| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||
1 Gray Institute for Radiation Oncology and Biology, Old Road Campus Research Building, Roosevelt Drive, University of Oxford, Oxford OX3 7DQ and 2 21C Institute of Particle Therapy Cancer Research, Particle Physics, Denys Wilkinson Building, Keble Road, University of Oxford, Oxford OX1 3RH, UK
Correspondence: Bleddyn Jones, Gray Institute for Radiation Oncology and Biology, Old Road Campus Research Building, Roosevelt Drive, University of Oxford, Oxford OX3 7DQ, UK. E-mail: Bleddyn.Jones{at}rob.ox.ac.uk
 |
Abstract |
|---|
|
Introduction |
|---|
The introductory talks included one by Mr Fujioshi, First Secretary at the Japanese Embassy in Paris, who noted the excellent 150-year co-operation between France and Japan, especially in the legal, policing, particle physics and aerospace domains. The Japanese Government Science Policy review of 2006–2010 had emphasised the need for Science and Technology to contribute to the direct benefit of the population. In medicine, an improvement in the cure of difficult-to-treat conditions was encouraged, with an acknowledgement that the use of radiotherapy for cancer was increasing. The Government was anxious to deploy the latest developments and to encourage other countries to follow.
Professor Jacques Balosso (Grenoble), Director of the Etoile project at Lyon, outlined the French proposals for a large carbon ion centre for medical and scientific cancer research at Lyon, commencing in 2014. The aim would be not only to treat the accepted indications for particle therapy (amounting to around 750 patients per year in France on the basis of cancer registry and epidemiology studies) but also to develop the evidence base for wider indications using prospective studies for all treated patients with an emphasis on randomised studies. Comparisons between proton and carbon ions will be uppermost on the research agenda. He suggested that France should have its own National Tumour Board for patient selection. Further details can be obtained at http://www.centre-etoile.org.
As well as the expansion in proton treatment capacity at Orsay in Paris, France is also embarking on an ambitious project at Caen, Normandy. The first clinical cyclotron capable of accelerating carbon ions to over 400 MeV will be installed there after its construction by IBA in Brussels. This centre will, after the expected start in 2013, be devoted to clinical research and technological development but with a small routine clinical workload. Consequently, France will be well placed to make major contributions to improving the radiotherapy of cancer over the next two decades.
The main contributors to the meeting were from the Japanese NIRS. They included an overview from Professor H Tsujii, who explained the encouragement provided by the Government in two successive 10-year cancer plans starting in 1994. The two government-sponsored particle therapy centres (protons at Tsukuba and ions at NIRS) have been followed by five further established centres, and three more are in construction. Japan will therefore have 10 centres in operation soon, almost approaching one per 12 million of their population. Four of these have 360° rotating gantries and three will deliver carbon ions.
Recently, the Japanese Government have scrutinised the clinical results and approved reimbursement costs of approximately $30 000 per patient in order to provide a kinder form of cancer treatment. This is exemplary.
The only other attempt at carbon ion therapy outside Japan has been the successful project at the German National Physics Laboratory in Darmstadt, where treatments were only permitted in three blocks of 20 days per year, which allowed development of a "19 fractions in 19 days" schedule, with impressive clinical results. The techniques used went beyond those in Japan, with raster spot scanning that applies the dose in successive layers of each cancer. This highly successful cross disciplinary project has led to the establishment of a large full-time carbon ion facility at the nearby University Hospital Heidelberg, adjacent to the German Cancer Research Centre. This is expected to treat cancer patients later in 2009, with the objective of comparing protons with ions ranging from helium through to carbon in a multimodality and interdisciplinary setting.
|
Clinical results from Japan |
|---|
Since 1996, 4500 patients have been treated at NIRS with impressive results, as well as a smaller experience in Germany [4, 5]. The updated clinical outcomes for the following cancers are presented below.
Skull base tumours
Results are consistent with other centres worldwide but with a slightly higher local control (78%) at 5 years compared with protons, although the selection bias is large in this category of rare tumours. So far, no severe toxicity of the brainstem, optic nerves or spine has been observed. The experience in this category is small (just over 100 patients) because of the rarity of these tumours.
Head and neck cancers
A 5-year local control rate of 70% is found following carbon ion therapy in a wide range of 328 head and neck cancers, of which 110 were recurrences after surgery. The treated adenocarcinomas, which predominate in this series, had an 80% local control rate, whereas squamous cell cancers had 60% control but was stable at 30% by 10 years. Skin reactions can be severe (as with X-rays) and in many cases of salivary gland tumours, where only the same side of the neck is treated, the acute morbidity is much reduced. There is no risk of spinal damage in most treatments and hearing can be better preserved. Patients with mucosal malignant melanoma show a better local control (60% compared with <40% if the tumour volume is <60 cm3). Such results cannot be achieved with X-rays. The role of concomitant chemotherapy needs to be explored carefully in non-melanoma patients.
Sarcomas
The results of 583 soft-tissue and bone sarcoma patients showed local control of 78% with a 57% 5-year survival rate. These include high-grade inoperable osteosarcomas treated also with neoadjuvant combination chemotherapy. Skin reactions were the main reported toxicity, with a surprisingly low incidence of sciatic neuropathy, a risk which can occur if the length of nerve treated exceeds 10 cm. The experience with sacral chordoma is excellent, with no instances of colostomy or renal diversion procedures.
Primary hepatocellular carcinomas and palliative therapy for liver metastases
Results continued to be equivalent to the best obtained following surgery, but with lower morbidity. Progress had been made with reducing fractionation to two treatments. There was evidence that tumours larger than 5 cm in diameter had a worse survival rate (14% vs 53%) at 5 years. Studies of apparently isolated colorectal liver metastases commenced using 36 GyEq, 40 GyEq and 44 GyEq in a single fraction. Carcino-embryonic antigen and CA 19-9 serum marker levels fell to the normal range, with an improvement in positron emission tomography (PET) uptake after a few weeks, but it is impressive that there were no adverse acute reactions.
Locally recurrent rectal cancer
In this situation, X-ray therapy is usually palliative in intent, but for carbon ions it is possible to deliver much higher doses, e.g.73 GyEq in 16 fractions, and so aim for local cure. The recent analysis showed that 94% local control is achieved with a 5-year survival of 
40%.
Pancreas
Experience over the past 4 years has found a 44% 1-year survival rate with carbon ions and gemcytabine chemotherapy for 47 patients with advanced local cancer, and with very little acute toxicity compared with X-rays. 22 patients with operable tumours have been treated pre-operatively with 30–35.2 GyEq in eight fractions, as there is a 50% recurrence rate after surgery alone. So far, in this second group, the 1-year local control is 86% with a 100% survival rate. The overall results indicate a 30% 3-year survival rate, with 51% after additional surgical resection. The Grade I toxicity rates are low, and so far no Grade 2–4 toxicity has occurred.
Uterine cancers
The cancers treated were largely bulky but localised adenocarcinomas, with the entire treatment delivered by carbon ions, thus eliminating the use of brachytherapy. The opposed fields can be extended to cover the lower para-aortic nodes while delivering acceptably low small bowel, bladder and posterior rectal doses, an arrangement that is not possible with X-rays. The 5-year local control rate is 64%, with overall survival at 46%. In the Phase II study of 66 patients, only one Grade 4 toxicity event occurred — a colostomy for fistula in a diabetic patient. Salvage surgery has also been performed for patients with a residual mass; the local control then appeared to increase form 46% to 69%.
Prostate cancers
The use of 16 fractions rather than 20 has resulted in reduced toxicity [4]. The latest analysis of the 57.6 GyEq dose in 16 fractions over 4 weeks in a population comprising 60% "high-risk" patients revealed a 1.9% and 4.8% Grade 2 toxicity rate in the rectum and bladder, respectively. These results are better than three-dimensional conformal and intensity-modulated radiotherapy studies in which rates of 4–5% and 7–11% are reported for each category, respectively. Local control is 97.5% and biochemical control 89%, with a survival rate of 95%; the high-risk category survival appears higher than in X-ray-based Radiation Therapy Oncology Group (RTOG) studies in the USA. A new protocol of 51.6 Gy in 12 fractions has been introduced over the past 2 years and, so far, there have been no biochemical failures or serious toxicities.
There is some interest in a pan-Japanese particle therapy fractionation trial, as there are now sufficient numbers of centres to make this possible.
|
The scope for randomised trials |
|---|
|
Radiation biology |
|---|
At present, the radiobiology is mostly studied in a one- or two-dimensional test tube-like situation and needs to be extended to three-dimensional studies, as the magnitude of the biological effects vary according to the dose gradients obtained in the clinic. Even in situations where dose gradients have been studied, the limitations of in vitro experiments on the survival range require doses that are far lower than would be required in clinical practice for tumour cure in, for example, one to four fractions. In the final analysis, the results of clinical treatments will be the only ones that can provide the correct parameters for use in humans. Even so, cellular and tissue experiments will be necessary to guide progress in some instances and for the key aspect of particle beam inter-comparisons using biological end points. The past data from neutron irradiations should not be forgotten, especially as the relative biological effectiveness (RBE) values are very close to those for carbon ions.
As an alternative to the former neutron RBE-based model applied to carbon ions [9], the Japanese are developing a new model that covers the entire range of linear energy transfer (LET) by using protons, helium, carbon and iron ions. This is essentially empirical modelling, but contains the important turnover of RBE with increasing LET, as determined in three cellular systems. The parameters are turnover point position, width of curve and vertical height. It leads, at very high LET values, to RBE values of <1; these are well beyond the therapeutic LET values but are of interest to radioprotection/astronomy.
In terms of molecular responses, some results from candidate gene expression and early and delayed apoptosis studies were presented, as well as repair rate changes after carbon ion cellular exposures. These studies may suggest ways of better patient selection and the use of dose modification strategies. It is thought that the use of carbon ions, for example, converts an X-ray-treated response into that which would be expected if a DNA inhibitor or a radiosensitising agent had been used. It will be important to simulate the use of all adjuvant therapies used in cancer alongside ion therapy. The appointment of Penny Jego from the University of Sussex to the NIRS faculty will ensure further investigation of DNA repair mechanisms.
|
Costs of ion beam radiotherapy |
|---|
19 500 for carbon ions compares favourably with novel biological therapies, e.g. temazolamide and avastin at around 25 000 (the same as for a hip replacement) and as high as 36 000 for herceptin, with a ceiling of 60 000 for cetuximab in recurrent rectal cancer. The study by Jäkel et al [9] showed how definitive therapy of chordoma would cost ~27 000 if surgery was followed by ion beam therapy. The average cost of 10 chordoma patients who had multiple surgical procedures amounted to over 100 000. In terms of cost-effectiveness, the breakdown points for 70% and 65% local control occurred at 20 and 16 fractions, respectively. Much further work needs to be done in this respect, as recommended by other authors [10], especially for the more commonly occurring cancers. |
Developments in physics |
|---|
The combined biophysical treatment planning method used in the NIRS is based on a transformation of neutron effects at the same linear energy transfer of 80 keV µM–1 as obtained with a carbon beam in the spread out Bragg peak (SOBP), where the RBE is 3 towards the end of the SOBP. Treatment planning takes 1 week to complete because of individual dosimetry and construction of customised beam shaping and modulating devices. Future plans for a three-dimensional scanning beam will avoid many of these delays and will reduce the waiting time, as customised devices will no longer be required.
The quality assurance systems used in the NIRS were described; owing to a combination of ageing equipment and increased workload per year, breakdowns are becoming more frequent. These appear to be of short duration, with only two examples of delays over 6 h in one year (2007). Most breakdowns are caused by hardware failure, such as motor drives controlling treatment couches. Faster throughput of patients is anticipated when rotational gantries will be used in 2010.
In France, work is proceeding on cryogenic gantries with one 90° magnet, which should reduce gantry weight by a factor of three and overall dimensions by 30%. The additional cost is estimated to be 2 million. A prototype is being considered. Quenching of the temperature control is a potential problem, as it would take 2 days to restore a sufficiently low temperature of 4.5 K.
Further work includes real time imaging improvements to check dose distributions using advanced PET scanning, improved scintillator detectors and proton radiography. The difficulties associated with treating mobile tumours were illustrated using high-resolution CT scanning along with synchrotron beam pulsing to match the movement. The NIRS has been using respiratory-gated radiotherapy since 1992.
Work on more compact synchrotrons for hospital use is underway in Japan, as well as in the UK.
|
Other European centres |
|---|
for summary). This list does not include the many different projects for proton beam therapy in Europe. It became apparent that Price Waterhouse Coopers, the London-based Management Consultancy agency, had advised the University of Kiel project.
|
|
The co-ordination of research and infrastructure by CERN |
|---|
1 million to create a community of workers in this field. Although this funding ended in 2005, EU support has continued in the award of:
5.6 million Marie Curie project called PARTNER (Particle Training Network for European Radiotherapy), which will fund 25 young researchers from 10 countries in this field (see http://www.cern.ch/partner); 8.4 million infrastructure called ULICE (Union of Light Ion Centres in Europe) aimed at integrating clinical, biology, physics and technology (see http://www.cern.ch/enlight). These projects effectively allow close collaboration with respected CERN scientists and access to grid computing and other advanced technologies and scientific expertise available there.
|
Discussion |
|---|
The progress made in Japan with the reduction in treatment sessions (fractionation of therapy) carries huge implications for the overall costs of radiotherapy: the use of a single or 4–15 treatments would curb costs sufficiently to allow much wider applications throughout the world. For this to be accepted, the clinical studies must be beyond reproach and avoid as much selection bias as possible in order to ensure a high degree of confidence.
The continued debate about the ethics of randomised studies in this area needs further input from lay people, and some degree of public debate should be encouraged. This has already been the case for issues such as embryonic research and transplantation, but similar comprehensive discussion and debate would be useful to guide the conduct of cancer radiotherapy. It would certainly be useful if there were to be more assessments of quality of life and symptom duration, as well as of cure statistics. It should be possible to use randomisation within Phase II studies where different policies were used, e.g. fractionation, treatment volumes and margins, use of different respiratory gating techniques, different radiobiological models and mixtures of protons/carbon and X-rays, along with other adjuvant therapies.
Japan is fortunate to have such exemplary screening programmes, e.g. for lung and stomach cancer. The higher yield of smaller tumours allows radical therapy by ion beams to be used to its best advantage. To further improve the results of pancreatic cancer, for example, a screening programme linked to early referral for definitive therapy is indicated.
There needs to be a common approach to the radiobiological issue of conversion from physical to biological dose. This is a complex subject containing many assumptions, and different countries use their own systems. A simpler and more reliable overall framework for more widespread applications is required.
Global governance of particle therapy by an international organisation would be a worthwhile future goal to ensure equity of access, quality and effectiveness of therapy. In addition, the establishment of worldwide data collection and analysis, together with trial promotion and organisation, is highly desirable. The CERN model is exemplary in this respect and should perhaps be extended in conjunction with the international authorities that control health and radiation activities, such as the IAEA (International Atomic Energy Agency) and World Health Organization, but only with the agreement of governments across the world. As particle physicists have made good progress by such an organisation, the medical application of physics in cancer needs a similar overall work ethic and coordinating body.
There are now very good prospects for controlling cancer more effectively and with far reduced toxicity than has previously been accepted. In future, the linkage of population screening by clinical radiological and molecular means, if coupled with the best available surgery and radiation techniques that physics can provide, should provide significant benefits to humanity.
|
Summary |
|---|
|
Acknowledgments |
|---|
Received for publication March 28, 2009. Revision received June 17, 2009. Accepted for publication June 23, 2009.
|
References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
Top
Abstract
Introduction