Radiat Environ Biophys (2010) 49:469–475 DOI 10.1007/s00411-010-0285-1
ORIGINAL PAPER
Measurement of microdosimetric spectra produced from a 290 MeV/n Spread Out Bragg Peak carbon beam Satoru Endo • Masashi Takada • Hiroki Tanaka • Yoshihiko Onizuka • Kenichi Tanaka • Nobuyuki Miyahara • Hiromi Baba • Ayumu Oishi • Masayori Ishikawa • Masaharu Hoshi • Shinzo Kimura • Masakazu Minematsu Yuki Morimune • Yasuaki Kojima • Kiyoshi Shizuma
•
Received: 7 September 2009 / Accepted: 30 March 2010 / Published online: 18 April 2010 Ó Springer-Verlag 2010
Abstract This study describes measurements on secondary particles produced by a 290 MeV/n Spread Out Bragg Peak (SOBP) carbon beam. Microdosimetric distributions of secondary fragments from the SOBP carbon
S. Endo (&) M. Minematsu Y. Morimune Y. Kojima K. Shizuma Quantum Energy Applications, Graduate School of Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima 739-8527, Japan e-mail:
[email protected] M. Takada N. Miyahara National Institute of Radiological Science, 4-9-1 Anakawa, Inage-ku, Chiba 263-8555, Japan H. Tanaka Research Reactor Institute, Kyoto University, Kumatori-cho, Sennan-gun, Osaka 590-0494, Japan Y. Onizuka H. Baba A. Oishi Department of Health Science, Kyushu University, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan K. Tanaka Center of Medical Education, Sapporo Medical University, 17, Minami 1 Jo, Chuo-ku, Sapporo 060-8556, Japan M. Ishikawa Department of Medical Physics and Engineering, Graduate School of Medicine, Hokkaido University, N-15 W-7, Kita-ku, Sapporo 060-8638, Japan M. Hoshi Research Institute for Radiation Biology and Medicine, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8553, Japan S. Kimura National Institute of Occupational Safety and Health, 6-21-1 Nagao, Tama-Ku, Kawasaki 214-8585, Japan
beam have been measured by using a new tissue equivalent proportional counter (TEPC) system at the Heavy Ion Medical Accelerator in Chiba of the National Institute of Radiological Sciences. The new TEPC system consists of a TEPC, two solid-state detectors (SSD) and a scintillation counter (FSC: forward scintillation counter). The SSDs and FSC can separately identify charged fragments and secondary neutrons produced by the incident carbon ions. Microdosimetric distributions were measured for secondary particles including neutrons produced by a body-simulated phantom consisting of various PMMA plates (thickness: 0, 34.81, 55.2, 60.95, 64.83, 95.03, 114.79, 124.69, 135.2 and 144.98 mm, respectively) to cover the SOBP (at 60–125 mm depth). The new system can separately determine produced fragments from the incident SOBP carbon beam in a body-simulated phantom.
Introduction At the heavy ion medical accelerator (HIMAC) of the National Institute of Radiological Sciences (NIRS) in Chiba, more than 4,000 patients with cancers in different organs have been treated until 2008 by means of carbon ions with an energy of 290, 350 and 400 MeV/n, respectively. Carbon ion radiotherapy, due to its physical and biological advantages over photons (Scholz 2000; Kraft 2001; Tu et al. 2002; Karger and Ja¨kel 2007), has provided an improved treatment in terms of high local control rates for various tumors and pathologically non-squamous cell type tumors (Tsujii et al. 2004, 2007). We have studied the radiation quality of the carbon beam used for these treatments and of the produced secondary particles by means of microdosimetric techniques using a tissue equivalent gas proportional counter (TEPC)
123
470
coupled with scintillation counters (Endo et al. 2005, 2007a, b). In these previous experiments, the radiation qualities of primary and secondary particles produced by a mono-energetic carbon beam, i.e., 290 and 400 Mev/n, have already been measured (Endo et al. 2005, 2007a, b). However, the actual therapeutic beam is not mono-energetic; rather, it is a moderated beam formed with a ridge filter, a wobbler magnet and a scatterer. In order to study the beam quality of a therapeutic carbon beam as actually used in practice, a Spread Out Bragg Peak (SOBP) beam of 290 MeV/n carbon ions was used in the present study. In addition, a time of flight (TOF) instrument has been adapted for particle identification in a previous experimental setup (Endo et al. 2005, 2007a, b). The TOF technique requires a few meters flight length, so the system becomes relatively large. We improved the system, which consists now of TEPC and solid-state detectors (SSDs), to establish a compact system. In the present paper, the new TEPC system (TEPC coupled with SSD) is introduced, and the microdosimetric spectra measured for a 290 MeV/n SOBP carbon beam and for the secondary particles produced are reported and discussed.
Materials and methods Microdosimetric lineal-energy distributions for the 290 MeV/n SOBP carbon beam measured using a TEPC (Far East Technology Inc. LET counter 1/2’’ model) which consists of a 0.18-mm-thick aluminum cap, a 3.7-mm-thick A150 plastic wall and an active cavity with a diameter of 12.7 mm. A phantom (Kanai et al. 1998) made of polymethylmethacrylate (PMMA) plates (density: q = 1.19 g/cm3; elemental composition: C5H8O2; thickness of phantom controlled by combining nine plates of different H2Oequivalent thickness of 0.6, 1.1, 2.3, 4.8, 9.3, 18.5, 37.1, 74.3 and 149.0 mm, respectively) was used to simulate the human body. The microdosimetric distributions of the secondary particles were measured using body-simulated PMMA of 0, 34.81, 55.2, 60.95, 64.83, 95.03, 114.79, 124.69, 135.2 and 144.98 mm H2O-equivalent thickness, respectively, to cover the 6 cm SOBP peak of 290 MeV/n carbon. The diameter of the incident carbon beam was set to 20 cm at the PMMA position using a wobbler, a scatterer and a ridge filter at the so-called Bio-course of HIMAC, which can produce a SOBP beam similar to that used for therapeutic irradiation. A schematic diagram of the experimental setup is shown in Fig. 1. A pickup scintillation counter (PSC; thickness: 0.1 cm; cross-section: 25 9 25 cm2) was set behind the PMMA phantom. The so-called forward scintillation counter (FSC; thickness: 0.3 cm; cross-section: 1.5 9 1.5 cm2) and two solid-state
123
Radiat Environ Biophys (2010) 49:469–475
detectors (SSD1 and SSD2; thickness: 500 lm; diameter: 7 mm; both totally depleted silicon charged-particle detectors, PIPS, Canberra Co. Ltd.) were positioned behind the phantom but front of the TEPC to identify chargedparticle fragments. The PSC is a monitor used to count the total number of incident carbon ions. The PSC counts are used for normalization. The FSC will be used in the future for neutron identification using a veto technique. Note, however, that neutral events are not discussed in the present paper, and the FSC was not used for the present study. The TEPC was filled with tissue equivalent gas (propane-based TE-gas) at a pressure of 4.4 kPa which corresponds to a spherical tissue volume with a diameter of 1 lm site size. The TEPC was operated at 600 V. Energy calibration was carried out using a 244Cm alpha-source present in the TEPC wall, and the energy resolution for the 244 Cm alpha particles was 8.5% (one standard deviation). To cover a wide lineal-energy region from 0.2 to 1,000 keV/lm, pulse height distributions of ‘‘low’’, ‘‘medium’’ and ‘‘high’’ gains from the TEPC were acquired by using three amplifier modules (ORTEC 671). The triggers were made through a timing-single-channel-analyzer (ORTEC Timing-SCA) module using the signals of the ‘‘high gain’’ amplifier. The pulse heights of the TEPC, FSC and SSDs were acquired by an analogue-to-digital converter (ORTEC, ADC 811) using a CAMAC controller (TOYO cc/7700).
Results and discussion Identification for secondary particles Charged particles produced by the SOBP carbon beam are identified analyzing the pulse height of the DE signals of the SSD1 and SSD2. Examples of the particle identification for 0, 60.95 and 124.69 mm depths are shown in Fig. 2a, c. The color in these density scatter plots indicates the population of each event. Six clusters are clearly identified in the figure, which correspond to proton (H), helium (He), lithium (Li), beryllium (Be), boron (B) and carbon (C) particles, respectively. Particle separation is estimated by fitting a Gaussian distribution for neighbor particles. Note that for measurements using PMMA phantoms thinner than 124.69 mm, the tail of each particle spectrum contains less than 10% of the neighbor particle. The worst particle separation is about 30%, when large He tails for a PMMA thickness of 144.98 mm was misidentified as Li. The particle separation is evaluated from the particleidentified SSD spectrum. An example of the spectrum for 60.95 mm depth is shown in Fig. 3. The same identification was made for all depths, and the resulting relative fluence for each particle is shown in Fig. 4. The figure
Radiat Environ Biophys (2010) 49:469–475
471
scatterer
PSC
SSD1,SSD2
ridge filter
PMMA
SOBP-C Beam
FSC
wobbler
Fig. 1 Sketch of the experimental setup used to measure the microdosimetric spectra of the 290 MeV/n SOBP carbon beam. The carbon beam was incident to the body-simulated PMMA phantom. Spectra were measured using a tissue equivalent proportional counter
TEPC
(TEPC). A forward scintillation counter (FSC) and two solid-state detectors (SSD1, SSD2) were used to identify particles incident to the TEPC. A pickup scintillation counter (PS) measured the number of carbon ions hitting the PMMA phantom
Fig. 2 Density scatter plot for the two SSDs, for a PMMA = 0 mm, b PMMA = 60.95 mm and c PMMA = 124.69 mm, respectively. Six clusters are clearly identified. Red corresponds to high statistics while purple corresponds to low statistics
104
C He C
Li
Relative fluence
H
103
Li He H 102
B Be
Be
B
101 0
50
100
150
BF thickness (mm) Fig. 3 Example of SSD pulse height distributions of particles identified at 60.95 mm depth
demonstrates that carbon particles dominate up to a PMMA phantom thickness of 100 mm. For a PMMA thickness greater than 100 mm, the number of secondary proton and helium particles exceed those of the carbon particles. The other secondary particles are less than 5% of the secondary proton and helium particles. These particle-identified events are used for determination of the microdosimetric spectra.
Fig. 4 Relative fluence for each particle identified. Arrow shows the range of 290 MeV/n carbon
Microdosimetric spectra for secondary particles The event of each identified fragment was scored into logbin of the lineal energy, y as fraw(y). Applying the basic microdosimetry technique, y f(y) and y d(y) R1 R 1 should be by 0 fraw ð yÞdy ¼ 1 and 0 yf ð yÞ dy ¼ Rnormalized 1 1983). The microdosimetric 0 d ð yÞ dy ¼ 1 (ICRU
123
472
Radiat Environ Biophys (2010) 49:469–475
spectrum was calculated separately, for each particle. After this, in order to estimate the total spectrum f(y) and y d(y)for all the particles, a particle- and dose fraction of each particle was calculated from yF and the dose values given in Tables 1 and 2 (see the next subsection), and multiplied and combined with the spectrum (Endo et al. 2005, 2007a, b) The obtained spectrum y d(y) for incident carbon ions for a PMMA thickness of 0, 34.81, 55.2, 60.95, 64.83, 95.03, 114.79, 124.69 and 144.98 mm, respectively, is shown in Fig. 5. The y d(y) peak of the carbon beam is located around about 20 keV/lm. The peak positions of carbon for less than 64.8 mm show a similar value of 20 keV/lm. The value is lower than that of mono-energetic carbon which is shifted depending on depth (Endo et al. 2007a). Events with a lineal energy larger than 100 keV/lm were caused by Bragg peak events. In the SOBP, low lineal-energy events around a few 10 keV/lm are due to the less degraded high energy carbon beams. This suggests that a small number of carbon Bragg peak events contributes to this spectrum. The dose of secondary particles, excluding those of the primary carbon beam, is mainly associated with helium fragments. Carbon may break up to 4 He and 8Be fragments, while a 8Be particle decays to two 4 He particles with a short half-life of 6.7 9 10-17 s. The performance of particle separation in this new system is similar to that of the previous one (Endo et al. 2005, 2007a, b). However, two advantages of the new system can be listed: (1) the system size is reduced to 30 cm, whereas the previous size was 3 m; (2) particle identification is carried out by the particles incident to the TEPC. The first advantage offers the possibility to use the system in a small room, such as a space station. The second advantage is that the new system does not require any correction for escaped events, which was needed in the previous system. This is because the TOF method used in the older system uses energy deposition information of the
Table 1 Estimated yF for carbon and secondary fragments Thickness (mm)
H
He
Li
0
0.98
1.80
3.51
4.19
34.81
0.76
1.90
3.65
6.51
10.9
18.1
55.2
0.73
2.04
4.01
7.32
12.7
21.9
60.95
0.74
1.95
3.92
8.11
12.6
23.4
64.83
0.73
1.92
3.71
7.24
12.9
23.7
95.03
0.81
2.33
5.68
11.3
17.6
34.0
114.79
0.84
2.47
5.95
11.8
21.6
56.6
124.69
0.95
2.60
6.27
12.2
16.8
31.0
135.2 144.98
0.92 0.96
3.11 3.27
8.60 8.20
16.1 15.1
28.9 31.5
– –
123
B 9.35
Effective thickness (mm)a
Dose (nGy/carbon) H
He
Li
Be
B
C
Total
7.7
0.03
0.07
0.05
0.11
0.22
14.6
15.1
42.51
0.07
0.21
0.11
0.11
0.32
13.8
14.7
62.9
0.10
0.31
0.15
0.16
0.44
14.7
15.9
68.65
0.12
0.34
0.15
0.15
0.52
15.0
16.2
72.53
0.11
0.30
0.14
0.12
0.42
16.5
17.6
102.73
0.21
0.60
0.46
0.25
0.78
12.2
14.5
122.49 132.39
0.30 0.35
0.69 0.78
0.56 0.88
0.51 0.46
1.00 0.36
7.55 1.12
142.9
0.33
1.24
0.86
0.34
0.60
–
3.36
152.68
0.36
1.32
0.81
0.31
0.26
–
3.06
10.6 3.95
a
Wall and other material thicknesses of the detectors (7.7 mm) were taken into account
particles passing through the system, whereas the new system allows identification of incident particles. Relative dose distribution There is a relation between the absorbed dose per incident particle R (D) and the frequency of the mean lineal energy yF yF ¼ yf ðyÞ dy as D ¼ kyF n d 2 ð1Þ where k = 0.204 is a constant related to the diameter of the counter (Onizuka et al. 2002; Rossi and Zaider 1996), n is the number of trigger events corrected by a ratio of SSD to PSC counts at PMMA thickness of 0 mm and d is the actual detector diameter in the unit of lm (Onizuka et al. 2002; Rossi and Zaider 1996). From Eq. 2, the dose for each particle per incident carbon was estimated using the particle fraction /i, X kyiF /i =d 2 ði ¼ H, He, Li, Be, B, CÞ ð2Þ Dtot ¼ i
yF (keV/lm) Be
Table 2 Dose for each particle obtained from Eq. 2
C 15.0
The yF values estimated for carbon and fragments are summarized in Table 1, and the dose of each particle per incident carbon estimated and the fragment dose fractions are listed in Table 2. The results show that the yF does not change much for phantom thickness until near monoenergetic 290 MeV/n carbon range (125 mm). The dose profile deduced is shown in Fig. 6a. The total dose of all the particles obtained by the TEPC measurement is quite similar to the measured value of the ionization chamber which was used to measure the beam condition before this study. The fraction of the produced fragments contributing to total dose (for a thickness of less than 120 mm) is 3–20% increasing with depth. The dose
Radiat Environ Biophys (2010) 49:469–475
473
(a)
0mm
(b) 55.2mm
µ
(c)
µ 64.83mm
(d)
95.03mm
µ
(e)
µ
114.79mm
µ
(f)
144.98mm
e
µ
Fig. 5 Obtained y d(y) distribution. Figures are for a PMMA thickness of a 0 mm, b 55.2 mm, c 64.83 mm d 95.03 mm, e 114.79 mm and f 144.98 mm, respectively
average lineal energy yD for each particle (i = H, He, Li, Be, B, C) can be calculated from each microdosimetric spectrum. The yD values for all particles as calculated using
the dose-weighted mean over all particles (Table 2) are listed in Table 3. The yD values for all particles shown in Fig. 6b are compared with those measured previously by
123
474
Radiat Environ Biophys (2010) 49:469–475
(a) Dose (nGy/carbon)
25
H He Li Be B C total Chamber
20 15 10 5 0
yD or averaged LET (keV/µm)
(b) 250 Present CR39 (JJAP 2005) HIBRAC (JJAP 2005)
200 150
Uncertainty of the microdosimetric results was estimated to be 10% for yF and 6% for yD (Waker 1985). On the other hand, the uncertainty due to particle separation is of similar or smaller magnitude, except for Li and He beyond the mono-energetic 290 MeV/n carbon range of 125 mm, where the uncertainty is about 30%, due to misidentification between Li and He. Uncertainty of the dose values obtained with the ionization chamber is a few percentage. Compared with this, the dose determined by the TEPC system has a relatively large uncertainty. However, because the TEPC system can separate not only dose, but also the secondary particle spectra from the incident carbon beam, the present technique is considered to be quite useful.
100
Conclusion 50 0
0
40
80
120
160
Effective thickness or H2O thickness of (mm)
Fig. 6 a Dose profile as a function of the effective thickness of the PMMA; b Average lineal energy as a function of the effective thickness of the PMMA, and averaged LET dependence of the H2O thickness obtained using HBRAC, CR39 (Kohno et al. 2005)
Table 3 Estimated yD values for carbon and secondary fragments; ‘‘total’’ shows the averaged lineal energy for all particles Thickness (mm)a
yD (keV/lm) p
He
Li
Be
B
7.7
6.37
4.03
10.5
11.1
15.4
18.5
18.3
42.51
3.41
4.06
10.6
12.2
15.9
27.2
26.3
62.9
3.32
4.13
8.48
14.7
21.4
37.1
35.3
68.65
2.92
5.41
7.91
21.7
22.0
46.7
44.2
72.53
4.64
3.90
7.51
17.6
20.9
46.2
44.1
102.73
3.92
4.39
11.6
17.5
26.3
72.4
63.3
122.49
2.34
4.71
10.7
23.6
33.4
138
132.39
2.51
5.76
25.6
30.6
123
44.1
142.9 152.68
2.26 2.43
7.29 5.83
35.0 30.5
55.3 48.0
– –
21.5 14.2
9.39 20.3 15.7
C
Total
103.4
a
Wall and other material thicknesses of the detectors (7.7 mm) were taken into account
means of CR-39 foils and calculated with HIBRAC (Kohno et al. 2005). Figure 6b demonstrates that the present results agree well with those from Kohno et al. (2005). The yD values become larger near the 290 MeV/n carbon range (125 mm), which shows that the radiation quality depends on depth. In addition, the figure demonstrates that the new system is working well, for both detection and evaluation of the radiation field.
123
A microdosimetric study of secondary fragments produced by the 290 MeV/n SOBP carbon beam from HIMAC has been carried out using a system including a scintillation counter, solid-state detectors and a TEPC. Lineal energy distributions for the charged fragments were obtained in the lineal energy range from 0.2 to 1,000 keV/lm on a body-simulated phantom. It was found that the present TEPC system allows separation of secondary fragments of B, Be, Li, He and H, and incident carbon. By means of this particle identification system, microdosimetric spectra for secondary particles and carbon were analyzed in terms of microdosimetry. The dose fraction of the produced fragments (for a thickness less than 120 mm) increased with depth from 3 to 20%. The yD values deduced became larger near the 290 MeV/n carbon range, which demonstrates the radiation quality to depend on depth. It was shown that the radiation dose and quality of fragments for a SOBP carbon beam can be measured and evaluated by this TEPC system indeed. The advantage of this system which allows evaluation of both radiation dose and radiation quality was demonstrated. In the future, this system will be used to measure microdosimetric spectra for neutrons as well. Acknowledgments This experiment was accepted by HIMAC as collaborative study P210. The authors express their sincere thanks to the staff members of HIMAC for stable operation of the accelerator and management of the experiments. This study was supported in part by Grant-In-Aid of Scientific Research No. 18510046 and 21310038 from the Ministry of Education, Science, Sports and Culture of Japan.
References Endo S, Onizuka Y, Takada M, Tanaka K, Ishikawa M, Yamaguchi H, Hayabuchi N, Shizuma K, Hoshi M (2005) Microdosimetric evaluation for 400 MeV/nucleon carbon beam at HIMAC. Med Phys 32:3843–3848
Radiat Environ Biophys (2010) 49:469–475 Endo S, Takada M, Onizuka Y, Tanaka K, Maeda N, Ishikawa M, Miyahara N, Hayabuchi N, Shizuma K, Hoshi M (2007a) Microdosimetric evaluation of secondary particles in a phantom produced by carbon 290 MeV/nucleon ions at HIMAC. J Radiat Res 48:397–406 Endo S, Tanaka K, Takada M, Onizuka Y, Miyahara N, Sato T, Ishikawa M, Maeda N, Hayabuchi N, Shizuma K, Hoshi M (2007b) Microdosimetric evaluation of secondary neutrons in a phantom produced by a 290 MeV/nucleon carbon beam at HIMAC. Med Phys 34:3571–3578 ICRU (International Commission on Radiation Units and Measurements) (1983) Microdosimetry, ICRU Report 36 (Bethesda, MD: ICRU) Kanai T, Tomura H, Matsufuji N, Futami Y, Higashi A, Miyahara N, Itoh-Koyama H, Endo M, Kawachi K (1998) Physical studies on the irradiation system at HIMAC. NIRS-M (Natl Inst Radiol Sci) 125:338–343 Karger CP, Ja¨kel O (2007) Current status and new developments in ion therapy. Strahlenther Onkol 183:295–300 Kohno R, Yasuda N, Himukai T, Kase Y, Ochiai K, Komori M, Matsufuji N, Kanai T (2005) Measurements of dose-averaged linear energy transfer distributions in water using CR-39 plastic nuclear track detector for therapeutic carbon ion beams. Jpn J Appl Phys 44:8722–8726 Kraft G (2001) What we can learn from heavy ion therapy for radioprotection in space? Phys Med XVII(Suppl 1):13–20
475 Onizuka Y, Endo S, Ishikawa M, Hoshi M, Takada M, Kobayashi T, Sakurai Y, Utsumi H, Uehara S, Hayabuchi N, Maeda N, Takatsuji T, Fujikawa K (2002) Microdosimetry of epithermal neutron field at the Kyoto University Reactor. Radiat Prot Dos 99:383–385 Rossi HH, Zaider M (1996) Microdosimetry and its applications. Springer, Berlin Heidelberg New York Scholz M (2000) Heavy ion tumor therapy. Nucl Instrum Method Phys Res B 161–163:76–82 Tsujii H, Mizoe J, Kamada T, Baba M, Kato S, Kato H, Tsuji H, Yamada S, Yasuda S, Ohno T, Yanagi T, Hasegawa A, Sugawara T, Ezawa H, Kandatsu S, Yoshikawa K, Kishimoto R, Miyamoto T (2004) Overview of clinical experiences on carbon ion radiotherapy at NIRS. Radiother Oncol 73(Suppl):S41–S49 Tsujii H, Mizoe J, Kamada T, Baba M, Tsuji H, Kato H, Kato S, Yamada S, Yasuda S, Ohno T, Hasegawa A, Nakajima M, Sugae N, Tamaki N, Takagi R, Kanedatsu S, Yashikawa K (2007) Clinical results of carbon ion radiotherapy at NIRS. J Radiat Res 48(Suppl):A1–A13 Tu Zh, Kobayashi Y, Kiguchi K, Watanabe H (2002) Applicability of radiosurgery with heavy ion beams to inactivate specific organs in living organisms. Radiat Environ Biophys 41:231–234 Waker AJ (1985) Experimental uncertainties in microdosimetric measurements and an examination of the performance of three commercially produced proportional counters. Nucl Instrum Method Phys Res A234:354–360
123 View publication stats
