However, the exquisite specificity of proton therapy has created a new set of challenges for radiotherapy planners. Tightly focused energy delivery might minimize collateral damage, but if the tumour is much larger than the treatment beam, then malignant cells could escape unharmed.

Most proton radiotherapy facilities use passive scattering techniques to expand the treatment beam's coverage. With this method, a number of material components are placed in the firing line to scatter the incoming protons. The broadened beam is then shaped to the desired size by collimators.

The downside of this strategy is that proton interactions with materials in the beamline will create high-energy secondary neutrons. The high linear energy transfer (LET) of neutrons makes them extremely efficient at ionization, and far more likely to cause cell death than low-LET particles, such as X-rays or protons. If these neutrons reach the patient, then the risk of a second cancer increases significantly.

Given the growing number of proton therapy facilities worldwide, and their almost-exclusive use of passive scattering technology, this is a risk that should be taken seriously, according to David Brenner and Eric Hall from Columbia University Medical Center (New York, NY).

"The issue of how damaging these low doses of neutrons can be has perhaps not been fully appreciated," Brenner told medicalphysicsweb. "Typical lead times between radiation exposure and a tumour appearing can be 20, 30 or even 40 years. Our concern is that we are setting up a problem for the future."

Number crunching

The two radiation-physics professors have now calculated the lifetime risk of neutron-induced secondary cancer to patients undergoing proton therapy (Radiother. Oncol. 86 165). As a starting point, they took previously calculated data on the neutron doses received by non-target organs during proton treatment of a lung tumour.

The next step was to calculate the relative biological effectiveness (RBE) weighting factor that would make standard low-LET cancer risk estimates applicable to the high-LET neutrons. They then derived the risks for male and female patients, of differing ages, of developing another cancer induced by secondary neutrons.

The lifetime second-cancer risk for a 15-year-old male and female patient came out as approximately 5% and 11%, respectively. But these risks could actually be four times higher or lower, owing to uncertainties in the RBE calculation process, Brenner cautioned.

"There is no good source of information on the cancer risk from neutrons," he explained. "The number we arrived at is very much influenced by animal data, but there are uncertainties about extrapolating from rodents to human beings."

Brenner is keen to stress that the work should not be used as evidence against proton therapy per se. It is more a call for facilities to investigate alternative beam-broadening methods.

One option would be a switch to active scanning technology, where magnets are used to sweep the proton beam across the tumour. No physical objects are placed in the beamline, so no secondary neutrons are produced at all. This technology is still in its infancy, though in the future, it may become the norm.

For now, changes to passive scattering systems that would minimize neutron production are more likely. Most secondary neutrons are generated by the final collimator, Brenner said. These collimators are typically made from brass or similar materials with a high atomic number. Switching to an alternative, lower-mass, high-density materials could cut the neutron dose significantly.

"Proton therapy has a great future ahead of it, but if you can reduce this unwanted dose as much as possible, then you might as well," Brenner added. "In the short term, passive scattering will probably continue to dominate the market, but there are still ways of reducing the neutron dose with these systems."