Hatano Y., Katsumura Y., Mozumder A. (Eds.) Charged Particle and Photon Interactions with Matter - Recent Advances, Applications, and Interfaces

Подождите немного. Документ загружается.

650 Charged Particle and Photon Interactions with Matter

24.2.2.1 head and neck Cancers

Carbon-ion radiotherapy was rst used for head and neck cancers in 1994. Most patients had locally

advanced lesions or postoperative recurrence, with little expectation of cure from other therapies.

In the rst phase I/II study, a fractionation regimen of 18 fractions over 6 weeks was employed,

after which the second phase I/II study using a fractionation regimen of 16 fractions over 4 weeks

was introduced. The comparison of the results of these two studies showed no difference in terms

of side effects and local control (Mizoe etal., 2004). Since April 1997, phase II studies using

57.6–64.0GyE in 16 fractions over 4 weeks, the optimal dose determined in the second study, have

been

conducted, and almost no severe side effects have been observed to date.

With

respect to effectiveness, local control has been favorable, being 80%–90% in adenocarci-

nomas, adenoid cystic carcinomas, and malignant melanomas. For the treatment of bone and soft

tissue sarcomas in the head and neck region, it was found necessary to increase the total dose to

improve local control. Thus, a total dose of 70.4GyE in 16 fractions was used, and local control has

far been quite satisfactory.

24.2.2.2

skull

ase

umors

As the term implies, these tumors develop in the base of the skull, and include primarily chor-

domas and chondrosarcomas. Because the tumors frequently develop in the vicinity of the brain

stem, optic nerve, and large vessels, complete surgical resection is usually quite difcult. While the

advent of proton therapy has improved the treatment results, it has become clear that the recurrence

of chordomas 5 years or more after treatment is not rare (Munzenrider and Liebsch, 1999). In this

respect, carbon-ion radiotherapy has the potential to improve long-term survival (Schulz-Ertner and

Tsujii, 2007).

The most common types of tumors treated with carbon-ions at NIRS are chordomas and menin-

giomas. As the dose was increased in dose escalation studies, an improvement in tumor control was

observed, and there were no severe adverse reactions. A fractionation regimen of 60.8 GyE in

16 fractions

over 4 weeks yielded the best local control with acceptable morbidity.

Number

700

600

500

400

300

200

100

1994

126

159

168

188

201

241

276

286

277

110

113

138

166

189

324

411

476

495

333

396

437

549

684

642

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Year

Research

Practice

Total no. of patients: 4504

Annual number of patients in carbon therapy

(June 1994 to February 2009)

Figure 24.2 Number of patients enrolled for carbon-ion radiotherapy (June 1994 to February 2009).

Cancer Therapy with Heavy-Ion Beams 651

24.2.2.3 lung Cancer (non-small-Cell lung Cancer)

NSCLCs were divided according to tumor location into peripheral type and central type. It was

thought necessary to change the fractionation regimen depending on the location, since the toler-

ance of the surrounding normal tissue would be different between the two. Clinical studies were

conducted to establish a short-course radiotherapy regimen for stage-I (T1-2/N0/M0) peripheral-

type cancer. The treatment was started with the use of 18 fractions over 6 weeks, and then the

number

of fractions and treatment time were carefully reduced.

conducted a clinical study to develop a short-course, hypofractionated radiotherapy.

Treatment regimens of nine fractions over 3 weeks and four fractions over 1 week were evaluated,

with both showing little severe toxicity and a local control rate of >90% (Miyamoto etal., 2007).

A phase I/II dose escalation study using a single fraction of carbon beam has been under investi-

gation for peripheral-type stage-I NSCLC. Single-fraction irradiation is the ultimate regimen utiliz-

ing the features of a carbon beam. We intend to adopt this regimen as a clinical practice as early as

possible. The respiratory-gated irradiation method, a technique that uses synchronization with the

respiratory movement of target organs, was developed and effectively used to reduce unnecessary

dose

to the surrounding normal lung in these clinical trials.

24.2.2.4

liver

Cancer (

hepatocellular

Carcinoma)

All

the clinical studies of liver cancer dealt with patients for whom other therapies were not expected

to provide sufcient effects or were not indicated. In the rst phase I/II study, patients were treated

with irradiation in 15 fractions over 5 weeks, and safety and efcacy were conrmed. The 3- and

5-year local control rates were both 81% (Kato etal., 2004). In the second phase I/II study, the dose

was escalated using increasingly shorter irradiation times of 12 fractions over 3 weeks, 8 fractions

over 2 weeks, and 4 fractions over 1 week, for the purpose of developing a short-course irradiation

method.

All of the fractionation regimens were conrmed to be safe in this study (Kato, 2004).

the basis of these results, a phase II study (third protocol) to examine the efcacy of treat-

ment with 52.8GyE in four fractions over 1 week was conducted, and there were no or only minimal

changes in the liver function after the treatment. The 3- and 5-year local control rates were both

96%, and the 3- and 5-year cumulative crude survival rates were 58% and 35%, respectively, indi-

cating

an excellent fractionation regimen in terms of safety and efcacy (Kato etal., 2005a).

Finally,

a phase I/II study with the fourth protocol (very-short-course irradiation of two fractions

over 2 days) was conducted, and no severe side effects were observed (Kato etal., 2005b). This

fractionation

regimen has now been adopted as a clinical practice.

When

the side effects after performing carbon-ion radiotherapy were analyzed, there were few

limitations due to liver dysfunction or tumor diameter, and more than 90% of the patients were almost

free of symptoms during and after the treatment. Patients having cancers with no contact between the

tumor and the gastrointestinal tract, moderate or higher liver function, and a tumor diameter of 10cm

or less are considered to be good candidates for undergoing carbon-ion radiotherapy.

24.2.2.5 prostate

Cancer

After

the early dose escalation study was terminated, 66GyE in 20 fractions over 5 weeks was used

as the standard irradiation regimen in prostate cancer. For a further decrease in the incidence of

adverse reactions, the dose was reduced to 63GyE. However, we have also been conducting treat-

ment with 57.6GyE in 16 fractions over 4 weeks in parallel with the standard regimen in order to

reduce the treatment period, and we recently employed this fractionation regimen as the standard

method.

Patients with prostate cancer were classied into two groups, high-risk and low-risk groups,

based on factors before treatment, including PSA, Gleason score, and TNM classication, to deter-

mine the need for the combination with the endocrine treatment for the initial 10 years. The high-

risk group was treated with carbon-ion radiotherapy in combination with the endocrine therapy,

652 Charged Particle and Photon Interactions with Matter

and the low-risk group was treated with carbon-ion radiotherapy alone (Akakura etal., 2004; Tsuji

etal., 2005; Ishikawa etal., 2006). Patients considered to not require long-term hormone therapy

were separated from the high-risk group as a medium-risk group, and the period of their hormone

therapy

was reduced to 6 months for the last couple of years.

The

incidence of adverse reactions of Grade 3 or more in the rectum or lower urinary tract (blad-

der/urethra) in all patients was 1.5%, but no reactions of Grade 3 or more were observed in any

patients after the optimal irradiation dose was established. The incidence of adverse reactions of

Grade 2 with the current technique was as low as 1.5% in the rectum and 5.7% in the lower urinary

tract, and it has recently tended to decrease after the total dose was reduced to 63GyE. Treatment

with 57.6GyE in 16 fractions has produced no reactions of Grade 2 during observation for 6 months

or more, and this regimen is considered to be safer than that of 63GyE in 20 fractions. According to

the recent survival analysis in these patients, the 5-year survival rate was 91.6%, the cause-specic

survival rate was 98.5%, the 5-year local control rate was 99.1%, and the biochemical nonrecurrence

rate was 88.5%. A comparison of the biochemical nonrecurrence rate in patients with a PSA level of

20ng/mL or higher before treatment and in those receiving other radiotherapies showed a remark-

ably higher nonrecurrence rate in those receiving carbon-ion radiotherapy. The high nonrecurrence

rate should also be associated with the effects of our sound use of combination hormone therapy, but

a comparison with clinical studies in Europe and North America combining hormone therapy and

x-ray radiotherapy showed that the survival rate in our results was 10%–15% higher, conrming that

the

high local effect of carbon-ion radiotherapy led to good treatment results.

Treatment

with 57.6GyE in 16 fractions, the present standard regimen for prostate cancer at

NIRS, is expected to produce results better than those produced with an irradiation of 20 fractions

over 5 weeks in terms of the antitumor effect with a lower risk of adverse reactions, and shows great

promise

for future long-term results.

24.2.2.6

bone

and s

oft

issue

umors

In bone and soft tissue tumors, a phase I/II dose escalation study was followed by a phase II xed-

dose

study, and it is now being performed as a clinical practice.

The

total dose started at 52.8GyE in 16 fractions over 4 weeks, and was increased to 73.6GyE

in the initial study. The local control rate improved as the dose increased, but some patients in the

group who were given the largest dose developed severe skin and soft tissue reactions (Kamada

etal.,

2002).

the phase II xed-dose study, the 3- and 5-year local control rates were both 84% and the 3- and

5-year survival rates were 68% and 49%, respectively. Skin and soft tissue toxicities, as severe side

effects, developed at an incidence of about 3%, but recently, as a result of decreased skin doses, almost

no side effects have been observed. In patients with osteosarcomas in the pelvis or spine, whose resection

was difcult, the 5-year survival rate was 25%. In patients with chordomas other than those developing

in the skull base, the 5-year local control rate was 96% and the 5-year survival rate was 81%. Chordomas

of the sacral bone were reported in Clinical Cancer Research, 2004, and appeared in the Year Book of

Oncology, 2006 (Imai etal., 2006; Leoehrer etal., 2006).

Bone and soft tissue tumors are considered one of the best indications for carbon-ion radio-

therapy. Although long-term observation should be continued, carbon-ion radiotherapy may replace

surgical resection in elderly patients and in patients whose function would be greatly reduced with

resection,

as well as provide a treatment for patients for whom resection is not indicated.

than 4500 patients have already received carbon-ion radiotherapy at NIRS. The results

have been revolutionary for cancer treatment. It has proven effective for cancers in the head and

neck, lung, liver, prostate, sarcomas, etc. Many of the diseases targeted in these clinical trials were

considered unlikely to be effectively treatable with other therapies. Although it is difcult to carry

out direct comparative studies with other therapeutics, it has become increasingly evident that

carbon-ion radiotherapy is capable of curing cancers that are incurable by other treatments, as well

in a shorter time and more safely than when treated by other modalities.

Cancer Therapy with Heavy-Ion Beams 653

24.2.3 analySiS of tuMor control probability

On the basis of dose escalation studies, the clinical results at HIMAC were analyzed in order to

reduce the number of fractionated irradiations. The clinical dose distributions of therapeutic

carbon-ion beam, currently used at NIRS HIMAC, are based on in vitro human salivary gland

(HSG) cell survival response and clinical experience from fast-neutron radiotherapy. The moder-

ate radiosensitivity of HSG cells is expected to be a typical response of tumors to carbon beams.

At rst, the biological-dose distribution is designed so as to cause a at biological effect on HSG

cells in the spread-out Bragg peak (SOBP) region. Then, the entire biological-dose distribution is

evenly raised in order to attain an RBE (relative biological effectiveness) = 3.0 at a depth where

dose-averaged LET (linear energy transfer) is 80keV/μm. At this point, biological experiments have

shown that carbon-ions can be expected to have a biological effect identical to fast neutrons, which

showed a clinical RBE of 3.0 for fast-neutron radiotherapy at NIRS.

The resulting clinical dose distribution in this approximation is not dependent on the dose level,

the tumor type, or the fractionation scheme, and thus reduces the unknown parameters in the analy-

sis of the clinical results. The width of SOBP and the clinical/physical dose at the center of SOBP

specify

the dose distribution.

The

clinical results of NSCLC treated by HIMAC beams were analyzed. They depicted a very

conspicuous dose dependency of the local control rate. A dose escalation study was performed with

a treatment schedule of 18 fractions in 6 weeks (Miyamoto etal., 2003). The dose dependency of the

tumor control probability (TCP) with the photon beam was given by the following formula proposed

Webb:

TCP = −

−













⋅ − − + +

( )

∑

0 693

πσ

α α

α α β

( )

exp exp /( )

N n d d

−−

( )

































(24.1)

α and β are coefcients of the linear quadratic (LQ) model of the cell survival curve. The cell

survival rate, S(D

), is dened as a function of the absorbed dose (physical dose), D

S D D D( ) exp( )

p p p

= ⋅ − ⋅−α β

(24.2)

In the analysis, α and β values of HSG cells were used. σ is the standard deviation of the coefcient

α, which reects the patient-to-patient variation of radiosensitivity. N is the number of clonogens

in tumor (a xed value of 10

was used). n and d are the total fraction number and the fractionated

dose, respectively. T (42 days), T

(0 days), and T

(7 days) are the overall time for treatment, the

kick-off time, and the average doubling time of tumor cells, respectively. Values used in the analysis

are shown in brackets. The analysis was carried out for 18 fractionations in order to determine the

tumor-specic radiosensitivity parameter, α, and its variation, σ. Once the parameters are xed, it

is possible to estimate the dose response curve in various fraction schedules. Figure 24.3 shows an

example of the estimation of the dose response curve of NSCLC. TCP curves of 1, 4, and 9 fractions

are estimated from the parameters derived from the data of 18 fractions. This approach would be

useful

in estimating appropriate prescribed doses when initiating hypofractionated radiotherapy.

24.3 radiation Quality oF Carbon-ion

When incident heavy ions travel in a patient’s body, some of them suffer fragmentation reactions

with target nuclei, and various species of fragments are generated and become widely distrib-

uted. The biological effectiveness of heavy ions is affected not only by the deposited energy, but

also by the particle species (Blakely etal., 1984). In heavy-ion therapy, therefore, precise calcula-

tions of the spatial distribution of radiation quality such as dose, dose-averaged LET, uence, and

energy distributions for each species of particles play a highly important role. In passive beam

654 Charged Particle and Photon Interactions with Matter

delivery, lateral distribution of the radiation quality is, at the rst approximation, regarded as

constant due to the equilibrium between incoming and outgoing particles in the region of interest

in the irradiation eld. Figures 24.4 and 24.5 show an axial uence and the LET distribution of

the therapeutic carbon beam measured at HIMAC.

In order to estimate the distribution of radiation quality in the irradiation eld in detail, or in

order to shape the irradiation eld by a scanning method with pencil beams, it is strongly required

0.2

0.4

0.6

0.8

0 20 40 60 80 100

HIMAC carbon against NSCLC

1 Fx

4 Fx

9 Fx

18 Fx

TCP

Clinical dose (Gy)

Figure 24.3 TCP curve of NSCLC by carbon-ion radiotherapy. TCP curves of 1, 4, and 9 fractionations

are

calculated by clinical data derived from 18 fractionations.

0.2

0.4

0.6

0.8

0 50 100 150

C-290 (MeV/n) SOBP (60 mm) fluence distribution

Z =1

Z =2

Z =3

Z =4

Z =5

Z =6

Normalized fluence

Depth (mm)

Figure 24.4 Particle composition of therapeutic 290MeV/n of carbon beam (60mm SOBP) as a function

thickness in water.

Cancer Therapy with Heavy-Ion Beams 655

to understand the spatial distribution of the radiation quality. The deection of primary particles

in a thick medium has been well described by Moliere’s multiple scattering theory (Molière, 1948;

Wong etal., 1990). On the other hand, multiple scattering alone is not sufcient to account for

the distribution of fragment particles. Our recent study revealed that the large deection of frag-

ment particles in a substance could be accounted for in the multiple scattering formula by con-

sidering one additional term representing a lateral “kick” at the production point of the fragment

(Matsufuji etal., 2005). This additional term can be explained as a transfer of the intra-nucleus

Fermi momentum of a projectile to the fragment, and its extent obeys the expectation derived

from the Goldhaber model (Goldhaber, 1974). The angular distribution of fragment particles was

measured with a monoenergetic 290 MeV/n

C beam through a nuclear reaction in a thick water

target (Matsufuji etal., 2005). Matsufuji et al. determined a parameter describing the extent of the

transferred momentum in the Goldhaber model so as to reproduce the observed angular distribu-

tions. On the basis of these studies, a semi-analytical beam transportation code was developed for

energetic heavy-ion beams, in which the 3D distribution of radiation quality can be calculated for

each species of particles (Inaniwa etal., 2007b). The productions of secondary and tertiary frag-

ments are considered, and the effects of Fermi momentum transfer are taken into account at their

production point. Despite its simplicity, the developed code could reproduce the experimental

result well. Weexpect that it can be used in treatment planning to evaluate the biological effective-

ness of heavy-ion beams precisely.

Especially for 3D scanning with a pencil beam, it is essential to estimate the biological-dose

distribution given by one pencil beam. The biological-dose distribution, D

(r), is expressed as the

product

of the physical-dose distribution, D

(r), and RBE(r):

D D

b p

( ) ( ) ( )r r r= × RBE (24.3)

The RBE is obtained using measured α and β values in the LQ model. On the other hand, the

(r) is directly measured. Figure 24.6 shows the integrated depth-dose distribution measured

by a large-area parallel-plate ionization chamber. It is noted that the Bragg peak of the pristine

beam (monoenergetic carbon-ions with 350MeV/n) is slightly broadened by the ridge lter for

0.5

1.5

500

1000

1500

2000

2500

0 50 100 150

C-290 (MeV/n) SOBP (60 mm)

dose and LET distribution

Dose

Relative absorbed dose

LET [MeV/g/cm

]

Depth (mm)

Dose avgd.

Track avgd.

Figure 24.5 Axial LET distribution of therapeutic 290 MeV/n of carbon beam (60mm SOBP) in water.

656 Charged Particle and Photon Interactions with Matter

3Dscanning. The α and β values are also shown in Figure 24.6, as a function of the depth in water.

Figure 24.7 shows the lateral size of the dose prole in one standard deviation under various range-

shifter thicknesses as a function of the depth in water. With increasing the depth, the lateral size

of the dose prole is gradually increased due to multiple scattering. On the other hand, the size is

rapidly increased beyond the primary beam range, which is caused by fragment nuclei.

0 100 200 300

Relative dose

Depth in water (mm)

(Interp.)

(Meas.)

Figure 24.6 Measured physical-dose distribution of a pencil beam, α and β values in the LQ model, as a

function

of the depth in water.

0 50 100 150 200 250

[mm]

Depth in water (mm)

RSF=0 mm

RSF=30 mm

RSF=60 mm

Figure 24.7 Lateral size of physical-dose prole as a function of the depth in water for various ranges.

Circles, squares, and triangles represent the measured size with range shifter plates of 0, 30, and 60mm thick-

ness,

respectively.

Cancer Therapy with Heavy-Ion Beams 657

Another approach to derive the spatial distribution of radiation quality in matter is the use of

Monte Carlo codes. Although Monte Carlo codes are time consuming and, currently, not practi-

cal for implementation into a treatment-planning system of heavy-ion therapy, some articles have

reported that they can provide precise estimates of radiation quality in matter (Kameoka etal.,

2008;

Nose etal., 2009).

24.4 development oF aCCelerator and beam-delivery systems

24.4.1 heavy-ion therapy facility

The existing carbon-ion radiotherapy facilities are HIMAC, the Hyogo Ion Beam Medical

Center (HIBMC), the Institute of Modern Physics in China (IMP), and the Gesellschaft fuer

Schwerionenforschung (GSI). At HIBMC, treatments were carried out using both protons and car-

bon-ions, and 454 patients were treated with carbon-ion beams. At IMP, using a carbon-ion beam

with 100MeV/n, 103 patients with supercially placed tumors have been treated since 2006. Further,

six patients with deeply seated tumors have been successfully treated with a 400 MeV/n carbon-ion

beam in March 2009. GSI has successfully carried out cancer treatments for around 400 patients

with

carbon-ion beams, applying the intensity controlled raster-scanning method, since 1997.

a typical carbon-ion radiotherapy facility, the HIMAC facility is introduced here. The design

parameters of HIMAC are based on the radiological requirements. Although the ion species ranged

from He to Ar in the design stage, those from p to Xe can be delivered at present. The beam energy

was designed to vary from 100 to 800MeV/n for efcient treatment. HIMAC consists of an injector

linac cascade, dual synchrotron rings with independent vertical and horizontal beam lines, and three

treatment rooms equipped with the beam-delivery system. For carbon-ion radiotherapy, a C

beam

as produced by the 10GHz electron cyclotron resonance (ECR) ion source is injected into the linac

cascade consisting of RFQ and Alvarez linacs, and is accelerated up to 6MeV/n. After the carbon

beam is fully stripped with a carbon-foil stripper, the beam is injected into the synchrotron rings by

the multiturn injection scheme, and is slowly extracted after acceleration to a desired energy. Finally,

the carbon beam extracted from the synchrotron is delivered to the beam-delivery system. Figure 24.8

shows a bird’s eye view of the HIMAC facility.

Synchrotron

(upper)

Synchrotron (lower)

Medium

energy

beam

Accelerator

control room

Alvarez

linac

Physics exp.

room

Secondary

beam exp.

room

Treatment

room C

Treatment

room B

Therapy control room

Horizontal beam

Treatment

room A

Biology exp.

room

Vertical

beam

RFQ linac

Ion sources

Figure 24.8 Bird’s eye view of the HIMAC facility for heavy-ion cancer therapy. (Adapted from Hirao, Y.E.

et al., Nucl. Phys. A., 538, 541C, 1992.)

658 Charged Particle and Photon Interactions with Matter

The HIMAC facility employs a passive beam-delivery method (Torikoshi et al., 2007), as

shown in Figure 24.9: single beam-wobbling for forming the lateral dose distributions and ridge-

lter methods for forming the SOBP as a distal dose distribution, respectively. The pencil beam

delivered from the accelerator is wobbled by a pair of dipole magnets placed in tandem with

their eld directions orthogonal to one another. The magnets are sinusoidally excited with an

alternating electric current of the same frequency of 56.4 Hz, but with a 90° phase shift between

them. Controlling the amplitudes of the magnetic elds, the pencil beam moves on a circular orbit

around the original beam axis. At the same time, the pencil beam is broadened with a scatterer,

placed just downstream from the wobbler magnets. With an appropriate combination of the ampli-

tudes of wobbling and thickness of the scatterer, a uniform dose distribution could be obtained at

the isocenter.

A bar-ridge lter has been used at HIMAC in order to spread out the Bragg peak, so as to match

its size with the target thickness. Passing each monoenergetic ion through a different thickness

of the bar-ridges, the ion looses its energy in proportion to the thickness under the high incident

energy. The cross-sectional shape of the bar-ridge is designed to deliver a biological dose to the tar-

get region homogeneously. In the case of a heavy-ion beam, the SOBP is composed of various LET

components with different weighting factors at each depth. The survival rate under a mixing LET

radiation eld can be described by a formalism proposed in the theory of dual radiation action on

the basis of the LQ model (Kanai etal., 1997), and was experimentally proven. We nally designed

the SOBP of mixed ions with different LET according to the procedure proposed (Chu etal., 1993)

and the formalism (Kanai etal., 1997).

Using the beam-wobbling and ridge-lter methods, the maximum lateral eld and the SOBP size

are designed to be 22cm in diameter with ±2.5% uniformity at the isocenter and 15cm, respectively.

A multi-leaf collimator (MLC) and a patient collimator are used in order to match the lateral dose

distribution precisely with a lateral target shape. On the other hand, a bolus collimator is used for

precisely matching the SOBP with a distal target shape.

At HIMAC, beam-delivery methods were improved in order to increase the irradiation accuracy.

These are (1) the respiratory-gated irradiation method and (2) the layer-stacking irradiation method.

24.4.1.1 respiratory-g

ated i

rradiation

ethod

Damage to normal tissues around a tumor was inevitable in the treatment of the tumor moving along

with the patient’s respiration. A respiratory-gated irradiation method, therefore, which can respond

quickly to irregular respiration, was strongly required. In this method, the irradiation-gate signal

is generated only when the target is located at the design position and the synchrotron can extract

Ridge filter

Multi-leaf

collimator

Target

Field

Patient

collimator

Wobbler magnets

Scatterer

Range shifter

Ionization

chamber

Patient

compensator

Figure 24.9 Schematic diagram of the HIMAC beam-delivery system.

Cancer Therapy with Heavy-Ion Beams 659

a beam. The beam is delivered according to the gate signal. Thus, one of the key technologies for

the respiratory-gated irradiation is the beam extraction method from the synchrotron according to

the gate signal. For this purpose, the radio frequency knockout (RF-KO) slow extraction method,

which utilizes a transverse beam heating by an RF eld tuned with a wave number of a horizontal

betatron oscillation, was developed (Noda etal., 1996). Advantages of the RF-KO extraction are a

quick response within sub-milliseconds to a signal of an extraction start/stop and a low horizontal

emittance as compared with that in the ordinary slow extraction method, due to a constant separa-

trix. In accordance with the RF-KO method, the respiratory-gated irradiation system was developed

(Minohara

etal., 2000).

In this system,

the

respiration signal is generated by the observation of the

movement of an LED set on the surface of the patient’s body through a position-sensitive detector,

and a beam can be delivered only with the gate-signal. The respiratory-gated irradiation system has

been

used for treatments of liver and lung tumors since 1996.

24.4.1.2

layer-stacking

rradiation

ethod

By the conventional method, the xed SOBP, which is produced by a ridge lter, results in unde-

sirable dosage to the normal tissue in front of the target, because the width of an actual target

varies within the irradiation eld. In order to suppress this undesirable dosage, the layer-stacking

irradiation method was proposed (Kanai etal., 1983), and the HIMAC irradiation system has been

upgraded to put the technique including the treatment planning (Kanematsu etal., 2002) into prac-

tice. In this method, as shown in Figure 24.10, a small SOBP, several mm in water equivalent length

(WEL), which is produced by a single ridge lter, is longitudinally scanned over the target vol-

ume in a stepwise manner by dynamically controlling the conventional beam-modifying devices.

Concerning the lateral dose distribution, the target volume is longitudinally divided into slices, and

each slice is irradiated by conforming the beam to cover each cross-sectional shape with the MLC.

The performance of this method was experimentally veried (Futami etal., 1999; Kanai etal.,

2006),

and it has been used mainly for the treatment of the head and neck cancer.

24.4.2 developMent of coMpact carbon-ion radiotherapy facility

24.4.2.1 standard-type Carbon-ion radiotherapy Facility in Japan

For the purpose of the widespread use of carbon-ion radiotherapy in Japan, the NIRS decided to

design a standard-type carbon-ion radiotherapy system in order to reduce construction costs by

downsizing the accelerator complex and beam-delivery systems. Based on the design study by NIRS

(Noda etal., 2007), a standard-type facility has been under construction at the Gunma University

Wobbler

magnet

Scatterer

Target

Monitors

Range shifter

MLC

Compensator

Ridge filter

Figure 24.10 Schematic diagram of the layer-stacking irradiation method. (Adapted from Kanai, T. et al.,

Med. Phys.,

33, 2989, 2006.)