| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Commentary |
Birmingham Cancer Centre, University Hospital Birmingham, Birmingham B15 2TH, UK
Some 13 years ago, the European Organisation for Research and Treatment of Cancer (EORTC) organized a large meeting devoted entirely to neutron therapy, its original promise and, of course, its clinical disappointments. The term "charged particles" has changed the emphasis, with realistic expectations of better clinical outcomes; the best available radiation dose distributions can now be achieved due to the Bragg peak effect that results from the deceleration of charged particles (CP) in tissues; neutrons and X-rays carry no charge and do not exhibit such selectivity in dose deposition. Consequently, the application of curative radiotherapy in the future is expected to rely more and more on these physical advantages [1]. The reader who may be unfamiliar with terms and some of the acronyms used should refer to Tables 1 and 2
. This article is written to inform those working in the various disciplines associated with radiotherapy in the UK about the present and intended developments in Europe, Japan and North America and how these are organized and funded. It is not intended to be an account that proves that these new forms of radiation therapy, which are being tested in phase I/II studies in an increasing number of cancer sites, are better than the techniques presently available in the UK; considerable prospective research and development will be necessary to show not only that this is unequivocally so, but also to investigate the actual cost-benefits compared with other X-ray based techniques such as intensity-modulated radiotherapy (IMRT)/tomotherapy/conformal radiotherapy. Other articles have been written elsewhere regarding this increasingly important topic [2–4].
|
|
Japan, followed by Germany, has pioneered the use of light ions in modern medicine, after some promising results were initially obtained at Berkeley, USA and PSI in Switzerland (initially with pions – or pi-mesons – using more traditional radiation field arrangements and treatment planning methods). The ions, in contrast to protons, have substantially increased linear energy transfer (LET) properties that cause additional biological effects, which can counteract tumour intrinsic and hypoxic radio-resistance; the latter properties were those from which clinical advantages were predicted for neutrons, but neutral particles cannot be deposited with selective advantage as in the case of CP and, largely as a consequence of this, they failed to improve the therapeutic index [5, 6]. The initial claims of the efficacy of neutron therapy at the phase I/II study level – and which were not substantiated in subsequent phase III randomized studies – cannot be compared fairly with the results presently being claimed for CP therapy because tissue side effects were classified in a rudimentary manner during the initial neutron studies, whereas CP studies are presently assessed with the most modern criteria.
The present international position
There is considerable interest in the formation of a network of active CP Centres situated in as many European countries as possible. Germany and Italy have already commenced construction work; the French Government decided in June 2005 to proceed with a CP Centre in Lyon, while also encouraging a second centre in Caen. The existing GANIL (Grand Accélérateur National d'Ions Lourds) facility at Caen can be adapted at relatively low cost for therapy and there is financial support from the Conseil Régional de Basse-Normandie. Austria is well advanced in its facility planning and Sweden, as part of the Nordic Light Ion Project, will soon finalize their options. These centres are separate from existing projects to develop high-energy proton therapy, which requires a lower energy facility than for light ions, e.g. at Munich, Essen, Cologne and an expansion of the service at Orsay, Paris. ENLIGHT is an ambitious 3-year network project that reports back to the EC this year (see estroweb.org); membership has included the scientifically advanced countries in Western Europe with the rather enigmatic absence of the UK. CERN, the European Organization for Nuclear Research based in Switzerland and to which all advanced countries including the UK contribute substantial sums of research monies, is also highly supportive of the now attainable goal of providing CP treatment to a large number of patients so that the promised benefits of Nuclear Research can be spread as widely as possible.
The USA, by comparison, is extending only proton therapy, but at a rapid rate with eight centres either in operation or under construction; more contracts are being considered. The throughput of proton treatments is increasing by 40% per year at the relatively new Massachusetts General Hospital facility. It seems likely that full treatment capacity can be obtained within 3–4 years of starting a new centre. Whether the USA will invest in Centres that can deliver ions remains to be seen; it appeared that there was some interest from the Boston group who attended the meeting.
There are presently no plans to extend proton or ion therapy in the UK beyond continuing the relatively low energy proton work at CCO, Clatterbridge, where targets are restricted to the eye.
Funding and organization of future CP therapy
The major problem appears to be reimbursement, rather than the initial purchase costs of around \#8364;100–140 million. In Japan, one third of the cost is paid by the patient, the remainder by health insurance companies, while the facilities receive separate subsidies. The Austrian state has agreed to fund treatments in a privately owned but university managed facility; some 40% of the beam time – at night and weekends – would be used to generate income from other research projects in physics, astronomy and medicine in order to cover the initial operating costs of approximately \#8364;18 million per year (www.medaustron.at). The funding pattern differs in Italy, where 90% of the funding will be from its state health service, with plans to treat 3000 patients per year at Pavia (see www.tera.it), where building has now commenced. This large facility is to have advanced electronic links to the existing nine large Italian Cancer Centres.
The French Etoile and ASCLEPIOS projects are a consequence of the French Plan against Cancer (2002). Their national programme for research in hadrontherapy (therapy using particle irradiation) has now officially commenced: funding will be via their Ministry of Health and Ministry for Research. The Royal College of Radiologists have recently nominated a team of three to initially represent the British health service in collaboration with ASCLEPIOS: these are Professors R G Dale, Dr I Rosenberg and the present author. Formal agreements are presently being processed. It is envisaged that a British team will need to be employed for selection and care of patients treated in France. The UK government will be obliged to fund treatments approved in other EC countries that are not available on our islands.
Construction of a new German facility at The University Hospital of Heidelberg is proceeding well and randomized clinical trials (RCT) are being planned: this is in marked contrast to neutron therapy where no RCT were performed in a plethora of German departments, quite unlike the exemplary efforts made in the UK under the auspices of the MRC.
There is widespread concern that the present European work directives will adversely influence salary costs and might limit staff recruitment. In the UK, the state would have to reimburse the treatment costs under present legislation. Somehow, one cannot see the present system – with its multiplicity of health purchasers – allowing this to occur efficiently. It is perhaps time for the UK to re-think the arrangements for funding radiation cancer services: a central (i.e. single) purchasing system would solve at least some of the problems encountered over the last 15 years.
Governments will need to decide not only the reimbursement scales and the mixture of private and public partnership, but also the ratio of proton to ion facilities; the proportion of research and development; how to ensure equity of access and consequently the geographic spread of Centres per 10 million population. It is feared that the newest privately managed proton centres in the USA and Europe will be of most benefit to rich patients with cancer, who may have relatively poor indications for such sophisticated therapy. This is potentially a serious political issue. In the UK, the excellent progress made in development of Cancer Networks would allow CP therapy to be mapped onto these existing structures, even if patients may require referral abroad for therapy.
Several epidemiological and cost effectiveness studies – to estimate the likely demand for CP – were presented. It is clear that the indications will extend to previously unconsidered conditions such as lymphoma where dose can be confined better to target structures in order to reduce second malignancy, e.g. of the breast. Also, for example, in cancer of the cervix, ions can be used to treat a U-shaped nodal volume in the pelvis and the para-aortic nodes without the delivery of substantial bowel and kidney dose, which can be a problem in X-ray based IMRT.
For costs, a detailed German study from the University of Mannheim, in conjunction with Heidelberg, showed that in 10 chordoma patients the overall management was more cost effective by ions with respect to conventional radiotherapy combined with multiple surgical procedures (these are based on the available tumour control rates of 74% at 4 years in Germany [7] and 80% in 5 years in Japan [8]), and the best published result for X-ray based therapy (50% with stereotactic guided radiotherapy) [9]. The permitted reimbursement for chordoma treatment in Germany is \#8364;20 000, which is consistent with charges of around \#8364;1000 per fraction proposed at the new Heidelberg facility and for the Italian, French and Austrian projects. This cost must be balanced against the overall average cost of \#8364;81 470 for multiple surgical procedures, mostly for recurrences. The break-even cost point for carbon ions given in 20 fractions is reached if the control rate is 70%, whereas for 16 fractions a control rate of 65% would produce the same degree of economy. Interestingly, the reimbursement costs for radiotherapy appear to be lower in Europe than the USA, almost by an order of magnitude.
A Belgian consortium has proposed a pan-European solution based on Airbus Industries, where technical specifications and standards would be agreed. This may have been feasible a few years ago, but with up to seven industrial companies capable of providing the necessary equipment this seems unlikely – at least in the short term – unless there is significant industrial collaboration or consolidation.
The organization of the medical service for hadrontherapy is quite advanced in Italy: the TERA foundation team (Novara) consists of five radiation oncologists and a secretary, who have dealt with approximately 1000 requests per year resulting in 54 patients being accepted for therapy since 1997; this is because referrals were limited mostly to base of skull tumours at that time and also there were large numbers with metastatic cancer. The UK will soon need to establish some form of selection system and could learn from this experience.
Education of existing staff is clearly a major problem; few specialists in conventional radiotherapy are fully aware of the potential of CP therapy. It is noteworthy that Germany and Italy will soon be offering Masters degree-courses in this subject for medical physicists; Marie Curie Fellowships are also being promised. Contracts are already operational at some European centres for training given by experts from PSI, Switzerland.
There are proposals for e-health related developments with specific health networks for fast transfer of imaging data sets and grid-computing for rapid analysis and treatment planning. Progress is already being made between some Italian cancer hospitals in this respect. Experts at CERN consider CP departments to be an excellent test bed for e-health care; a similar system was proposed in the UK as part of the unsuccessful CASIM project. Existing projects such as EGEE (egee-intranet.web.cern.ch) and MIAS, developed for mammography in the UK (www.wiau.man.ac.uk/services/MIAS) should be considered for further development in cancer care.
The future scope for CP therapy is being actively assessed using evidence based medicine methods. As part of ETOILE, and linked to ENLIGHT, the following steps are being followed: (1) definition of a rationale based on physical and biological properties; (2) collegial screening of potential indications; (3) assessment of current therapy results using evidence based processes; (4) incidence and prevalence studies; (5) a second round of collegiate work as advisory committees to validate previous steps, ascertain indications and estimate potential recruitment; (6) proposition of a set of clinical trials in a European framework. Currently these multiple projects dealing with multiple anatomical sites of interest (cancers of the head and neck, thoracic, pelvic, central nervous system (CNS), and sarcomas) have reached stage 5.
New technologies
Work in CERN, China, and at the TERA foundation in Italy explores new technologies for the construction of smaller cyclotrons; the cyclinac concept, for example, involves use of existing small cyclotrons coupled to a linac booster with multiple klystrons capable of delivering carbon ions up to 400 MeV. Clatterbridge (UK) had requested funding for a similar research project previously, only to be repeatedly turned down because of cost – which could not be born by an NHS Trust – and, of course, demands for evidence based medical proof. Technical developments at Boston include the setting up of a dedicated radiosurgery system, with novel developments including miniature multileaf collimation, beam steering and focusing. The use of PET/CT imaging to verify dose distributions due to nuclear activation after initial 1–2 Gy exposures is being investigated in several centres worldwide. The use of robotic treatment tables is also increasing, as well as more sophisticated immobilization devices and methods for the correction of organ movement, with continuous modifications of tissue density with time, as in the lung, leading to more realistic four-dimensional treatment planning. Various Monte Carlo computational codes, containing fundamental physics interactions, are now capable of accurate dose predictions. Better methods for beam monitoring, dosimetry and the use of proton radiography show considerable promise. These developments should change the conduct of proton and ion therapy.
The topic of intensity-modulated proton therapy (IMPT) – using the "spot scanning" technique – was described in detail by Tony Lomax of PSI (Switzerland), who has pioneered this technique in proton therapy [10, 11]. In "passive scattering" – the usual method for creating a broad particle beam – all fields are homogeneous over the target volume, whereas in IMPT the fields are individually non-homogeneous. IMPT is used in patients where single optimized fields cannot provide similar target coverage or critical organ sparing cannot be achieved with single optimized fields. Patch fields are often required; these have greater dose uncertainties in the case of IMPT. The possibility of an enhanced risk of delivery error and additional physical effects such as secondary particle production exist when using IMPT. They have used IMPT in only 43 of 209 patients between 1999 and 2004, sometimes as boosts or in the later part of treatment. Considerable further research is required with this powerful newly developed method for radiotherapy delivery in the most difficult clinical situations to determine if outcomes are significantly improved.
Radiobiology
One interesting development is that of micro-ion beams, which can study selective radiation down to individual cells or even to sub-cellular structures. The bystander effect, where adjacent non-irradiated cells die, saturates at a relatively low dose and so cannot dominate the tumour cure probability achieved by radiotherapy, particularly at high dose per fraction [12]. Nevertheless, a fuller understanding of the mechanisms might usefully inform new pathways or mechanisms that might be exploited within normal tissues. Of particular relevance to CP beams is that bystander effects have been found with respect to malignant induction, so that the possibility of reducing the yield of second malignancies might enhance the potential benefits predicted for proton radiotherapy.
The thorny question of how best to use relative biological effect (RBE) within treatment planning arose repeatedly in the meeting. The longstanding international recommendation is to use constant weighting factors to compensate for RBE; the dose used in treatment is described as the cobalt equivalent Gray and this factor should be varied with dose per fraction in the case of ion beams; a similar correction although theoretically required for protons will only produce small differences in the equivalent dose. However, more recent British modelling developments based on biological effective dose have revised the method of dose conversion [13, 14], with full continual allowance for changes in RBE with dose per fraction, which the previous method did not fully allow. Some Centres, as at Darmstadt/Heidelberg and CHIBA, Japan, have used sophisticated methods to derive an RBE at various depths of the beam. In Germany 
/
ratios are also used to determine the actual dose given to patients. Calculations presented from the UK showed that the cobalt equivalent Gray method may incorrectly estimate the actual BED. Consequently, it was not surprising to learn that the Japanese authorities have further increased their single fraction lung dose from 28 Co-Gy Eq to 34 Co-Gy Eq; our estimated equivalent single fraction doses for X-rays using the newer method was 18 Gy instead of 28 Gy, hence the scope for dose escalation was entirely predictable. There were several pleas for more radiobiological funding in a clinically applied direction, with scope for improved model based understanding and outcome assessment.
Further presentations included alternative models for the estimation of second cancer risk, e.g. using the concept of organ equivalent dose with epidemiological data from A-bomb exposures and the risk of second malignancy developing in tissues after radiotherapy for Hodgkin's disease [15]. In the example of prostate cancer, risks were compared with conventional conformal radiotherapy: the use of IMRT appears to increase the risk by 15%, and by 20% if the X-ray beam energy is 15 MeV, whereas spot scanned protons should reduce the incidence by up to 50%.
The Japanese clinical experience
Carbon ion therapy is considered as "Highly Advanced Medical Technology" in Japan, where over 2000 patients have now received carbon ion radiotherapy within 40 different protocols. The dose distributions appear better to those of protons, although there remains some concern over the straggling effect beyond the Bragg peak and enhanced auto activation when compared with protons; this might result in more second malignancies than with protons. Prof. Tsujii and others gave detailed presentations of their experience in large pelvic and other inoperable sarcomas, prostate cancer, in recurrent rectal cancers and in T1 and T2 lung cancer. Their most recent publication is from the previous ENLIGHT and HCPBM meeting [8], where the specific patient numbers treated until 2003 can be seen. The following summary is of unpublished work; numbers of patients are small in some categories as these are essentially phase I/II dose escalation studies with more limited follow up of only a few years in the highest dose categories. Interested readers should seek out their future publications and judge the results for themselves; it is possible that higher rates of late toxicity might occur.
In rectal cancer recurrences, three dose levels have been tested in 16 fractions: 67.2 Co-Gy Eq, 70.4 Co-Gy Eq and 73.6 Co-Gy Eq; the overall local control is 73% at 2 years (surgery provides 10–20% control in various series). The control rate at the highest dose level is so far 100% and side effects are extremely low if Gore-Tex sheet displacement of bowel is also used. For chordomas the 5 year survival is 80% with low morbidity. In large mucosal malignant melanomas, concomitant carbon ion and chemotherapy pilot schedules in small numbers of patients appear to be giving better survival at 3 years than the use of both treatments given sequentially.
Their impressive experience in liver cancer [8] was not fully presented, although there was a statement that the only serious side effect appears to be grade 3 acute liver failure in 3% of cases – all of whom had pre-existing cirrhosis – and which tends to slowly recover. Perhaps it was thought that this was less of a problem in Europe, although the prevalence of biliary and hepatic cancers is increasing. It is clear that the results are extremely good and that the incidence of serious complications is very low in the pilot studies that have been completed since 1994. Urethral dose appears to be a problem in prostatic ion-therapy, where uniform dose has been given to the gland in 272 patients (stages T1–T3), the overall survival being 79.8% at 7 years with grade 2 or above rectal toxicity rate of only 0.4% and 2.85% for haematuria originating from the urethra or bladder; because of the latter, future protocols will use the "dose painting" technique – using spot scanning of targets – to reduce the urethral dose in the same way as selective peripheral loading of radioactive sources in prostate brachytherapy. Recently, research has started on pancreatic cancers and primary brain tumours in Japan. Techniques continue to be refined and fractionation protocols reduced to levels that cannot be used in the case of X-ray based therapy.
In lung cancer, 125 patients with 131 peripheral primary lung cancers were treated with increasingly hypofractionated carbon ions (see above) between 1999 and 2003 with 91.9% local control (100% for T1 stage and 78% for T2 stages) with crude survival rates of 54.7% and 46.1%, respectively. The 28 Co-Gy Eq single fraction results (66.7% local control) have been improved to 90% with 32 Gy; so far 34 Gy equivalent single fractions are without detectable side effects but these schedules have only been used over the past 12–18 months, so follow-up is short. The progressive changes in dose fractionation have allowed them to construct dose response curves so that any changes in technique can be rapidly compared with the most recent cohorts.
There is a clear concept that a Japanese fisherman can be rapidly screened for lung cancer, and flown at short notice to the National Institute for Radiological Sciences for treatment in a single day with high probability of success and almost no side effects; this is not a viable prospect for the UK population (e.g. for a Scottish crofter), until perhaps the French CP Facilities are complete, and even then our treatment slots may be limited to a few hundred patients per year. In contrast, a London stockbroker – if well informed – might negotiate privately funded CP treatment in Europe or elsewhere. Clearly, we face very difficult decisions in the UK.
The UK position
Our present commitments, and future dilemmas about how best to improve research in radiation therapy, can be summarized by the following points:
Conclusions
Just as we seek better early diagnosis of cancer, so we should also seek a safe alternative to radical cancer surgery through the development of safer radiation techniques; the fear of cancer will then be much alleviated by the benefits promised by nuclear physics. At a time like this, with the availability of such advanced medical technology, priority should be given to research and development aimed to find the best curative cancer therapies along these lines. As an absolute minimum, the UK needs a large high energy proton therapy centre as soon as possible and to continue its present cooperation with France in ion beams until longer follow up data is available in larger numbers of patients. There must be no complacency about this matter – short term expediency and prolonged prevarication may prove to be politically embarrassing in the long term, with additional costs and consequences to British cancer patients. The UK is capable of making a substantial contribution to this area of medicine, providing a positive decision is made quite soon.
Acknowledgments and declarations
The author was funded to attend this meeting entirely by ASCLEPIOS & Conseil Regional de Basse Normandie (France) and is a member of the UK EPSRC Medical Applications of Ion Beams Network and a Trustee of The Cyclotron Trust (UK). The author regrets not being able to mention the names of so many excellent speakers.
Received for publication June 23, 2005. Revision received July 21, 2005. Accepted for publication July 21, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
