now 60, is a multi-disciplined, registered professional engineer who specializes in the design of particle accelerators, cyclotrons and isotope production. He works with nuclear physics researchers and government laboratories through Langdon Engineering & Scientific Services. Swank constructed his first cyclotron by age 17.

But at the time of my call, nobody writing about the situation seemed to share Swank’s view of the story: That a small-time entrepreneur was fighting a political battle against constantly moving regulatory targets in his effort to bring cancer-fighting drugs to Alaska.

Swank, who lost his father to cancer years ago, wanted to use the cyclotron to produce a range of radiopharmaceutical isotopes that are too short-lived to be flown in from Seattle, the nearest such American source. To this day, there is still no production of these isotopes in Alaska. Yet no one in the nuclear medicine community thought of Swank in those terms, if at all.

“We started off with one PET camera in Anchorage in 2006, and now we are up to our fourth,” Swank says, “and everybody is screaming because they need PET isotopes,” he says.

A Personal Quest Meets an International Crisis

In May 2009, perception of Swank’s agenda shifted from quixotic to prescient when the reactor went offline at the Ontario, Canada-based Chalk River Laboratories, the chief supplier of radioisotopes to the American medical community. A worsening heavy water leak at the facility shuttered the entire operation for more than a year.

The impact of that discovery was compounded by a routine maintenance shutdown of the High Flux Reactor in the Netherlands in February 2010. In the six months that the two facilities were offline concurrently, the global imaging community lost 65 percent of its regular production of medical isotopes. Only three major reactors were left to make up the shortfall: BR2 in Belgium, OSIRIS in Franceand the SAFARI-1 reactor in South Africa.

The American imaging community was ablaze. With exactly zero of the five chief global producers of radioisotopes located on domestic soil, the term “strategic reserve” began to apply not just to political conversations about petroleum. Agencies like the Society of Nuclear Medicine (SNM) have pushed discussion on this issue to the forefront of national policy.

“What happens to supply lines when these materials are traveling farther and farther from the U.S. market?” asks Dr. Robert Atcher, SNM Past President and the Chair of its Domestic Isotope Availability Task Force.

“When the volcano in Iceland erupted, it disrupted isotope supply for a week,” Atcher says. “They actually trucked materials down from the Netherlands to Rome in order to get them to North America, and that day cost them. 9-11 closed the borders and we had to get a special dispensation for medical isotopes.

Even with ancillary agreements penned to source isotopes from reactors as far off as Australia, the physical limitations of distance may be indomitable for the preservation of materials the half-life of which are measured in hours, says Atcher.

“If we have any kind of incident, particularly one involving radioactivity, there’s not going to be anything moving from offshore to the United States,” he says.

The Workhorse Isotope

The biggest concern of the isotope shortage is the availability of Technetium-99m (Tc-99m). Tc-99m is used in about 90% percent of all single-photon imaging procedures, making it the backbone of the diagnostic imaging world. The isotope has a half-life of six hours and is minimally radioactive, making it ideal for imaging because it doesn’t remain in the body for a long time and is easily traceable.

Procedures for which Tc-99m is optimized include cardiac screenings, bone cancer detection and pediatric imaging. This versatility combined with its relatively low cost (as compared with other isotopes) makes Tc-99m “the workhorse isotope,” according to Atcher.

However, the advantages of Tc-99m also describe its innate flaws. The isotope is the “radioactive daughter product” of molybdenum-99, which produces Tc-99m as it decays. This fact makes stockpiling Tc-99m impossible, and also means that 25 percent of the world’s supply of molybdenum decays daily.

Molybdenum-99 is produced in a nuclear reactor by splitting highly enriched uranium.

The world’s reactors run on an annual production schedule that requires significant logistical calculation, says Atcher, given the half-life of molybdenum-99 and the operational costs associated with fissile materials.

“You can’t make the molybdenum unless the reactor’s running, and the real substantial production cost is associated with running the reactor,” Atcher says. “At the beginning of a fiscal year, you decide how many operating cycles you can run with your budget and that’s what you do.

“The bad news,” Atcher says, “is that all the reactors that are used to do this are multitask facilities. Chalk River does fuel testing and structural studies; trying to divide up what percentage of the reactor operations should go to isotope production is some kind of guesswork.”

“Isotopes don’t make money for the reactor guys because they’re a pain to deal with,” says Swank. He also dismisses the notion that domestic nuclear power-generating facilities could be retrofitted to produce isotopes in the event of a shortfall.

“A commercial power reactor is not designed to bombard the materials to make isotopes,” Swank says.

So aside from the fact that the source materials required to make Tc-99m are constantly eroding due merely to their atomic makeup, and that the isotope itself decays too fast to store any reserves, production must be perpetual to have a chance of meeting demand. But who’s going to foot the bill to run a nuclear reactor for a year?

Picking up the Tab

In Europe, national governments foot some of the reactor operating costs in the interests of public health. When the Chalk River reactor went down Atcher says it was rumored that the Canadian government made up the difference with $20 million from its own coffers. However, he also argues, although subsidies keep the atoms splitting, they effectively stifle any free-market investments that would eventually wean the isotope producers off public funds.

“There’s sort of this line that’s been drawn in the sand that we can no longer operate with subsidies because it’s killing the incentives for private organizations to maintain the supply,” says Atcher. “If you reach the point where private industry is willing to invest to make molybdenum, then everything’s good.”

However, Atcher says, the switch cannot come all at once, for obvious reasons. If subsidies disappeared, even if they were phased out gradually, there is no way of telling what would happen to production. Moreover, as we’ve already seen, even a minor hiccup can have a devastating effect throughout the global medical environment.

When the 2009/2010 shortfall hit, Atcher says that private parties within the isotope manufacturing industry itself provided the funds to add additional operating cycles at BR2 and SAFARI-1.

“The additional funding they needed for operations and fuel was provided by the industry,” he says. “My presumption was that the operating entity did not do that out of the kindness of its heart, but that’s something that’s never been discussed publicly.”

The shortage’s reach stretched beyond the U.S., Swank says. “Many patient imaging procedures were not available or were delayed due to the unavailability of Tc99-m in Alaska. The local medical community started to consider the use of PET imaging rather than SPECT imaging during the shortage.” The shift toward PET became common all over the U.S., Swank says. In the last five years, Alaska has acquired three PET imaging cameras. Before, there was only one in the state.

Atcher serves on the High Level Group for Medical Radioisotopes, an international task force within the Nuclear Energy Agency and the Organisation for Economic Co-operation and Development (OECD-NEA), which is headquartered in Paris, France. The Group has proposed a plan to increase independently the worldwide price of molybdenum in order to generate enough funds to operate reactors in the event of a shortfall or outage – sort of like an isotopic OPEC.

“But you’ve got five countries doing five different accounting principles,” Atcher says. “The other issue there is that you’re almost proposing something that’s impossible to do. The United States represents 50 percent of the market for molybdenum, and Medicare sets the tone for reimbursement in the medical world.”

Third-party insurers tend to follow how the Centers for Medicare and Medicaid Services (CMS) designates reimbursement costs for any medical procedure, Atcher says, and CMS typically treats radiopharmaceuticals as a supply item bundled with the cost of the technical procedure.

“So these people can keep raising the price of molybdenum, but the only people they’ll hurt is the clinic doing the procedure,” he says.

“In most of the rest of the world, if the single-payer systems agree to increase their price for molybdenum, that’s fine,” he says. “It won’t work in the United States. But because all the people involved in production are all outside of the U.S., we might be beholden to the way they dictate policy.”

 A Domestic Distributed Model

In the nearly 10 years it’s taken Al Swank to broker his cyclotron deal with Johns Hopkins, the Anchorage town council, the NEA and the state of Alaska, he’s never once considered that the time it took him to meet the various regulatory hurdles might also have primed the public pump for his venture.

As Swank prepares to take his initiative online in September 2011, he envisions a future in which a distributed network of much smaller, non-fission-based isotope producers could complement or replace the larger reactors and the hassles that go along with them. It’s a future in which he naturally sees his own operation as a model enterprise.

“As far as I’m concerned, what’s going to happen is that reactors will no longer be the main suppliers of a few critical isotopes,” Swank says.

“There will be some national half-dozen sites that won’t be reactor-based, as well as the development of some high-energy accelerators and cyclotrons that they can transition over to that support base.

Swank hypothesizes that suppliers will include companies like GE and Siemens, private entities and hospitals.

“Then everybody will have access and it won’t be contingent on nuclear reactors anymore,” he says.

In Swank’s favor are environmental impact and safety arguments – an accelerator or cyclotron is far less risk to manage and operate than a nuclear reactor – as well as cost calculations. Swank ballparks the total cost of ownership for a 24 to 70 million-electronvolt (MeV) cyclotron site at anywhere from $25 to 80 million. That there are only five major isotope-producing reactors in the world should give you an answer as to the costs of building one of those.

“With a dozen specialized accelerators and cyclotrons, you could outfit the United States with all the technetium and other nuclear medicine isotopes it needs: iodine, thallium, palladium and others” he says. “At 24 MeV, you can make your own molly-99 onsite. The TRIUMF research lab and cyclotron facility in Vancouver is already doing this, with governmental support.”

“You’d distribute the technetium in a very similar way that you do now,” Swank says, “with multiple smaller production facilities. You wouldn’t have the big reactor making thousands of curies of molybdenum, with a single point of failure.”

Atcher believes that although Swank’s vision is certainly plausible, it’s by no means easily realized.

“The installed cyclotron base in the United States is all mostly used for making fluorine-18,” he says. “They’re not physically capable of making technetium directly.

             “On one hand, you can make the case that it’s a better model than relying on the Canadian and Dutch reactors,” Atcher says, “but when you look at the procedural issues associated with using a cyclotron to make technetium, it’s a much more capital-, time-, and personnel-intensive procedure than making fluorine-18.”

Atcher thinks that a model such as that which Swank proposes “will break down in the economics.”

“Nobody’s demonstrated the ability to manufacture isotopes with a cyclotron day after day with all the additional steps and complications in the side products that build up with that target material,” he says. “Plus the FDA would make people re-qualify technetium produced by a different process than the one that we have used for the past twenty years,” Atcher says.

Atcher’s money is on more material coming to the United States from Australia, which he says is capable of supplying 10 percent of the world market now, or 40 percent with added processing facilities.

“The problem is that they don’t have the added processing capacity,” he says.

Whatever the solution, it is true that other reactor systems worldwide have also almost outlived their intended life spans. Atcher says the Dutch are in the process of siting and building a reactor to replace HFR; another project is underway to replace the OSIRIS reactor in France. Both are 50 years old.

Amidst that turnover, he says if the U.S. can establish some sort of domestic production capability covering even 20 percent of its own domestic market, that’s a tremendous stride for patients.

“We have to start including the economic effect of doing these studies in terms of value to the patient.”

Originally published in Medical Dealer Magazine