Share |

Computing to advance new cancer treatment

Imagine yourself in a tough fight against an entrenched opponent that is small and elusive, and holding an innocent hostage.

What weapon would you prefer to use: a cannonball or a bullet?

That is the basic concept behind hadrontherapy, a technology that provides more effective, tightly targeted cancer treatment with fewer side effects than conventional radiotherapy. Soon researchers and clinicians will be able to share hadrontherapy-related data via an online 'data hub'  supported by grid middleware.


Model of precise irradiation of tumor with hadrons using raster scanning (horizontal radar sector scan that changes in elevation) technique. Image courtesy GSI, HIT, Siemens.

The reasoning is simple. In the world of cancer therapy, X-rays can sometimes be ‘cannonballs,’ releasing their energy relatively indiscriminately over a broader area than doctors would like when aiming at a tumor. This is especially true when dealing with tumors near the brain and the eyeball, in which there is little room for error.

In contrast to this ‘shotgun’ approach, hadron particles such as protons and neutrons can be fired, sniper-like, to directly hit and destroy a cancer target alone while leaving surrounding healthy tissue untouched.

In addition, hadrons can be calibrated to hit a given cancer at its ‘highest-damage level’ in one dose, while X-rays (high-energy photons) require repeated dosages over time.

The efficacy of hadrons is due to an effect known as the ‘Bragg peak:’ hadrons deposit most of their energy at the end of their range with no loss of intensity; immediately afterwards, their energy level drops to zero. Thus, you can come right up against a critical organ and kill the tumor without causing any collateral tissue damage. (The concept of using the Bragg peak in this manner on cancer was first proposed by Harvard physicist Robert Wilson in 1946.)

The idea has been around for decades, with great strides happening relatively recently. Recently, carbon ions have been shown to have more effectiveness at treating cancers that are resistant to photons and even protons. These ions hit tumours with 24 times more energy than a proton and are especially suited to ‘radio-resistant’ tumors.

What may surprise you is that hadrontherapy comes straight out of research into high-energy particle physics.


European proton-carbon ion facility in Heidelberg, Germany. Image courtesy GSI, HIT, Siemens.

Size matters

The first patient to be treated with protons was at Berkeley, California, in 1954 — a state that is also home to the first proton clinical facility, the Loma Linda University Medical Center, built in 1990.

Much research and many hundreds of millions of dollars of US federal funding went into developing the technology to accelerate protons, much of which was done at the Fermi National Accelerator Laboratory. Fermilab physicists and engineers built the proton accelerator that exists at the Loma Linda facility today.

As of 2010, over 67,000 people worldwide have undergone proton therapy, according to the Particle Therapy Cooperative Group of Switzerland.

(Cancer treatment via fast-neutrons has developed in parallel with proton therapy, with most treatment centers based in the US; one example is the Neutron Therapy Facility based at Fermilab, which has treated over 3,000 patients since 1976.)

But the accelerators needed to generate the particle beams are very expensive. Furthermore, in order to attack cancer from different angles around a patient, the particle beam must be steered using large and powerful magnets, using a heavy and complex device known as a gantry.

No one thought that particle accelerators would be used for medical applications, or need to be mass-produced or even portable. Some particle accelerators, such as the cyclotron  at the PSI institute, have gantries that are over 15 meters in diameter and weigh 100 tons. This has not helped the field to grow.

Encouragingly, a new prototype using a linear accelerator has just been developed. The LIGHT (Linac for Image-Guided Hadron Therapy) accelerator offers the benefits of being both cheaper and smaller. (The expense was often an issue; in the US, proton therapy has often been called “the rich man’s therapy.”) With a more mobile and affordable tool, a key element seems to be in place to make hadrontherapy more widely available.

But the therapy is about more than just hardware. It is also about software, a database, and a trained core of researchers and clinicians with expertise drawn from a variety of fields.

Enter the PARTNER (Particle Training Network for European Radiotherapy) project, coordinated by CERN under a four-year Marie Curie grant funded by the EC Framework Program 7.

The PARTNER project funds PhD researchers to be the next generation of hadrontherapy experts. They are currently developing a software prototype, supported by grid middleware, for all research data. This consolidated hadrontherapy database will eventually be accessible to hospitals and doctors, enabling them to securely import, transfer and view patient information. They hope that this ‘data hub’ will allow for easy extrapolation of results to test new treatment ideas and permit patient referrals anywhere in the world.

One CERN fellow, Vassiliki Kanellopoulos, of the University of Surrey, UK, envisions that the end result will be a ‘superhighway’ of information exchange across disciplines. “Look at it like the Autobahn in Germany; everyone uses it for different purposes,” said Kanellopoulos.


Conceptual outlook for a prototype Hadrontherapy Information Sharing Platform (HISP) which supports patient referral, research and acts as a rare tumor database. Image courtesy PARTNER project.

Challenges

However, while the system  connects physicists and doctors, it also needs to be accessible to the larger world; biologists and epidemiologists must be able to use it too.

Consequently, it has to speak their languages.

The problem is that each field often seems to describe the same data in different terms. The system will have to be able to interpret and re-use this data in different contexts; for example, when a physician needs to understand research data from a hospital in another country in order to create the best hadrontherapy treatment regime for a given patient. “Capturing and describing data using semantic web technologies is important,” said Faustin Laurentiu Roman, a CERN researcher from Spain’s Instituto de Fisica Corpuscular. Just like the semantic standards being developed to help the next generation of the World Wide Web comprehend the meaning of data, standards need to be established for hadrontherapy.

This medical information superhighway also has to deal with the sharing of patient data, something which patients, doctors, hospitals and insurance companies are often reluctant to do. Daniel Abler, a CERN fellow from Oxford University, working on this problem said: “The role of grids is to overcome some of these issues, as it provides tools for managing distributed resources within a strong security framework.”

Consequently, it would seem that an easily accessible scientific application for the medical and physics community is a long way off.

To help overcome these obstacles, the first Physics for Health conference was held in February 2010, at CERN, in which all parties could discuss the issues. The next one will be held in 2012.

Adoption by doctors is one more hurdle to the wide availability of hadrontherapy. Jane Barrett, president of the Royal College of Radiologists in the UK explained, “Radiotherapy plays a very small part in a doctor's training . . . compared with drugs and surgery.” Barrett's statement is sobering; if general practitioners receive little training on radiotherapy, it is no wonder that they remain poorly informed about hadrontherapy.

The PARTNER researchers will be giving an oral presentation of their prototype system at the upcoming EGI User Forum, which will be held in Vilnius, Lithuania, during the week of April 11-15.

Your rating: None Average: 4 (2 votes)

Comments

Post new comment

By submitting this form, you accept the Mollom privacy policy.