Methods

A systematic review of the literature was carried out through a search of the CINAHL, MEDLINE, and ScienceDirect databases. The keywords are paediatric, proton therapy, ion therapy, carbon therapy, PBT, and hadron therapy. Papers were selected on the basis of their relevance to the topic. All the articles were available in English, and any articles that appeared to have vested interests were identified. Articles presented in the form of case studies were included if they added any significant information.

The physical principles of x-rays and particles

Particles offer considerable advantages over X-rays regarding the sparing of normal tissues. This is due to the physical characteristics of the particle beam when compared with an X-ray beam. X-rays are highly penetrating electromagnetic waves which deliver dose throughout any volume of tissue irradiated.Reference Metz¹¹ This means that X-rays will always deliver substantial unnecessary dose both proximally and distally to the tumour volume. Furthermore, the depth at which the maximum dose of X-ray radiation is delivered (Dmax) ranges from as little as 0.5 cm, to a maximum of 3 cm, depending on the energy utilised. Tumours are often located deeper than these ranges and as X-ray beams decay exponentially throughout the body, a higher dose is invariably delivered to the healthy tissue anterior to the tumour,Reference Loeffler, Smith and Suit¹² while the tumour is treated in the region of the beam where the energy deposition is falling off. This can be overcome with the use of beams from multiple angles which centre on the tumour, allowing the dose to accumulate within the tumour volume. However, since the beam travels throughout the entire thickness of the body, all normal tissue from the beam entrance to its exit will be affected, increasing the total or integral dose.

The absorbed dose of a particle beam, however, increases very gradually with increasing depth and then suddenly rises to a peak at the end of the particle range, which is known as the Bragg Peak (Dmax).Reference Loeffler, Smith and Suit¹² The particle beam can be directed and spread out to create an area of uniform dose, ensuring that the Bragg Peak effect occurs precisely within the tumour volume, something that cannot be achieved using X-rays. As particles have little side scatter, and their range is energy-dependantReference Loeffler, Smith and Suit¹² the dose received by the surrounding healthy tissue is much less than that received by the tumour. This means that the dose beyond the Bragg Peak is essentially zero, which allows for the sparing of the healthy tissue around and beyond the tumour volume. In contrast to an X-ray beam, a single particle beam can be shaped to deliver a homogenous dose of radiation to an irregular 3D volume.Reference Hug, Nevinny-Stikel, Fuss, Miller, Schaefer and Slater¹³ The improved dose distribution obtained when using particles has been found to reduce the integral dose by more than 50% in many treatments,Reference Jones and Burnet¹⁴ reducing the likelihood of secondary radiation-induced malignancies occurring later in life.Reference Miralbell, Lomax, Cella and Schneider¹⁵

The exit dose experienced when using CRT and intensity-modulated radiotherapy (IMRT) can restrict the use of certain angles as the beam frequently exits through critical structures distal to the target, causing unnecessary damage. These critical structures can also restrict the total dose delivered when using CRT and IMRT as the tolerance dose of these structures may be less than the dose required for tumour control, compromising the treatment. Simpler beam arrangements are possible in PBT owing to the absence of any dose exiting through distal critical structures (owing to the Bragg Peak effect) allowing for increased dose in these situations, improving the probability of cure.

Current role of particle therapy in paediatrics

Particles are currently used in many centres around the world to treat various malignancies in both adult and paediatric population and each centre has the capability to treat several thousand patients a year. Approximately 55,000 patients have been treated with particle therapy since its introduction in 1954.Reference Goitein and Jermann¹⁶

Paediatric malignancies currently treated include medulloblastoma, craniospinal ependymoma (boost), pineal tumours, astrocytoma, retinoblastoma and orbital rhabdomyosarcoma. The centre at Loma Linda University is also exploring the use of particles to treat non-cancerous paediatric diseases such as intractable childhood epilepsy.Reference Slater¹⁷

There is currently only one, short range, 62 MeV facility in the UK (Clatterbridge) which only treats adult patients with tumours of the eye. Over 1,400 patients have received proton therapy at Clatterbridge with an excellent local control rate of 98%.Reference Jones¹⁸ A higher energy facility is required in the UK to treat deeper situated cancers and to transfer these exceptional control rates to other cancer sites. If the British Government fails to invest in a higher energy facility, informed parents may take their children abroad to receive the best possible treatment available. Jones suggests that these referrals abroad would have a severe and detrimental effect on UK oncology treatments.

Glimelius et al. (2005) estimated that approximately 120 out of 330 new paediatric cancer cases a year benefit from particle therapy in Sweden.Reference Glimelius, Ask and Bjelkengren¹⁹ With 1,400 new cases a year in the UK, extrapolation of these figures would mean approximately 500 children a year would benefit from a particle therapy centre in the UK.

Clinical advantages of particles for paediatric malignancy treatment

The avoidance of even moderate amounts of healthy critical structures is extremely important in the treatment of paediatric patients and combined with the fact that 50% of solid paediatric tumours occur in the brain, an increased need for improved conformity and reduced integral dose becomes even more essential in these treatment areas. Head and neck tumours are frequently situated near dose limiting critical structures, making it difficult to deliver a curative dose of CRT. Because of this, cranial irradiation can cause many late sequelae to the irradiated child such as losing hearing ability and sight, interfering with intellectual development, leading to obesity, increasing the chance of secondary malignancies occurring and affecting the child's physical growth. These late-effects of cranial irradiation are caused by a number of critical factors including total radiation dose, dose per fraction, volume of tissue irradiated, age of patient at treatment, the anatomic area irradiated and the combination of radiotherapy with other treatment modalities, that is chemotherapy and radiotherapy.Reference Halperin, Constine and Tarbell²⁰

Modern 3D treatment techniques such as IMRT and PBT have helped improve dose conformity to the treatment site and numerous dosimetric studies have been performed, comparing the dose distribution of PBT with CRT and IMRT in paediatric malignancies. The fact that many long-term effects of treatment appear to be related to dose and volume of tissue treated warrants the need for this research into less damaging treatment techniques that provide at least similar cure rates.

It has been found that the predominant long-term effect of moderate to high-dose cranial radiotherapy is reduced intellectual capacity.Reference Kirsch and Tarbell²¹ A recent report found that children exposed to 18–24 Gy of cranial irradiation were three to seven times more likely than the national average to be unemployed later in life,Reference Mulhern and Palmer²² highlighting the damaging late-effects of irradiation. It has also been found that brain tumour survivors who are irradiated at a younger age are more vulnerable to cognitive dysfunction before radiotherapy.Reference Mulhern, Hancock, Fairclough and Kun²³ A review of 22 studies of childhood brain tumour sufferers, treated with radiotherapy, found that those treated at a younger age had a 14-point larger deficit in IQ when compared with children treated at an older age.Reference Mulhern, Hancock, Fairclough and Kun²³

The reduction of dose to healthy tissues is critical in this area and Lin et al. (2000) compared the normal tissue dose-sparing capabilities of proton radiotherapy with photon therapy in the treatment of the posterior fossa.Reference Lin, Hug, Schaefer, Miller, Slater and Slater²⁴ With identical coverage of the target volume, doses to 50% and 10% of the temporal lobe volume were limited to 2% and 67% when using protons, compared with 56% and 100% with 3D conformal photons, respectively. This demonstrates that particle therapy allows for escalation of tumour dose while reducing dose to non-target tissue, potentially decreasing the chance of severe late-effects, including cognitive dysfunction, while increasing the chances of survival. Another common option available to children suffering from brain cancer is open surgery, but this treatment method can prove extremely daunting to a young child and can also pose a considerable risk to the patient. In these instances, particle therapy offers an excellent alternative.

Treatment of rhabdomyosarcoma, the most common primary orbital malignancy in children and the most common childhood soft-tissue sarcoma,Reference Souhami and Tobias² also presents its own distinct problems. During radiotherapy of this particular malignancy the exit dose needs to avoid both the ipsilateral and contra lateral critical structures. Ninety percent of children treated for rhabdomyosarcoma fortunately survive,Reference Hug, Adams, Fitzek, De Vries and Munzenrider²⁵ but the long-term side-effects such as cataracts, orbital hypoplasia, conjunctivitis, keratopathy, corneal ulcers, exophthalmia, vitreous haemorrhage and hypopituitarism, caused by irradiation of developing neighbouring structures can be severe. A recent study on the long-term effects of orbital radiation on paediatric patients found that all patients in the cohort experienced soft-tissue changes and 50% of patients developed bone changes that resulted in visible facial asymmetry.Reference Guyuron, Dagys and Orth²⁶

Yock et al. (2005) conducted a study on seven patients aged between five and eight, comparing PBT with CRT in orbital rhabdomyosarcoma.Reference Yock, Schneider, Friedmann, Adams, Fullerton and Tarbell²⁷ These patients were treated with PBT, and retrospective photon plans were created, to compare the dose distribution differences between the treatments. This study concluded that PBT provides excellent dose distribution to the tumour while sparing healthy tissue. All the structures evaluated in this study received lower doses from the PBT treatment plans and a dramatic benefit to sparing the contra lateral structures was recorded. The reduced incidence of late-effects of PBT are also highlighted in this study. Follow-ups were conducted on the patients and it was found that none of the patients suffered from side-effects of the treatment, and none of the patients required hormone replacement from scattered dose received by the hypothalamic/pituitary axis (a commonly experienced side-effect after CRT).

Studies on medulloblastoma, one of the most common paediatric CNS tumours,Reference Yock, Schneider, Friedmann, Adams, Fullerton and Tarbell^27–Reference Livesey, Hindmarsh, Brook, Whitton, Bloom, Tobias, Godlee and Britton³⁰ have found that particles deliver superior target dose coverage and sparing of healthy tissues, when compared with both IMRT and CRT. Advances in the treatment of this particular malignancy, which was once considered incurable, have led to an increase in survival of children with this disease.Reference St Clair, Adams and Bues²⁸ With this improvement in long-term survival of children suffering from this disease, there is increased concern regarding late side-effects. Dramatic differences were highlighted throughout these studies in most of the critical structures examined.Reference Yock, Schneider, Friedmann, Adams, Fullerton and Tarbell^28–Reference Livesey, Hindmarsh, Brook, Whitton, Bloom, Tobias, Godlee and Britton³⁰ By comparing the dose volume histograms (DVH), it was found that the use of particles in the treatment of medulloblastoma could reduce the dose to 50% of the pituitary volume from 84.5% with CRT to 26.7% and 0.5% with IMRT and protons, respectively,Reference St Clair, Adams and Bues²⁸ significantly reducing the probability of late-effects occurring.

With neuroendocrine dysfunction, such as growth hormone deficiency (GHD), being a common dose and site-related sequelae following irradiation of the brain,Reference Oeffinger and Hudson⁹ reduction of unnecessary dose to glands such as the pituitary (which can result in reduced height), becomes essential. GHD is reported as the most common endocrinopathy following cranial irradiation.Reference Oeffinger and Hudson⁹ Livesey et al. (1990) supports this stating that out of a total of 144 paediatric patients treated with conventional cranial radiotherapy, 140 had evidence of GHD.Reference Livesey, Hindmarsh, Brook, Whitton, Bloom, Tobias, Godlee and Britton³⁰ Treatment of this group of patients with growth hormone substitutes fortunately results in near normalisation of stature, but the annual cost of GHD has been estimated at a massive £9,170 per patient.Reference Lundkvist, Ekman, Ericsson, Jonsson and Glimelius³¹ The use of particles in this treatment would not only reduce the need for continuous use of growth hormone substitutes, but would also drastically reduce on-going health care costs.

Medulloblastoma treatments also require irradiation of the whole of the spinal cord as the disease has a propensity to metastasise via the cerebrospinal fluid.Reference Souhami and Tobias² This brings a substantial amount of healthy tissue along the spinal axis into the treatment field. This method places other vital structures at risk including the thyroid. Ionising radiation that penetrates the thyroid causes nodule development which can lead to thyroid cancer later in life.Reference Hancock, Cox and McDougall³² St Clair et al. (2004) concluded that the risk of thyroid cancer was 16 times the expected risk in their cohort of paediatric patients treated with CRT.Reference St Clair, Adams and Bues²⁸ By reducing the amount of radiation received by the thyroid, PBT can reduce the subsequent risk of thyroid cancer occurring. Studies have found that the volume of thyroid receiving 10 Gy can be reduced from 100%, when using CRT, to only 7% with PBT.Reference Lee, Bilton and Famiglietti³³ This reduction in dose to the thyroid would have remarkable effects on the child's future quality of life, not only reducing the risk of secondary malignancy but also minimising the chance of hypothyroidism occurring from 33% to >1%.Reference Livesey, Hindmarsh, Brook, Whitton, Bloom, Tobias, Godlee and Britton³⁰ Raney et al. (2000) support these findings by stating that 8% of patients receiving CRT require hormone replacement later in life.Reference Raney, Anderson and Kollath³⁴

Recent research has also concluded that particle therapy is much quicker and less complicated than IMRT in the treatment of medulloblastoma.Reference St Clair, Adams and Bues²⁸ The spinal IMRT technique in this particular study required the use of seven isocentric beams, which were chosen to avoid critical structures, compared with the simpler single beam arrangement used in the particle treatment. The use of multiple beams during IMRT not only increases the complexity and time taken for treatment, but also increases the integral dose,Reference Hug, Sweeney, Nurre, Holloway, Slater and Munzenrider³⁵ escalating the chances of secondary malignancies and long-term sequelae.

Research into the treatment of medulloblastoma and rhabdomyosarcoma has found that the use of particle beams reduced the expected incidence of radiation-induced secondary malignancies by a factor of >2 in rhabdomyosarcoma treatments and by a factor of 8–15 when compared with either IMRT or CRT.Reference Miralbell, Lomax, Cella and Schneider¹⁵ The author of this particular study concluded that this potential reduction in secondary cancers when using particles represents a significant case in favour of the use of PBT for most radiotherapy treatments in paediatric oncology.

Exposure of the paediatric chest to ionising radiation is also common in many types of radiotherapy treatments including craniospinal, mantle, chest, mediastinum and lung fields. These treatments often expose the lungs, kidneys, liver, heart and developing breast tissue to significantly high doses of ionising radiation. These non-target organs are then at increased risk of late sequelae developing as a direct result of treatment, including secondary malignancies. Treatment of medulloblastoma and Hodgkin's disease in particular delivers a significant exit dose to the developing breast tissue of female paediatric patients and is accountable for an increased risk of breast cancer. It has been shown that girls treated for Hodgkin's disease between the ages of 10 and 16 face a significantly increased risk of breast cancer later in life.Reference Bhatia, Yasui and Robinson³⁶

At present most patients with Hodgkin's disease are cured,Reference Van Leeuwen, Travis, DeVita, Hellman and Rosenberg³⁷ but are at increased risk of late sequelae as a result of the treatment. It has been estimated that children treated for this malignancy are 22 times more likely to develop a secondary cancer than the general population.Reference Donaldson³⁸ In 2002, Ng et al. studied a large population of patients (n > 1,000) with Hodgkin's disease and found that the risk of death from late complications was greater after 14 years than the risk of death from the initial disease.Reference Ng, Bernardo and Weller³⁹ Studies have found that the use of particle therapy in this particular malignancy can eliminate any exit dose, substantially reducing subsequent morbidity.Reference Kirsch, Ebb, Hernandez and Tarbell^40, Reference Suit⁴¹

Irradiation of the heart in CRT is also often unavoidable in paediatric mantle, chest and spinal radiation fields. Hull et al. (2003) estimates that 16% of children who receive chest or spinal irradiation would have considerable cardiovascular morbidity within 20 years of treatment owing to the asymmetric growth of the irradiated heart.Reference Hull, Morris and Pepine⁴² Similar late-effects are noted in the lungs of children treated with these fields, with radiation being associated with chronic cough, exercise related dyspnoea, and secondary lung cancer.Reference Souhami and Tobias² Miralbell et al. (1997) found that paediatric patients receiving spinal irradiation were 4.3 times more likely to have restrictive lung disease as a consequence of treatment, despite only a small portion of the lung being in the treatment field.Reference Miralbell, Lomax and Russo²⁹

It has been found that in the treatment of chest, mantle and spinal fields, particles can not only completely eliminate any unnecessary dose to the heart, lungs and breast tissue, but can also help avoid the liver and kidneys.Reference Hug, Nevinny-Stikel, Fuss, Miller, Schaefer and Slater^13, Reference Miralbell, Lomax and Russo^29, Reference Lee, Bilton and Famiglietti^33, Reference Van Leeuwen, Travis, DeVita, Hellman and Rosenberg³⁷ Kirsch et al. (2005) found that due to the rapid dose fall-off experienced with particles, the same biological dose could be delivered to the target as conventional RT while reducing the dose outside the target volume, when treating Hodgkin's disease,Reference Van Leeuwen, Travis, DeVita, Hellman and Rosenberg³⁷ minimising the risk of severe late-effects occurring.

Radiation therapy to the abdomen, pelvis and spine may also significantly affect either the ovaries or the testes. Particles, again, have the ability to eradicate any scattered dose received by these organs.Reference Lee, Bilton and Famiglietti³³ It has been found that PBT can reduce the volume of ovary receiving 10 Gy from 86% when using CRT, to 0% with PBT, in the treatment of pelvic sarcomas.Reference Lee, Bilton and Famiglietti³³ The use of particles in this situation would help preserve the future fertility of patients undergoing this treatment.

In their paper of 2001, Neglia et al. report on 116 survivors of childhood malignancies from the childhood cancer survivor study cohort who developed subsequent malignant and benign tumours of the CNS.Reference Neglia, Friedman and Yasui⁸ The researchers arrived at three main conclusions. First, the single most significant risk factor for the development of a further CNS tumour in survivors of childhood malignancies is exposure to therapeutic radiation. Second, after adjustment for radiation dose, neither initial tumour diagnosis nor chemotherapy was related with the risk of CNS tumours. And finally, the higher risk of subsequent glioma in children irradiated at a very young age (before 5), may reflect greater susceptibility of the developing brain to radiation.Reference Neglia, Friedman and Yasui⁸

Potential disadvantages

If particle therapy is so good then why isn't it available in the UK? The main disadvantage with particle therapy is the initial expensive cost of the equipment and the subsequent increased on-going operating, maintenance and personnel costs. The recently completed MD Anderson centre in America cost 120 million pounds.Reference Sisterton⁴³ The initial cost is reflected in the individual treatment and Goitein and Jermann (2006) calculated the cost of a complex intensity-modulated proton treatment at 2.4 times the cost of an IMRT plan.Reference Goitein and Jermann¹⁶ However, when discussing the cost of particle therapy the initial and running costs must be weighed against the improved health gain and reduced late-effects and morbidity. The superior dose distribution experienced with particle therapy would undoubtedly reduce long-term toxicity and prevent late side-effects occurring, especially in paediatric malignancies. This in turn would help to reduce the need for long-term drug use, (growth hormone replacement for patients receiving irradiation of the pituitary gland), or extensive follow-up and screening (mammography for female patients treated for Hodgkin's disease). Combining these facts with the simpler beam arrangements available makes the use of particles even more attractive. Jones (2006) estimated that a single UK particle centre should recoup its own initial running costs within 6 years if it is able to treat 2,500 patients by its third year of operation.Reference Jones¹⁸

Orecchia et al. (1998) estimated the cost per patient of proton therapy to be approximately £4,200 compared with £2,000 for CRT £12,500 for an intensive course of chemotherapy and £4,500 for conformal radiotherapy.Reference Orecchia, Zurlo and Loasses⁴⁴ Considering PBT is still relatively new, and the costs of new technologies have a tendency to decrease, the price for this highly effective treatment seems extremely appealing. However, a full cost benefit analysis, weighing the initial and subsequent running costs of a particle therapy centre against the potential savings is beyond the scope of this paper but would provide a valuable future contribution to the evidence base.

The particle therapy centre also has the drawback of its reliance on a single accelerator. This accelerator injects the particles into a synchrotron and particles are then further accelerated by the synchrotron before being extracted and delivered to the different treatment rooms. This reliance on one accelerator means that if the accelerator breaks down then the whole department comes to a stand still. This could prove very distressing for paediatric patients and their families who may have travelled long distances to receive their treatment.

Another disadvantage of PBT when treating paediatric patients is the intimidating size of the gantry. An isocentric rotating gantry is required for PBT which consists of a huge rotating cylindrical structure containing the beam-bending magnets. These structures can be massive; weighing over 100 tonnes for protons and 200 tonnes for ionsReference Jones¹⁸ and these massive gantries can prove very distressing for small children.

The main reason for acute and late-effects from PBT is non-conformity of treatment dose which places a great deal of pressure on the physicist to ensure maximum accuracy when planning the treatment and the radiotherapist when positioning the patient. The results of treatments do show, however, that this is achievable.

Discussion

Many of the studies mentioned above did not actually treat patients with particle therapy but are quantitative planning studies that have used generated particle plans. The resulting plans and DVH have then been analysed and compared with the IMRT and CRT plans and the doses received by amount of normal tissue spared by photon therapy was then calculated. This method means that follow-up of the patients in these studies is impossible and long-term follow-up of patients is essential to assess the actual value of PBT. The studies that did actually use PBT were small scale, and retrospective, limiting the results of the studies. Long-term follow-up from larger studies into particle therapy is required to confirm and record the expected reduction in late sequelae. Whether phase III studies that directly compare particles with conventional therapies are possible, however, remains very doubtful as this may be seen by some as unethical. Clinicians and patients are already aware of the dose advantages associated with PBT and may therefore refuse to allow their patients to participate in this type of study. For accurate assessments to be made, prospective studies will need to be performed on a large patient populations, with long-term follow-ups helping to evaluate the expected improvement in quality of life.

Measuring this quality of life, however, can prove difficult, especially in the paediatric population. One such example of this is a T-cell lymphoma survivor with severe restrictive lung disease and moderate cardiomyopathy who described her health as ‘great’.Reference Oeffinger and Hudson⁹ The use of quality-adjusted life years (QALY) may prove a more appropriate method to asses the cost-effectiveness of the differing treatment techniques.

The building of a national centre would mean children travelling long distances from all over the country to be treated at a single centre. These children would therefore need to be accompanied by parents or carers. This would result in missed schooling for the children and absence from work for their carers. Although this may be difficult in the short-term, the eventual reduction in side-effects and the subsequent decrease in future hospital visits, would far outweigh this temporary problem. This situation is nevertheless, more attractive than the scenario painted by Jones (2006), who suggests that between 5,000 and 12,000 patients will demand treatment abroad within the next decade if the UK does not develop a PBT centre. This situation would undoubtedly have a catastrophic effect on British oncology.Reference Jones¹⁸

Clinical results from countries using particles are very impressive and consistently highlight the reduced side-effects experienced as a result of the improved dose distributions. Combining this with the fact that the United States, Japan, China and many parts of Europe are investing billions of dollars in the expansion of particle facilities emphasises the significant benefits of this treatment technique and the need for development of particle beam facilities within the UK.Reference Jones¹⁸

The treatment of childhood malignancies alone, however, could not possibly justify the construction of a particle therapy centre but the paediatric population must be considered in the eventual design of a national centre. Waiting areas and treatment rooms should be built with children in mind and staff with knowledge of children's needs should be employed.

Conclusion

All of the above studies conclude that the dose distribution and resultant sparing of healthy tissue realised with PBT is far superior to that of CRT. These studies have also demonstrated substantial advantages in PBT, when compared with CRT and IMRT, not only because PBT offers a significant prospect of tumour control but also in the reduction in the probability of serious late-effects occurring, since the dose received by normal tissue is minimised. PBT has also been shown to be quicker and less complicated that CRT and IMRT. Reductions in side-effects in a population of patients with a high probability of cure and a lengthy life expectancy will not only substantially improve the individuals future quality of life, but will also help reduce the accumulative cost of treating each patient.

Clinical research with long-term follow-up, is however necessary to assess the real value of particles in paediatric tumour treatment and to evaluate whether the benefit in the related long-term side-effects persist. For many children suffering from cancer, particle therapy has the potential to limit the late-effects of radiotherapy, including the risk of second malignancies, and therefore offers a valuable advantage and increased long-term hope for the children and their families.

The development of a PBT centre should therefore include paediatric patients with the construction of child friendly waiting areas and treatment rooms, and the employment of staff who are comfortable working with children and their families. Accommodation should also be made available to the patients and immediate family to reduce their expenses and ease their worries.