Introduction
Radiation therapy has been widely used in the management of cancers, and has become one of the major modalities for cancer patient care. In the past two decades, modern radiation therapy technologies, three-dimensional conformal radiation therapy (3DCRT), and intensity modulated radiation therapy (IMRT) have been developed and gradually applied in clinics. Given that these techniques could concentrate irradiation doses on the tumor while sparing the adjacent normal tissues and organs, significantly improved local controls and survivals were obtained without an increase in irradiation-induced toxicity. Therefore, this technique becomes a common tool in cancer irradiation. However, a number of problems and defects are exhibited by these techniques. For example, these approaches deliver low doses to a large volume of normal tissues and expose the patient’s entire body to a low irradiation dose, thus increasing the risk of irradiation-induced second malignancies, particularly in young patients. Moreover, tumors that are highly resistant to irradiation, such as melanoma and sarcomas, cannot be controlled even with very high doses. Therefore, further exploration of new irradiation technology in the radiation therapy field is needed.
Particle beam therapy is one of the new irradiation tools under development, which shows promising potential in radiation therapy. Therefore, this type of therapy has drawn considerable attention from the oncology community over the past two decades. This article reviews radiation physics, radiobiology, and clinical experience in particle irradiation to present this frontier in radiation therapy and to promote it as one of the multidisciplinary approaches to cancer patient care.
History of particle therapy
Particle irradiation for cancers started in 1952 in Lawrence Berkeley National Laboratory, USA. The particle used for cancer therapy in large-scale clinical trials was the fast neutron in the 1970s. However, this particle was later abandoned because of intolerable toxicity and complications. Several particles, including proton, helium, carbon, neon, -π, and so on, have been clinically studied to treat malignancies. Lately proton and carbon beams have been demonstrated as the optimal particles for the clinical irradiation of cancers. In 1991, Loma Linda University Medical Center (LLUMC) established the first dedicated synchrotron for cancer therapy in the world [
1]. The National Institute of Radiological Science (NIRS) in Chiba, Japan is most famous for carbon beam irradiation. NIRS started to treat cancers in 1994, and over 6 000 patients have been treated to date [
2]. Preliminary results demonstrate that the local controls and survivals after carbon ion therapy were as good as those treated by surgery for prostate carcinoma, early-stage non-small cell lung carcinoma (NSCLC), and liver cancers. Moreover, particle therapy is a non-invasive treatment and is thus suitable for more patients, such as those suffering from cardiovascular diseases and poor liver function. The encouraging outcome of proton and carbon irradiation reported in the literature motivates an increasing number of hospitals to show interest in particle irradiation. Although particle irradiation is believed to be a new tool for radiation oncology in the coming years, not many hospitals in the world are planning or building particle therapy facilities because such facilities are very expensive.
Physics and radiobiology of particle irradiation
The physical dose distribution of particle beams, including proton and carbon, have been found to be the most appropriate for cancer therapy. At present, the irradiation beams widely used in clinics are photons, including the γ ray from 60Co, orthorvoltage X-ray, and megavoltage X-ray from liner accelerators. The physical dose distribution of photon is characterized by a build-up effect in the first centimeters, followed by an exponential decrease as the photon beam goes deep into the tissue. In contrast, the physical dose distribution of charged particle beams, such as proton and carbon ions, is shown in Fig. 1. The energy deposited at a certain penetration depth is inversely proportional to the ion energy, and nearly all energy is released at the end of beam track, forming a very high-dose region with little or no dose deposited beyond this region. This high-dose region is named the “Bragg peak” after William Bragg, who discovered it in 1904. When a 135 MeV proton beam penetrates a tissue, the deposited dose follows the following pattern: low dose at the first 10 cm, called the plateau phase and progression to the “Bragg peak” until almost no dose remains. For carbon beam, however, a “tail” dose exists beyond the “Bragg peak,” indicating that a fraction of the dose remains after the “Bragg peak.” Overall, the physical dose distribution of particle beams is optimal for cancer irradiation therapy. For example, for a tumor between 10 and 16 cm beneath the skin surface (Fig. 1), photon beam irradiation delivers a 100% to 150% dose to the tumor at the entrance irradiation dose, assuming that the entrance dose in the skin is defined as 100%, and the normal tissues in front of and behind the tumor receive 150% to 200% and 100% of the entrance dose, respectively. Thus, when a tumor is irradiated by photons, the normal tissues adjacent to it also receive significant doses, as high as or even higher than that the tumor receives. In contrast, the tumor receives 400% to 500% of the entrance dose for particle beams, whereas the normal tissues in front receive 100% of the entrance dose and those behind the tumor receive no dose. These favorable physical properties of particle beams permit highly precise dose localization for cancer therapy, which can spare normal tissues in front of and beyond the “Bragg peak.” However, “Bragg peak” width is quite narrow, and the diameter of tumors is typically large, which means that the “Bragg peak” can not cover the total tumor volume. Thus, a method called spreading-out of the “Bragg peak” is developed, which could be achieved via passive beam modification systems with ridge filters for proton beam with the fixed energy generated by cychrotron or by adjusting beam energies for protons from synchrotron. However, the dose in the plateau phase increases when the “Bragg peak” is spreading out (boarding) because of multiple beam scans with different energies, which adds up each plateau phase dose.
High-dose distribution must conform to the three-dimensional tumor shape to spare the adjacent normal organs and tissues. The most commonly used methods are scattering systems that enlarge beam profile and allow the geometric beam shape to conform to the tumor shape with collimators or blocks. A compensator should be applied to allow the distal edge of the high dose shape to conform to the tumor distal edge. A block and a compensator should be constructed for each irradiation beam, which is a time-consuming and labor-intensive task. Therefore, intensity-controlled raster scanning technique is developed, which uses small pencil beams of variable energy. For geometric matching with tumor shape, a scanning beam can actively draw any shape conforming to the tumor shape and can also deposit “Bragg peaks” at certain depths by adjusting beam energies. For the beam-scanning technique, the tumor is scanned one slice at a time, starting with the furthest slice and continuously to the closest slice. Finally, an extremely precise filling of the target volume is achieved, which could include up to 5 000 voxels within 3 min to 5 min with an individually calculated number of particles. With this particle irradiation technology, dose distributions are far superior to photon 3DCRT or IMRT as demonstrated by Chiang in lung cancer [
4]. In Chiang’s dosimetric study a very high dose could be concentrated on the lung tumor with good conformity in both proton and photon IMRT therapies. However, proton delivered a significantly less dose to the normal lung compared with photon. When stage I NSCLC (
n = 10) was irradiated to 66 Gy by photon 3DCRT, the mean lung V
5 (percentage of lung volume that receives≥5 Gy in total lung volume), V
10 (percentage of lung volume that receives≥10 Gy in total lung volume), and V
20 (percentage of lung volume that receives≥20 Gy in total lung volume) were 31.8%, 24.6%, and 15.8%, respectively, whereas they were 13.4%, 12.3%, and 10.9%, respectively when those patients were irradiated to 87.5 Cobalt Gray equivalent (GyE) by proton beam (
P = 0.002). For stage III NSCLC (
n = 15) the mean lung V
5, V
10, and V
20 were 54.1%, 46.9%, and 34.8%, respectively when lung tumors were irradiated to 63 Gy with photon 3DCRT, whereas those percentages were 39.7%, 36.6%, and 31.6%, respectively, when irradiated by proton with 74 GyE (
P = 0.002). Doses to the lung, spinal cord, heart, esophagus, and integral doses in all cases were lower with proton therapy compared with IMRT. Irradiation-induced complications would be significantly reduced in proton therapy compared with photon because of the low doses deposited to normal tissues and organs.
Linear energy transfer (LET) for ionization beams defines the energy deposited along the ion’s path. LET is found to be a universal parameter for radiobiological effects and correlates with relative radiobiological effect (RBE), that is, higher LET results in higher RBE. The track diameter of different ions is not proportional to LET but is dependent on particle energy. However, although RBE tends to increase with LET, each type of beam can produce a range of LET values, such that LET alone does not totally correlate with RBE. Moreover, given that carbon ions have larger particle size than protons, proton and carbon ion beams with the same LET display different RBE.
The RBE of proton beam is 1.0 to 1.1 in in vitro and in vivo experiments, indicating that the biologic effect of proton is slightly higher than that of 60Cobalt. However, RBEs for carbon are mixed with low LET in the entrance plateau and high LET in the “Bragg peak” area. Thus, the RBEs of carbon differ. Cell irradiation damage is the damage to DNA. DNA damage is of three types, namely, base, single stain break (SSB), and double stain break (DSB). SSB is more easily repaired than DSB damage. Moreover, low LET DSB damage is generally more readily repaired than high LET DSB damage because high LET causes “clustered” damage, whereas low LET damage is more detached.
After photon and proton irradiation, majority of DNA damage is SSB, which is repairable. However, after carbon irradiation with “Bragg peak” dose, at least 70% of DNA damage becomes DSB. Therefore, carbon produces significantly more severe DNA damage compared with photon and proton, thus inducing more cell death by apoptosis. As illustrated in Fig. 2 by colony-forming assay the survival curve bends with a “shoulder” at low dose ranges of photon irradiation (acute hypoxic cells), which indicates damage repair. In contrast, the cell survival curve is a straight line exponentially without a “shoulder,” suggesting the absence of repair after carbon irradiation within the “Bragg peak” dose (acute hypoxic cells). Therefore, carbon has high RBE in the “Bragg peak.” Compared with photon and proton irradiation, carbon has a stronger effect in terms of cell killing, which includes three aspects. First, cell radiosensitivity is increased by the carbon beam. Cell killing in photon and proton irradiation is generally cell cycle phase dependent, with the S and G0 phases being irradiation resistant, whereas cells in the G2 and M phases are sensitive. When cells are irradiated with “Bragg peak” of carbon beam, no radiosensitivity differences are observed among the G0, G1, S, G2, and M phases, and all cell cycling phases are sensitive. Hence, radiosensitivity is increased by carbon irradiation, especially for irradiation-resistant S and G0 phases. Second, hypoxic tumor cells that are highly resistant to photon and proton have no longer been resistance to carbon irradiation. For low-LET beams in hypoxic tumors, cells are 2.5 times to 3 times more radioresistant than normoxic cells, that is, cell killing exhibits oxygen dependence with an oxygen enhancement ratio (OER) of 2.5 to 3 for hypoxic cells, implying that hypoxic cells need 2.5 times to 3 times more photon dose to be sterilized. In contrast, carbon beam has an OER of 1, suggesting that hypoxic cells are as sensitive as oxygenated cells. As shown in Fig. 2, both the survival curves of acute and chronic hypoxia cells exhibit “shoulders” after photon irradiation. However, the “shoulders” disappear, and the slope of curve becomes steeper after “Bragg peak” irradiation of carbon ion, which indicates an increase in the radiosensitivity of hypoxic cells. From Fig. 2, the shape and slope of survival curves of the oxygenated, acute, and chronic hypoxic cells are evidently alike, which implies that photon-resistant hypoxic cells are as sensitive as oxygenated cells to carbon ion. Third, carbon irradiation can effectively kill intrinsic photon-resistant tumors, such as bone and soft tissue sarcomas, melanomas, and so on. Such malignancies are highly resistant to photon beams, but could be controlled by carbon irradiation. Overall, a carbon beam produces stronger cell killing in “Bragg peak” area compared with photon and proton beams. However, carbon therapy is a double-edged sword. If normal tissues and organs are exposed to “Bragg peak” of carbon beam, the damage is also severe.
Clinical experience
Up to 2009, a total of 78 275 patients had been treated by particles, including 67 097 by proton, 7 151 by carbon, 1 100 by -π, 2 054 by helium, and 873 by other particles. In radiation physics, radiobiology, and clinical practice, nowadays proton and carbon are recognized as the most optimal particles for cancer irradiation. The outcome reported in the literature showed superiority to treatment by photons in terms of tumor control, survivals, irradiation-induced complications, and quality of life. At present, 28 particle therapy facilities are in use worldwide, which are located in Japan, USA and Europe, and 18 facilities are currently under construction. A carbon therapy facility is available in Laizhou, Gansu, China, and a new one is under construction in Shanghai.
Proton therapy
LLUMC is the first to use a dedicated synchrotron to generate proton for cancer therapy since 1991. All cancers that are suited to photon irradiation are also indications for proton therapy. However, proton is more advantageous to tumors in the base of the skull, orbit, central nervous system, and pediatric patients.
(1) NSCLC. LLUMC treated 67 cases of stage I NSCLC by proton with 51 GyE to 60 GyE in 10 fractions over two weeks. All patients were not qualified for surgery because of cardiovascular disease or poor pulmonary function. None of the patients showed irradiation-related symptoms, such as early and late complications in the esophagus, heart, and lungs. With median follow-up time of 30 months, three-year local control and survival were 74% and 72%, respectively [
9]. Nihei reported 37 cases of inoperable stage I NSCLC with proton of 70 GyE to 94 GyE in 20 fractions, resulting in two-year progression-free survival of 80%, two-year overall survival of 84%, and locoregional progression-free survivals of 79% (stage Ia) and 60% (stage Ib) [
10]. Iwata reported the Nagoya experience on stage I NSCLC using proton (57 patients) and carbon beam (23 patients). Irradiation doses were 80 GyE in 20 fractions or 60 GyE in 10 fractions for proton and 52.8 GyE in four fractions for carbon. For all 80 patients with the median follow-up period of 35.5 months for living patients, three-year overall survival, cause-specific survival, and local control rates were 75% (IA: 74%; IB: 76%), 86% (IA: 84%; IB: 88%), and 82% (IA: 87%; IB: 77%), respectively. Irradiation-related toxicity was slight, with grade 3 of pulmonary toxicity in one patient [
11].
Proton therapy had been tried for locally advanced NSCLC. A retrospective study was conducted for proton therapy in 35 stage II to IIIb patients (stage II,5; stage IIIA, 12; stage IIIB, 18). Their median age was 70.3 years. The median proton dose given was 78.3 GyE (67.1 GyE to 91.3 GyE) without chemotherapy. Local progression-free survival was 93.3% at one year and 65.9% at two years. Four patients (11%) developed local recurrence, 13 (37%) regional recurrence, and 7 (20%) distant metastases. The progression-free survival rate was 59.6% at one year and 29.2% at two years. The overall survival rate was 81.8% at one year and 58.9% at two years. Grade 3 or greater toxicity was not observed. A total of 15 patients (42.9%) developed Grade 1 and 6 patients (17.1%) developed Grade 2. Proton irradiation for stage II-III NSCLC without chemotherapy resulted in good local control and low toxicity, and hence it could be intergraded with chemotherapy [
12].
Sejpal
et al. from M.D. Anderson retrospectively analyzed treatment-related toxicity in NSCLC treated by concurrent chemotherapy and irradiation (3DCRT, 74 cases; IMRT, 66 cases; proton, 62 cases). The study found fewer incidences of≥Grade 3 pneumonitis and esophagitis in the proton group (2% and 5%). On the other hand, the incidences were 30% and 18% in the 3DCRT group and 9% and 44% in IMRT group (
P<0.001 for all) despite higher irradiation dose to tumor in proton (74 CGE) than in 3DCRT (63 Gy) and IMRT (63 Gy) [
13].
Overall, from a limited number of studies, proton therapy showed encouraging outcomes for medical inoperable early-stage NSCLC and locally advanced NSCLC with mild irradiation-related toxicity.
(2) Prostate carcinoma. LLUMC is the first to report proton therapy for prostate carcinomas. As early as 1998, they reported 643 patients with localized prostate carcinoma treated with protons, with or without photons. Treatments were 74 GyE to 75 GyE at 1.8 GyE to 2.0 GyE per fraction. The overall disease-free survival rate was 89% at five years. The 4.5-year disease-free survival was 100% for patients with initial prostate-specific antigen (PSA) of<4.0 ng/ml, and 89%, 72%, and 53% for patients with initial PSA levels of 4.1 ng/ml to 10.0 ng/ml, 10.1 ng/ml to 20.0 ng/ml, and>20.0 ng/ml, respectively. Patients with post-treatment PSA nadir below 0.5 ng/ml did significantly better than those whose nadir values were between 0.51 and 1.0 ng/ml or>1.0 ng/ml. The corresponding five-year disease-free survival rates were 91%, 79%, and 40%, respectively. Minimal radiation proctitis was seen in 21% of patients, and toxicity of greater severity was observed in less than 1% [
14]. LLUMC later summarized 1 255 cases of prostate carcinoma irradiated by proton and reported PSA-free survival of 73% at 10 years. Acute complication of grade≥3 in gastrointestinal (GI) and genitourinary (GU) systems occurred in<1%, and late GI and GU toxicity of grade≥3 occurred in 1% [
15].
(3) Chordoma in base of skull. Igaki’s summary of the local control rates in the literature for proton therapy showed that the five-year local control rates were 46% to 73% at five years, in contrast 17% to 39% for photon treatment [
16].
(4) Hepatocellular carcinoma (HCC): Kawashima reported 30 inoperable HCC patients irradiated with proton. A dose of 76 GyE in 20 fractions was delivered over five weeks. After a median follow-up time of 31 months, local progression-free rate and overall survival rate at two years were 96% and 66%, respectively. Four patients died of hepatic insufficiency [
17].
Chiba
et al. conducted a retrospective study on proton beam therapy for HCC. A total of 192 lesions in 162 HCC patients were treated by proton with or without transarterial embolization and percutaneous ethanol injection. The patients were medically unsuitable for surgery because of hepatic dysfunction, multiple foci, or recurrence after surgery. The median total dose of proton irradiation was 72 Gy in 16 fractions over 29 days. The overall survival rate for all 162 patients was 23.5% at five years. The local control rate at five years was 86.9% for 192 tumors among the 162 patients. For 50 patients associated with the least impaired hepatic functions and a solitary tumor, the survival rate at five years was 53.5%. Patients had very few acute reactions to treatments and a few late sequelae during and after treatment [
18].
For medically inoperable or unresectable HCC, transarterial chemoembolization is the most commonly used treatment, but yields about 10% of three-year survival [
19]. Therefore, proton provides a good treatment alternative for HCC patients who are unfit for surgery.
Carbon therapy
Only a few hospitals worldwide provide carbon therapy, with the majority from NIRS [
2,
20]. This institute has treated over 6 000 patients using carbon beam. The high LET in the “Bragg peak” results in stronger radiobiological effect of carbon compared with photon and proton, which requires the physical irradiation dose to be transferred to biologic dose by a RBE of 3. Therefore, the carbon dose thereafter is expressed as GyE, which is the physical dose multiplied by three.
(1) NSCLC: Since 1994, NIRS has conducted a series of clinical trials to determine the most suitable irradiation schedule for early-stage NSCLC. NIRS started with a fractionation of 18 fractions in six weeks (18fr/6wks), and the total doses were escalated from 59.4 GyE to 95.4 GyE. Shift fractionation was then conducted at nine fractions in three weeks (9fr/3wks), four fractions in one week (4fr/1wk), and single dose. A range of doses was applied, beginning with low dose and escalating to high dose. A total of 322 stage I NSCLC patients were enrolled, with all patients being medically inoperable or refusing surgery. The study is still ongoing. The optimal fractionations are 90 GyE for 18fr/3wks, 72 GyE for 9fr/3wks, and 52.8 GyE and 60 GyE for stage Ia and Ib for 4fr/1wk, respectively. However, the optimal total dose for the single-dose schedule remains undefined. For the 131 patients who received 4fr/1wk and 9fr/4wks fractionation irradiation, the five-year local control rates were 98.6% and 89.7%, and the five-year overall survivals were 63.1% and 50% for stage Ia and Ib, respectively. For the 121 patients irradiated by single dose of 36 GyE, the five-year local control and overall survival was 79.2% and 63.6%, respectively. On the other hand, irradiation induced side-effects and toxicity were slight, with<5% of Grades 3 to 4 early toxicity in skin and lung, and 2% of Grade 2 late toxicity. Moreover, in this cohort, 56% of the patients were medically inoperable, and 66% were over 80 years old. The local control and overall survival treated by carbon are as good as that treated by surgery [
21-
23].
(2) HCC: NIRS conducted a trial similar to that for NSCLC to determine the optimal fractionation in the 1990s. They started with 49.5 GyE to 79.5 GyE in 15 fractions in 5 weeks (15fr/5wk), followed by total doses of 32 GyE to 69.6 GyE in 12 fractions in 3 weeks (12fr/3wks), 8 fractions in 2 weeks (8fr/2wks), 4 fractions in 1 week (4fr/1wk), and 2 fractions in 2 days (2fr/2d). Schedules of 4fr/1wk and 2fr/2d were repeated after a more advanced irradiation technique was used in 2005. To date, 392 HCC patients have been enrolled. The five-year local control rates were 93%, 86%, 86%, and 81%, respectively, for 4fr/1wk, 8fr/2wks, 12fr/3wks, and 15fr/5wks, comparable with that reported from surgery [
24]. Moreover, five-year local controls were slightly better for a tumor diameter≤5 cm than for>5 cm (90% vs. 86%). With respect to hepatic toxicity, which was evaluated via Child-Pugh scores three months after irradiation, 40% to 65% of the patients maintained their scores, and the rest had scores reduced by 0 to 1 point in different fractionations. At three to 12 months after irradiation, 40% to 55% of the patients still had unchanged scores, whereas others dropped by 1 to 2 points [
25,
26].
(3) Prostate carcinoma: Three different fractionation regimens were tried at NIRS, namely, 66 GyE in 20 fractions over five weeks, 63 GyE in 20 fractions over five weeks, and 57.6 GyE in 16 fractions over four weeks. The last regimen (57.6 GyE) has been demonstrated to be appropriate with minimum side effects. A total of 903 prostate carcinomas have been summarized, and carbon irradiation yielded five-year overall survival and PSA-free survivals of 94.9% and 90.9%, respectively. For high risk patients (GS 8-10 and T3), the five-year overall survival was 87%. Notably, 2.3% of GI and 5.1% of GU toxicity rates were recorded. The NIRS study on prostate carcinomas irradiated with carbon beam showed an outcome similar to that reported in surgery and proton and photon IMRT. Local control and survival for high risk patients were better than those treated by other modalities.
(4) Head and neck cancers: Several tumors are photon and proton resistant. Such tumors include adenocarcinomas (AC), adenoid cystic carcinomas (ACC), papillary carcinoma (PC), and malignant melanomas (MM). NIRS reported 195 cases (AC 27, ACC 70, PC 13, MM 85) irradiated with carbon ion. The five-year local control rates were 76%, 74%, 81%, and 75%, and the five-year overall survivals were 47%, 70%, 26%, and 75%, respectively, for AC, ACC, PC, and MM. However, their outcome would be dismal if treated with photons [
27].
(5) Osteosarcoma: This tumor is extremely resistant to photon irradiation. NIRS irradiated 81 patients with carbon ion, which yielded a five-year overall survival of 34%, and 42% of sarcomas in pelvis and sacrum were administered with doses of 64 GyE to 70.4 GyE in 16 fractions.
(6) Skull-base tumors: A national institute of heavy ion in Germany (GSI) treated 96 patients with chordoma in the base of skull by carbon ion with a median total dose of 60 GyE delivered in 20 fractions over three weeks, and they showed local control rate of 81% and 70% and overall survival of 91.8% and 88.5% at three and five years, respectively [
28].
(7) Soft tissue sarcomas and chondrosarcomas: Soft tissue sarcomas and chondrosarcoma are radioresistant and are not indicated for photon irradiation. Imai reported a five-year local control rate of 96% for unresectable sacral chordomas (
n = 32) after carbon irradiation with a median of 70.4 GyE [
29]. NIRS also used carbon to irradiate unresectable bone and soft tissue tumors (
n = 57), resulting in three-year local control and overall survival of 73% and 46%, respectively [
30]. GSI also employed carbon beam to treat spinal and sacrococcygeal chordomas and chondrosarcomas with similar results to that in NIRS.
(8) Recurrent rectal cancer: Locally recurrent rectal cancer after surgery is generally incurable with modern therapies. NIRS irradiated 65 patients with local recurrences after surgery with total doses up to 73.6 GyE, resulting in three-year local control of 82% and three- and five-year overall survivals of 65% and 55%, respectively. Irradiation-related toxicities were acceptable [
26].
Debate of particle therapy
The value of particle therapy in cancer therapy is currently being debated. In terms of physical dose distribution of particle beams, particle beams unquestionably deliver significantly less doses to normal tissues, which could reduce irradiation-induced mortality and morbidity and allow for dose escalation to improve disease control and survival without increased toxicity. On the other hand, the photon IMRT technique could satisfy most dose requirements in terms of dose conformity in tumors. However, considering the high cost of facilities, some people think that the development of proton therapy is not worthwhile because no difference is observed between the biologic effect of photon and proton. Moreover, clinical trials from single-arm studies show encouraging outcomes by proton therapy, but no clear evidence of benefit is shown in well-designed prospective studies [
32]. However, people with a positive attitude toward proton believe that proton beam should be further developed based on the optimum characteristic of physical dose distribution and available data. Heavy-ion therapy is more controversial. More favorable opinions have been given in Japan and Germany, and a number of heavy ion centers are being constructed based on their clinical experience and favorable outcomes in carbon therapy. In contrast, no hospitals are building heavy ion centers in the USA at present despite the heavy-ion therapy started from Lawrence Berkley National Laboratory, USA. The predominant concern is the irradiation injuries of normal tissues and organs in heavy ion therapy.
Challenges to particle therapy techniques include dose uncertainty in the moving targets of lung and liver cancer patients, dose calculation inaccuracy in inhomogeneity regions, and so on. Hypofractionation has been used in particle therapy, but further study is needed to determine the optimal fractionation.
In summary, the following points could be drawn from this review: (1) Preliminary data in the literature on particle therapy (proton and carbon beams) show more promising local control, survival, and significantly reduced irradiation-related toxicity for several cancers compared with photon 3DCRT/IMRT; (2) The local control and survival of patients treated by proton, especially carbon, are as good as those of patients treated by surgery for early-stage NSCCL, HCC, prostate carcinoma, and some radiation-resistant head and neck cancers. Nevertheless, good outcomes should be reconfirmed by large-scale prospective clinical trials; (3) Particle irradiation provides more opportunities of treatment for cancer patients, especially for elderly patients with medical co-morbidities; (4) Proton therapy is at a mature stage with over 60 000 patient treated, but only 7 000 cases have been treated with carbon ion. Given the lethal damage to both tumor and normal tissues, an accurate irradiation is absolutely necessary. Moreover, such an approach will require more advanced technology and experience, as well as a greater understanding of radiobiological effects.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(195KB)
