1
Moderate-energy Carbon Ions for Intra-Operative Radiation Therapy: A Feasibility
2
Study
3
M. Seimetz,1, a) P. Bellido,1 P. Conde,1 A.J. González,1 A. Iborra,1 L. Moliner,1
4
J.P. Rigla,1 M.J. Rodrı́guez-Álvarez,1 F. Sánchez,1 A. Soriano,1 and J.M. Benlloch1
5
Instituto de Instrumentación para Imagen Molecular (I3M),
6
Universidad Politécnica de Valencia, Valencia, Spain
7
(Dated: 23 May 2014)
1
Purpose: At present, the primary medical use of carbon ion beams is the precise
treatment of deep-lying tumours. Existing, large-scale therapeutic facilities are optimised for the application of high-energy particles. The superficial irradiation of
surgically resected tumour beds with beams of carbon ions at moderate energies
might provide a cost-effective possibility to make use of their advantageous characteristics for a much larger number of pathologies. We sketch the outline of a compact
device for the acceleration and application of these particles and study its technical
feasibility.
Methods: The key component of the compact therapy device is a carbon ion
source, based on laser-plasma interaction, with a maximum energy of 480 MeV
(40 MeV/u). Its feasibility is assessed in a review of published data. While the
energy and spectral distribution of ions accelerated by laser are often considered inadequate for the treatment of deep-lying tumours the physical requirements for the
proposed application are less stringent. Based on realistic ion spectra various aspects of a superficial irradiation are investigated, like the depth-dose profile and the
production of secondary isotopes, as well as practical details of the therapy system.
Results: Carbon ions in the required energy range, an order of magnitude below
current external beam therapy facilities, have already been demonstrated in laserplasma interactions. Further experiments are required to achieve similar results at
reduced laser power. GATE simulations show that continuous carbon ion spectra
in the range 200-480 MeV provide an interesting depth-dose profile for a radiation
boost after a surgical intervention. The absolute dose can be locally as high as 50 Gy
for realistic, single-pulse particle numbers, making a complete treatment with about
100 laser shots feasible. Prompt gamma emitting isotopes are produced in sufficient
abundance to allow for online-monitoring of the administrated radiation dose.
Conclusions: The proposed, compact treatment system takes advantage of the
physical characteristics of laser-accelerated carbon ions. It may represent a promising alternative to existing Intra-Operative Radiation Therapy with photons and electrons. Since carbon ions of the required energies have been obtained in previous
experiments we conclude that an irradiation device may be realised in the near future.
2
PACS numbers: 87.56.-v, 87.55.-x, 52.38.Kd
8
a)
Electronic mail: [email protected]
3
9
I.
INTRODUCTION
10
The emerging technique of accelerating protons and heavier ions by highly intense laser
11
pulses has attracted considerable interest for its potential to provide compact particle
12
sources.1 These are especially interesting for medical applications where the size and cost of
13
classical accelerators hinder the prevalence of hadrons for cancer therapy. Radiation treat-
14
ment with positively charged particles is currently limited to very few, specialised centres,
15
despite their very advantageous characteristics with respect to photons and electrons. Cur-
16
rent therapy concepts for deep-lying tumours require ion energies which are an order of
17
magnitude above the maximum achieved in laser-plasma interactions. Therapeutic applica-
18
tions of laser-driven particle sources are therefore often considered far beyond the current
19
technical possibilities.
20
However, the requirement on high ion energies is justified mainly by the range of particles
21
inside the patient. A superficial treatment may well be conducted with low-energy ions with
22
broad energy spread. We propose a new therapeutic modality which combines the versatility
23
of Intra-Operative Radiation Therapy (IORT) with the demonstrated advantages of carbon
24
ions as compared to photon and electron radiation. In order to motivate the usefulness of the
25
new technique we briefly review IORT and current proton and ion therapy in section II. The
26
novel concept of Intra-Operative Ion Therapy (IOIT) will be presented in section III. In order
27
to address its feasibility we review in detail the status of carbon ion acceleration by laser-
28
plasma interactions (section IV A). As we will show in section IV B the energy distribution
29
and particle numbers reported in recent experiments are already close to the necessities of
30
the proposed application. We therefore hope to motivate further investigation on this novel
31
concept which may be beneficial for the treatment of a wide range of pathologies.
32
II.
RADIATION THERAPY CONCEPTS
33
A.
Intra-Operative Radiation Therapy
34
The irradiation of a tumour bed immediately after resection is a widely used therapeutic
35
strategy to erradicate remaining cancerous cells. Several methods exist to apply a well-
36
controlled, elevated radiation dose inside the operation room. Compact, mobile accelerators
37
provide electron beams of several MeV or secondary photons in the 50 keV range. Another
4
38
popular option is the (temporary) implantation of radioactive sources by automated after-
39
loading devices. In IORT, a large radiation boost (typically 10-20 Gy) is applied immediately
40
after surgery, with a small number of follow-up treatments. This implies not only a higher
41
cost-effectiveness, but also an important increase in comfort for the patient as compared
42
to standard radiation therapy distributed over several weeks. With the radiation applied
43
directly to the affected organs IORT allows for reduced dose as compared to External Beam
44
Radiation Therapy (EBRT). This effect is multiplied by a factor 2-3 increase in biological
45
effectiveness of a large single fraction as compared to conventional fractionation.2 A draw-
46
back of the high initial dose, especially with electrons, is a unique profile of late toxicity.
47
The intrinsic blood vessels and connective tissue of organs suffer long-term damage, and the
48
irradiation of close-by nerves also increases the risk of late side effects.3 Therefore an even
49
more localised radiation treatment is desirable, as offered by carbon ions with very limited
50
range. These promise, in addition, a further reduction of the overall dose thanks to their
51
increased relative biological effectiveness (see section II B).
52
The clinical relevance of IORT has been proven for many kinds of cancer, also in di-
53
rect comparison with standard external beam radiation.3 It has been studied in depth for
54
breast cancer within the TARGIT-A4 and ELIOT trials5 , with soft X-rays and electrons,
55
respectively. For this pathology IORT reduces the risk of local tumour recurrence.
56
For the long-term prospect of the patient the importance of complete resection of can-
57
cerous tissues has often been stressed, also in combination with standard IORT. However,
58
a more efficient IORT may prove to be especially useful where completely clear margins
59
cannot be achieved, e.g. in the vicinity of critical organs, or even with unresectable, small
60
tumours close to the surface. In these cases carbon ions with very limited range may more
61
effectively destroy all remaining cancer cells without harming other tissues.
62
B.
Proton and Carbon Ion Therapy
63
Positively charged ions have been used in External Beam Radiation Therapy (EBRT) for
64
several decades. As compared to electrons and X-rays they present several advantageous
65
properties. The most prominent one is the deposition of their bulk energy in the so-called
66
Bragg peak at the end of their trajectory while the dose in tissues on top of the target volume
67
is minimized. Behind the target (at known depth) no primary ions are present. However, in
5
68
this region the dose may be non-vanishing due to projectile fragmentation (in case of Z > 1
69
beams) or the production of other secondary radiation. These effects are important for ion
70
energies of several hundred MeV/u which are necessary to reach penetration depths around
71
20 cm inside the human body. Due to their high mass the ions propagate practically without
72
lateral scattering.
73
Another outstanding feature of positive ions is their capacity to kill malignant cells. For
74
carbon ions it is about 3-4 times higher than the one of soft X-rays at the same radiation
75
dose while for protons this so-called relative biological effectiveness (RBE) is only slightly
76
increased.6 This does not only imply that the total dose can be reduced to achieve the same
77
effect as the one of photon or electron radiation therapy. A further, important advantage
78
resides in the capacity of carbon ions to efficiently destroy two major kinds of cancer cells
79
which show increased resistance to photons and protons. These are hypoxic tumour cells
80
and cancer stem cells. Hypoxia is a well-known complication of many pathologies. Cancer
81
cells in sparsely oxigenated regions are more likely to survive irradiation with X-rays and
82
are held responsible for local tumour recurrence and the generation of distant metastasis
83
through migration. Carbon ion irradiation has proven to be less dependent on the oxy-
84
gen concentration, neither at the time of treatment nor during post-irradiation recovery, as
85
compared to photon or proton therapy.7 Cancer stem cells (CSC) are less active than dif-
86
ferentiated tumour cells, but after a treatment (by radiation or chemotherapy) can give rise
87
to various kinds of cancer cells and thereby cause tumour recurrence and systemic spread.8
88
For some types of CSCs the superiority of carbon ion therapy as compared to X-rays has
89
been demonstrated.9,10 Further positive effects have been reported such as the suppression
90
of angiogenesis.7 These findings underline the necessity to augment the availability of car-
91
bon ion therapy and to extend its use beyond deep-lying lesions in order to improve the
92
long-term prospect of cancer therapy.
93
Existing proton and ion therapy centres are designed to take maximum advantage of the
94
physical peculiarities of these particles. They typically aim at deep-lying, non-resectable
95
tumours for which they are clearly superior to external photon or electron beams. Ion
96
energies are precisely adjusted to reach a given depth. Extended tumours are scanned
97
layerwise with very thin (“pencil”) beams. The direction of the incoming ion beam is varied
98
over time in order to minimize the radiation dose along the entrance path. The technological
99
effort of this kind of treatment is huge. Modern proton and ion therapy centres, such as
6
100
the Heidelberg Ion-Beam Therapy Centre (HIT, Germany), are equipped with fully rotating
101
gantries (25 m diameter, 600 metric tons), in addition to the accelerator sections necessary
102
to provide a 400 MeV/u, monoenergetic carbon ion beam. Such facilities, with building
103
costs exceeding 100 million dollars, are affordable only for the most developed countries. At
104
present only four carbon ion therapy centres are operative worldwide, two in Japan (HIMAC
105
at Chiba and HIBMC at Hyogo), CNAO at Pavia (Italy), and HIT (after a previous pilot
106
project at GSI, Darmstadt). Approximately 10000 patients have been treated with external
107
carbon ion beams to date. High-energy proton therapy is offered at more than 20 centres
108
in the world with more than 70000 patients. The treatment capacity of these facilities is
109
limited to approximately 1000 patients per year.
110
Proton and ion acceleration by ultra-intense lasers has been widely discussed as possi-
111
ble means to reduce the size and cost of therapeutic facilities.11 This emerging technique,
112
detailed in section IV, in principle allows for replacing the large radiofrequency accelerator
113
structures and most of the electromagnetic beam control elements of classical accelerators by
114
much more compact, optical components, as was suggested already in 2002 by Bulanov and
115
Khoroshkov12 . It requires highly intense, femtosecond laser pulses which, when focused onto
116
suitable targets, liberate large numbers of charged particles that are concentrated around
117
a known direction. Despite considerable progress throughout the last decade it has not
118
yet been possible to demonstrate proton or ion acceleration to energies as those applied
119
in EBRT, even at the largest, petawatt scale laser facilities. Medical applications of laser-
120
accelerated ions therefore are often considered to be still far in the future. Further, most
121
experiments observed broadly spread particle spectra which are incompatible with the re-
122
quirements of treatment plans based on pencil beams. Therefore some electromagnetic beam
123
control elements behind the laser target are generally considered indispensable, e.g. in the
124
proposed design of the ELIMED proton therapy facility.13 If a laser-driven carbon beam
125
is to be applied to similar, deep-lying tumours as in current therapy centres, the particle
126
energy has to be equally high. This, in turn, implies a huge gantry system since the limiting
127
factor is the bending power of the magnets. The novel concept at ten times lower carbon ion
128
energies (up to 40 MeV/u) which we present in section III may constitute an opportunity for
129
exploiting the possibilities of laser-ion acceleration on a shorter time scale, offering some of
130
the most important advantages of carbon ion therapy to patients at virtually every hospital
131
worldwide. It combines the compactness of Intra-Operative Radiation Therapy with the
7
132
enhanced cell-killing power of carbon ions.
133
III.
INTRA-OPERATIVE ION THERAPY: A NOVEL CONCEPT
134
What does it take to provide a carbon ion beam for intra-operative radiation treatment?
135
The ion range should be adapted to the superficial irradiation of recently resected tumour
136
beds. To reach a depth of 5 mm, say, inside water-equivalent tissue, carbon ions need some
137
40 MeV/u or, in other words, a total kinetic energy of 480 MeV (Figure 1). Note that
138
this is a factor 10 less than what is used at external-beam carbon therapy centres. The
139
same particles have a range of 5 m in air. Thus they may leave the vacuum system of
140
the accelerator and gantry through a thin applicator window some centimeters above the
141
patient. For the proposed, superficial treatment the carbon ion beam does not have to
142
be monoenergetic. Each ion will be completely absorbed at the depth corresponding to its
143
energy. If we suppose that possible, remnant tumour cells after resection are most likely close
144
to the operated tissue (at the margins of the tumour bed) this dose profile will effectively
145
attack them.
Carbon ion range in air
Range [m]
Range [mm]
Carbon ion range in water
5
5
4
4
3
3
2
2
1
1
0
0
5
10
15
20
25
30
0
0
35
40
45
Carbon energy [MeV/u]
5
10
15
20
25
30
35
40
45
Carbon energy [MeV/u]
FIG. 1. Range of carbon ions in water (left) and in air as function of kinetic energy (plotted with
the SRIM code14 ).
146
We thus propose a radiation system as sketched in Figure 2. It is not as small as photon
147
or electron IORT devices, but much more compact than existing proton or ion treatment
148
facilities. The overall layout combines the basic outline of laser-acceleration experiments, as
149
described in section IV, with some practical concerns. A fundamental part of the accelerator
150
section is the laser source (1) with its corresponding optical beam line (2). Ti:sapphire or
8
FIG. 2. General layout of a compact carbon ion irradiation system (components: see text). Inset:
Zoom on optical and beam selecting parts.
151
Nd:glass lasers are most commonly used. Laser pulses are focused (3) on a carbon-rich
152
target (4) within a focal spot of some micrometers diameter. Carbon ions are released and
153
accelerated in a direction close to the target normal (5). A collimator (6) selects the central
154
part of the beam where ions reach the highest energies. Carbon ions pass a stripper foil (7)
155
to remove remaining electrons and enrich the C6+ charge state. A pair of dipole magnets
156
(8) is applied to eliminate electrons and to define a minimum accepted energy through
157
a second collimator (9). All these components are housed inside a vacuum system (10)
158
which accelerated ions can leave through a thin kapton window at the end of an applicator
159
section (11). The final beam spot diameter and shape may be adjusted in a third collimator
160
inside the applicator. As typically a relatively large area (some cm2 ) will be irradiated
161
it is advantageous to use a beam with a few mm width and scan the entire surface. A
162
very narrow pencil beam, which is standard at high-energy facilities, is not required here.
163
Realistically, only this final applicator should be housed inside the operation room. The
9
164
laser and most of the radiation shielding (12) may be mounted in a separate room with
165
independent cooling system and better accesibility for technical maintenance. However,
166
the applicator must provide some flexibility to adjust the beam hit position and angle, an
167
operation which requires reverse interaction on several up-stream components, including the
168
laser target and focusing system and the magnetic momentum selector. A gamma detector
169
(13) allows for monitoring the applied radiation dose by measuring the activity of isotopes
170
produced by projectile fragmentation and other nuclear reactions inside the patient.
171
Before addressing some key aspects of the feasibility of such a therapy system in section
172
IV B we will present an overview on the status of the underlying technique, the acceleration
173
of ions by highly intense laser pulses. Note that our proposal is not limited to carbon
174
ions; other ion species heavier than protons may be accelerated and applied in the same
175
way. However, since fractionation increases at higher Z carbon ions are considered an ideal
176
tradeoff between elevated biological effectiveness and negative side-effects.
177
IV.
178
A.
ION ACCELERATION BY ULTRA-INTENSE LASER PULSES
Results from ion acceleration experiments
179
During the last decade proton and ion acceleration by high-intensity lasers has been
180
demonstrated at several laboratories and the experimental results have been extensively
181
reviewed1 . At the same time the theoretical understanding of the underlying acceleration
182
mechanisms in different regimes has progressed15 . The experimental setup for laser-plasma
183
acceleration generally comprises a high-power laser (starting from a few TW) providing fem-
184
tosecond pulses which are focused on a suitable target, and detectors for the characterisation
185
of the accelerated particles in terms of energy spectrum, particle numbers, and angular dis-
186
tribution. Typical target materials are thin metal or mylar foils with thicknesses between
187
some nm and about 20 µm. In this case the most efficient acceleration is achieved in forward
188
direction, i.e. ions are detected at the rear side of the target, opposite to the laser incidence.
189
The majority of studies are centred on the acceleration of protons. Here, focused laser
190
intensities between 1018 and 1021 W/cm2 have been applied at various facilities, with single
191
pulse energies between 1 mJ and more than 300 J. The most important findings for laser-
192
accelerated protons may be summarized as follows. Particles are emitted from a narrow spot
10
193
around the laser hit position, forming a beam along (or close to) the target normal direction.
194
The proton energies, E, are widely spread, and the particle numbers per energy interval can
195
be parametrised, e.g. with an exponential decay inspired by a Boltzmann distribution16 ,
N0 −E/kB T
∆N
=
e
,
∆E
E
(1)
196
up to a sharp cutoff, Emax . This maximum energy depends on several experimental para-
197
meters, like the laser pulse energy and focused intensity (denoted Wp and I0 , respectively),
198
the pulse contrast, and the target material and thickness. Nevertheless, general trends have
200
been observed by comparing a major number of experimental data1 : Emax is approximately
q
√
proportional to Wp , and Emax ∝ I0 for metallic foil targets, while Emax ∝ I0 for dielectric
201
target materials. In total, a few percent of the laser pulse energy may be converted into
202
kinetic energy of accelerated protons. These findings have been explained in the concept
203
of Target Normal Sheath Acceleration (TNSA). In this model, the wavefront of the high-
204
intensity laser pulse creates a plasma at the target front surface. In the subsequent laser
205
fields (with several tens up to hundreds of femtoseconds duration) the plasma charges are
206
separated and electrons are accelerated through the target foil. When leaving the target rear
207
side these electrons create an electrostatic field with field strengths of the order of TV/m.
208
Protons from the target back surface follow this field forming a secondary beam. On their
209
way through the target the electrons are spread out. Thus, the surface area from which
210
protons are pulled out usually is much bigger than the focal spot of the laser.
199
211
Experimental results on the acceleration of “heavy” (Z > 1) ions currently are much less
212
abundant than those for protons. The reasons for this are threefold: The mass-normalized
213
ion energies (in MeV/u) are lower because the transfer of electrostatic potential into kinetic
214
energy is less efficient for ions with reduced charge-over-mass ratio (q/mu = 1/2 in the best
215
case); the detection and identification of ions, especially at low energies, is more demanding
216
than for protons; and in the presence of protons the acceleration of heavier ions is suppressed.
217
The last point implies the use of high-purity targets. For the acceleration of carbon ions
218
(the most widely studied nuclei) mainly two target materials have proven to be useful:
219
pure carbon in special configurations or metal foils with hydrocarbon impurities where the
220
hydrogen content is eliminated by resistive heating.
221
Table I provides an overview on published experimental data for Z > 1 ions for which, to
222
the best of our knowledge, a similarly detailed review has not been published before. Similar
11
223
to proton acceleration, many collaborations have made use of thin metallic targets17–23 .
224
Willingale and coworkers18 applied 6 µm thick Al foils coated with 1 µm of deuterated
225
polystyrene. They observed accelerated deuterons with similar energies when coating either
226
the front or the back side of the target. Hegelich et al.20 coated aluminium and tungsten
227
foils with carbon and CaF2 , respectively, and demonstrated the highest C and F ion energies
228
after elimination of hydrogen impurities by resistive heating. With heated palladium foils a
229
monoenergetic band in the carbon ion spectra has been obtained19 . Ions of several different
230
elements have been accelerated in a similar way21,23 . Kar et al.22 claim the observation of
232
narrow-band features in carbon ion spectra at focused intensities up to 3 × 1020 W/cm2 . A
√
√
comparison of these data (Figure 3) reveals roughly a WL and I0 scaling of the maximum
233
ion energy per nucleon, similar to what is known from protons in the TNSA regime.
231
234
Several groups have made use of pure carbon targets, especially diamond-like carbon
235
(DLC) foils which are typically only some tens of nm thin25,28–30 . Here, about 5 times higher
236
ion energies have been obtained as compared to metal foils (Fig. 3). They are attributed
237
to more efficient acceleration mechanisms like Radiation Pressure Acceleration (RPA) and
238
Break-Out Afterburner (BOA) due to lasers with ultra-high intensity and contrast. Al-
239
though the data are still not too abundant, the general trend indicates indeed a Emax ∝ I0
240
scaling predicted for these regimes. The relatively low ion energies achieved by Carroll and
241
coworkers27 , both with pure carbon and Al targets, do not correspond to the general trend
242
at the very high focused intensity of 7 × 1020 W/cm2 which they claim. In terms of focused
243
pulse energy (5.8 J on target) they are, however, compatible with other experiments. With
244
circularly polarized pulses a peaked structure in the C6+ ion spectra has been found25 .
245
With gas jet31 and cluster-gas targets24,26 high-energetic ions of different species have
246
been reported, partly at moderate laser pulse power (≤1 J) and focused intensities (<
247
1018 W/cm2 ). It is not clear how the maximum ion energy scales with the laser parameters.
248
Apart from the maximum ion energy, another important parameter for possible hadron
249
therapy applications is the number of accelerated ions, especially at the high-energy end of
250
the spectra. Many authors present the number of particles, normalised to ions/(MeV/u)/sr,
251
but collected in a detector with limited aperture, such as a Thomson parabola spectrome-
252
ter with some nanosteradian acceptance angle located along the target normal direction.
253
For a concise extraction of total particle numbers the angular distribution and the useful
254
divergence of the ion beam should be taken into account. However, it is fair to assume a
12
TABLE I. Experimental data of laser-accelerated ions with Z > 1.
Reference
Pulse
Laser
energy
power
Irradiance
Target
Ion
Emax
WL [J]
PL [TW]
I0 [W/cm2 ]
material
species
[MeV/u]
Fujii17
0.088
20
6.8 × 1018
Cu
C
0.03
Fukuda24
0.15
4
7 × 1017
He-CO2 gas
C, O, He
20
Henig25
0.7
30
5 × 1019
DLC
C
6
Fukuda26
1
1 PW
7 × 1018
He-CO2 gas
C, O, He
50
Carroll27
5.8
115
7 × 1020
C foil
C
5
Al foil
C
4
2.5 × 1019
Al + polymere
D
0.17
Willingale18
6
Hegelich19
20
30
1 × 1019
Pd+CH2
C
3
Hegelich20
30
100
5 × 1019
Al+C, W+CaF2
C, F
5
Hegelich21
30
100
5 × 1019
metal+CH2 /CaF2
C, O, F, Be
5.5
Henig28
80
100
7 × 1019
DLC
C
15
Jung29
80
1 PW
5 × 1020
diamond
C
54
Hegelich30
90
150
2 × 1020
DLC, CH2
C
44
Kar22
200
250
3 × 1020
Cu, Al
C
11
Willingale31
340
1 PW
5.5 × 1020
He gas
He
3
McKenna23
400
1 PW
2 × 1020
Fe
Fe
12
255
uniform energy distribution up to 1◦ opening angle (corresponding to ∼ 1 msr). Thus, we
256
will use this angular acceptance for the following estimations. We will also present the data
257
normalised to an energy interval of 1 MeV/u and therefore scale them with the nuclear mass
258
number, NM , whenever the original authors have given their results in ions/MeV.
259
Kar et al.22 report 108 ions/(MeV/u)/msr in the energy range 3-7 MeV/u for Z/A = 1/2
Carroll et al.27 found about 108 C6+ ions/(MeV/u)/msr for particles above
260
(carbon).
261
4 MeV/u. Henig and coworkers28 observed, in the best case, some 107 C6+ ions/(MeV/u)/msr
262
around Emax = 15 MeV/u. This is much more than the total particle number of some 105
263
integrated for energies above 200 MeV quoted by Jung et al.29 . Hegelich et al.30 detected
264
some 106 ions/(MeV/u)/msr for the highest energy bins (above 40 MeV/u). Similar results
13
EI [MeV/u]
Maximum ion energy
10
Metal
DLC/Dielectric
Gas/Cluster
1
10-1
1
102
10
(a)
W L [J]
EI [MeV/u]
Maximum ion energy
10
Metal
DLC/Dielectric
Gas/Cluster
1
1
102
10
(b)
I0 [1018 W/cm2 ]
FIG. 3. Experimental data for maximum ion energies as a function of laser pulse energy and
focused intensity. The dotted trend lines correspond to A × (WL /J)0.5 , A = 0.75 MeV (red) and
A = 4.0 MeV (blue) for (a), and 0.75 × (I0 /1018 W/cm2 )0.5 (red) and 0.15 × (I0 /1018 W/cm2 )
(blue) for (b), respectively.
265
have been reported at lower cutoff energies21,25 . With cluster-gas targets26 the observed
266
particle numbers are much lower. Single-particle pits in CR-39 detectors are counted and a
267
total of the order of 104 − 105 for 50 MeV/u ions is estimated taking into account the low
268
detection efficiency of the somewhat indirect experimental method. However, the elemental
269
composition (possibly a mixture of He, C, and O ions) remains unspecified. In contrast,
14
270
Willingale and coworkers31 claim 109 He2+ ions/(MeV/u)/msr at 5-10 MeV/u along the
271
target normal direction.
272
Finally, it is interesting to compare the production of different charge states of the same
273
isotope. In general, higher charge states are accelerated to higher energies. Many authors
274
thus restrict a detailed presentation of their results to the highest ionization states and, to
275
the most, show some example spectra for the rest. In Carroll et al.27 the maximum ion
276
energy seems to be roughly proportional to the charge number when comparing C1+ , C2+ ,
277
C4+ , and C6+ ions. In contrast, for He1+ and He2+ ions from a gas target31 the maximum
278
energy ratio seems to be rather 1:4. Some indications exist that not necessarily all possible
279
charge states are populated and that full ionization of the target material may not always
280
be achieved.19,20 In summary, the increased maximum energy of higher charge states fits to
281
an acceleration in an electrostatic field. However, the relatively scarce experimental findings
282
require further confirmation and deeper understanding in order to be properly taken into
283
account for possible practical applications.
284
B.
Implications for carbon ion therapy
285
Is an intra-operative carbon ion therapy device, as depicted in section III, feasible in
286
the near future? In 2007, Linz and Alonso11 critically reviewed the application of laser
287
accelerated protons for radiation therapy, with a very skeptical outcome for most of the
288
aspects like particle energy, beam opening angle, and even cost. However, they considered
289
the needs of current facilities for the treatment of deep-lying tumours (section II B), whereas
290
the requirements for a superficial carbon ion treatment are far less stringent.
291
The most critical parameters are the ion energy and particle numbers. Maximum energies
292
above the required 40 MeV/u have already been demonstrated, yet not without drawbacks.
293
When solid carbon targets were applied29,30 the laser pulses were produced at petawatt
294
facilities with 80-90 J single pulse energy which may not easily be shrinked to tabletop size
295
and run by hospital staff. With a gas-cluster target similar maximum energies have been
296
observed with only 1 J per pulse26 , however providing a yet uncharacterised mixture of
297
different ion species (He, O, C) which is not obviously suitable for medical purposes. These
298
findings should be confirmed in independent experiments. In any case, with underdense
299
target materials such as carbon foams32 the conversion of laser energy may become much
15
Particles / MeV / msr
Generic carbon ion distribution for single laser shot
6
10
5
10
0
50
100
150
200
250
300
350
400
(a)
450
500
Ep / MeV
Dose [Gy]
Dose deposition in water
70
C ions, >100 MeV
60
C ions, >200 MeV
electrons, 3 MeV
50
photons, 50 keV
40
30
20
10
0
0
2
4
6
8
(b)
10
12
14
16
Depth [mm]
FIG. 4. (a) Generic energy distributions of laser-accelerated carbon ions, both with a total of 108
ions, but different low-energy cutoff. (b) Dose profile as function of depth in water for 108 carbon
ions, compared to monoenergetic electrons and photons.
300
more efficient than in the TNSA regime and the acceleration of carbon ions up to 40 MeV/u
301
with 1-10 J pulses does not seem to be impossible.
302
The absolute particle numbers around 106 /(MeV/u)/msr for energies around 40 MeV/u
303
look rather low. However, as a large part of the spectrum is useful for superficial irradiation
304
and the numbers increase by 1-2 orders of magnitude towards lower energies the total may
305
be 1000 times higher, making 108 useful ions per shot over the full spectral range a realistic
16
306
estimate. As an example one can consider a generic spectrum following eq. (1), with the
307
energy range and particle numbers similar to the experimental results by Hegelich et al.30 ,
308
however with a low-energy cutoff at 100 MeV (Figure 4(a)). We have applied the GATE V6
309
code33 to simulate the dose deposition of 108 such carbon ions in a uniform beam with 10 mm
310
diameter when absorbed in water (Figure 4(b)). The maximum radiation dose at a depth of
311
0.4 mm is as high as 65 Gy, decreasing rapidly for deeper-lying tissue, dropping to 10 Gy at
312
1.5 mm depth. With a higher cutoff (200 MeV), but the same total number of accelerated
313
ions, the peak position is shifted to 1.1 mm and the dose at larger depths is significantly
314
higher up to the maximum range of the carbon ions of 5 mm. The relatively large dose
315
values are due to the very small target volume. The irradiation of the entire operated bed
316
(some tens of cm2 ) then requires only ∼100 laser shots within a few minutes treatment time,
317
i.e. a pulse rate of the order 1 Hz. This is still challenging both for high-intensity lasers and
318
precisely positioned targets, but may be a realistic goal within a few years from now.
319
The broad energy spectra of the carbon ions and the presence of (hot and cold) electrons
320
behind the laser target necessitate some magnetic beam control elements, albeit much less
321
than in classical accelerators. Their design is challenged by the mixture of different charge
322
states. Therefore we suggest to homogenize the ionization spectrum by stripping the re-
323
maining electrons off carbon ions with charge states lower or equal than C5+ . This can be
324
achieved with a carbon stripper foil of a few microns thickness placed directly behind the
325
laser target, similar to injection systems of synchrotrons34 . A pair of dipole magnets with
326
0.1 Tm field integral each can then be applied to deflect and redirect the initial beam. Note
327
that in a single dipole field (such as a 90◦ bending magnet which is typical for therapeutic
328
gantries) the spectral components of the beam would be spatially separated. For C6+ ions
329
at 100-480 MeV the deflection angle is between 3.1◦ and 6.9◦ , sufficient for collimation be-
330
hind the first dipole to separate particles with lower energies. Electrons will be efficiently
331
eliminated in this arrangement, as well as most lighter ions such as protons which should be
332
rare anyway when pure carbon targets are used.
333
A major concern for radiation treatment is the safety of the patient and the operating
334
staff. In the laser-plasma interaction not only carbon ions are released, but also electrons
335
and X-rays. With the laser section housed in a separate room effective shielding can be
336
provided for all kinds of secondary radiation. Direct X-rays from the laser target, emitted
337
in all directions, can be eliminated from the particle beam behind the first chicane magnet.
17
338
In the same dipole, electrons will be deflected opposite to the carbon ions. Thus, a pure
339
carbon beam is provided in the applicator section. With the very limited range of these ions
340
the complete absorption of the halo is not very demanding. At a depth above 5 mm inside
341
the patient, the dose due to fractionation of the carbon ions or other, secondary radiation
342
is below 0.025 Gy, as demonstrated with the GATE simulations mentioned above. Note
343
that in photon or electron IORT a major part of the total dose is deposited outside the
344
target volume, limiting the applicability of the technique in the vicinity of critical organs.
345
To illustrate the very distinct properties of carbon, electron, and photon beams we have
346
calculated dose-depth profiles of monoenergetic, 3 MeV electrons and 50 keV photons as
347
typically used in IORT (Figure 4(b)). Note also that at high carbon ion energies the load
348
of secondary radiation behind the target volume is not negligible. In total, the use of low-
349
energy carbon ions can be considered a safe technique which may be applied in the presence
350
of the operating personnel.
351
C.
Control of applied dose
352
The superficial radiation dose, applied by a single shot of carbon ions as described above,
353
may be very high (locally up to 50 Gy for 108 particles), albeit confined to a very small
354
volume. Thus only a small number of pulses is needed inside the total, irradiated area.
355
This implies that, for dose control purposes, the shot-to-shot stability of ion energies and
356
intensities must be guaranteed up to a few percent. Apart from separate control measure-
357
ments (performed before an intervention) the radiation can be monitored directly. Since
358
the accelerated beam has a relatively large aperture part of the off-axis (halo) ions can be
359
detected and characterised, e.g. directly in front of the last collimator.
360
Some of the incident carbon ions undergo nuclear reactions and produce β + isotopes which
361
511 keV annihilation photons can be detected with Positron Emission Tomography (PET)
362
systems for a spatial reconstruction of the dose deep inside the patient. This technique is
363
well known in External Beam Radiation Therapy35 . In order to verify its feasibility at much
364
lower ion energies we have estimated the yield of β + nuclei as follows. The cross section of
365
the
366
hitting a carbon target.36 In inverse kinematics this corresponds to 25-40 MeV/u carbon ions
367
incident on protons. Taking into account the range and energy loss of the ions in water a
12
C(p, np)11 C reaction is roughly 80 mbarn for a proton beam with Ep = 25-40 MeV
18
368
spectral distribution like the one of Figure 4(a) (with 200 MeV minimum energy) translates
369
into some 30000
370
17.2 Bq which is insufficient for a reliable monitoring.
11
C nuclei produced from 108 incident
12
C ions. Their β + activity is only
371
Further information on the production of β + isotopes can be obtained from the GATE
372
simulations mentioned in the previous section which provide a list of all nuclear frag-
373
ments. For the simulated absorption of carbon ions in water these include the 12 C(p, X) and
374
12
C(16 O, X) reaction channels. Again, for a single shot of 108 carbon ions with 200 MeV
375
minimum energy we find between 4000 and 40000 atoms of different PET isotopes with a β +
376
activity of 118 Bq, the most important contribution coming from 15 O. Even if more activity
377
is accumulated with each pulse a precise dose control by this method is quite demanding. For
378
a superficial therapy a 3D reconstruction is not required and thus it may be more efficient to
379
detect only single, 511 keV annihilation photons inside a predefined energy interval instead
380
of requiring a pair coincidence. However, taking into account the finite (small) aperture of
381
a gamma detector the count rate will still be too low for a precise measurement.
382
Nevertheless, an alternative way of monitoring looks feasible, based on direct gamma de-
383
cays of 12 C reaction products. The GATE simulations indicate, for example, the production
384
of two excited states of
385
the ground state by emission of 1.37 MeV and 2.76 MeV photons. The appearance of this
386
isotope is plausible from the
387
may efficiently be identified with an inorganic crystal (say, 60 mm of BGO) close to the irra-
388
diated area. We therefore propose to include such a detector in the layout of the treatment
389
system.
390
V.
24
Mg, with about 5700 atoms each, which immediately decay to
12
C+16 O→
24
Mg+α fusion channel. These energetic photons
CONCLUSIONS
391
Ion acceleration by high-intensity laser pulses may provide carbon ion beams of sufficient
392
energy and particle numbers for superficial radiation treatment in the near future. Pub-
393
lished experiments from various laser facilities worldwide have already demonstrated the
394
feasibility of this underlying technique. This conclusion does not strictly contradict the of-
395
ten uttered evaluation that laser-accelerated protons and ions are still far from being useful
396
for medical applications. Our more optimistic point of view is justified by the proposal of a
397
completely new treatment modality, Intra-Operative Ion Therapy, with by far less stringent
19
398
requirements on ion energies and spectral dispersion. However, further basic research and
399
optimisation of the experimental conditions are necessary to achieve particle beams with
400
laser pulse energies in the 10 J range and maximum carbon ion energies around 40 MeV/u.
401
Due to the highly nonlinear dependence and the interplay between different parameters it
402
is hard to predict the composition of an optimum setup. Low-density target materials, like
403
carbon foams, seem to be promising components of a compact therapy device.
404
We have evaluated the feasibility of several aspects of the proposed therapeutic modality,
405
starting from carbon ion spectra similar to those of published experiments. After homoge-
406
nization of charge states in a thin stripper foil the interesting, broad ion momentum interval
407
may be selected by a pair of dipole magnets which, in addition, allow for the elimination
408
of other particles such as electrons. In GATE simulations we have calculated the dose de-
409
position which is limited to a maximum depth of 5 mm, contrary to photon and electron
410
radiation in current IORT. Thanks to an elevated single-shot dose only a small number of
411
laser pulses (of the order 100) is necessary for the irradiation of an operated bed. We have
412
critically evaluated the possibilities of monitoring the applied dose and concluded that in our
413
case the detection of prompt γ rays is the most promising alternative while the production
414
of β + (PET) isotopes, preferred in current carbon EBRT, may be insufficient due to reduced
415
ion energies.
416
With the proposed, intra-operative treatment scheme it will be possible to exploit the
417
elevated biological effectiveness of carbon ions for a large number of cancer patients. The aim
418
is to provide a sufficiently compact device to be used at local hospitals, contrary to the huge
419
therapy centers of today. The medical benefits will require numerous studies. Conceptually,
420
IOIT does not intend to fully replace photon and electron IORT; many aspects, including
421
practicability, cost effectiveness, treatment efficiency, and side effects, will have to be assessed
422
for many pathologies on a long-term scale. IOIT does not even compete with external-beam
423
carbon ion therapy at high energies because the treated tumour sites (superficial vs. deep-
424
lying) are completely distinct. IOIT may thus be established as an independent treatment
425
modality and, at the same time, promote research on laser-ion acceleration for medical
426
purposes.
20
427
ACKNOWLEDGMENTS
This work has partially been funded by the Centre for Industrial Technological Develop-
428
429
ment (CDTI), ref. IPT-20111027, through the Science Industry Subprogramme.
430
Conflicts of interest: All authors declare that they have no conflicts of interest.
431
REFERENCES
432
1
their applications,” Reports on Progress in Physics 75, 056401 (2012).
433
434
2
3
F. A. Calvo, R. M. Meirino, and R. Orecchia, “Intraoperative radiation therapy: Part 2.
clinical results,” Critical Reviews in Oncology/Hematology 59, 116 – 127 (2006).
437
438
C. G. Willett, B. G. Czito, and D. S. Tyler, “Intraoperative radiation therapy,” Journal
of Clinical Oncology 25, 971–977 (2007).
435
436
H. Daido, M. Nishiuchi, and A. S. Pirozhkov, “Review of laser-driven ion sources and
4
J. S. Vaidya, F. Wenz, M. Bulsara, J. S. Tobias, D. J. Joseph, M. Keshtgar, H. L. Flyger,
439
S. Massarut, M. Alvarado, C. Saunders, W. Eiermann, M. Metaxas, E. Sperk, M. Stterlin,
440
D. Brown, L. Esserman, M. Roncadin, A. Thompson, J. A. Dewar, H. M. R. Holtveg,
441
S. Pigorsch, M. Falzon, E. Harris, A. Matthews, C. Brew-Graves, I. Potyka, T. Corica,
442
N. R. Williams, and M. Baum, “Risk-adapted targeted intraoperative radiotherapy versus
443
whole-breast radiotherapy for breast cancer: 5-year results for local control and overall
444
survival from the TARGIT-A randomised trial,” The Lancet 383, 603 – 613 (2014).
445
5
U. Veronesi, R. Orecchia, P. Maisonneuve, G. Viale, N. Rotmensz, C. Sangalli, A. Luini,
446
P. Veronesi, V. Galimberti, S. Zurrida, M. C. Leonardi, R. Lazzari, F. Cattani, O. Gentilini,
447
M. Intra, P. Caldarella, and B. Ballardini, “Intraoperative radiotherapy versus external
448
radiotherapy for early breast cancer (ELIOT): a randomised controlled equivalence trial,”
449
The Lancet Oncology 14, 1269 – 1277 (2013).
450
6
Oncol. 7, 37 (2010).
451
452
M. Durante and J. S. Loeffler, “Charged particles in radiation oncology,” Nat. Rev. Clin.
7
N. Hamada, T. Imaoka, S.-i. Masunaga, T. Ogata, R. Okayasu, A. Takahashi, T. A. Kato,
453
Y. Kobayashi, T. Ohnishi, K. Ono, Y. Shimada, and T. Teshima, “Recent advances in the
454
biology of heavy-ion cancer therapy,” Journal of Radiation Research 51, 365–383 (2010).
455
8
F. Pajonk, E. Vlashi, and W. H. McBride, “Radiation resistance of cancer stem cells: The
21
4 R’s of radiobiology revisited,” Stem Cells 28, 639–648 (2010).
456
457
9
X. Cui, K. Oonishi, H. Tsujii, T. Yasuda, Y. Matsumoto, Y. Furusawa, M. Akashi, T. Ka-
458
mada, and R. Okayasu, “Effects of carbon ion beam on putative colon cancer stem cells
459
and its comparison with X-rays,” Cancer Research 71, 3676–3687 (2011).
460
10
K. Oonishi, X. Cui, H. Hirakawa, A. Fujimori, T. Kamijo, S. Yamada, O. Yokosuka, and
461
T. Kamada, “Different effects of carbon ion beams and X-rays on clonogenic survival and
462
DNA repair in human pancreatic cancer stem-like cells,” Radiotherapy and Oncology 105,
463
258 – 265 (2012).
464
11
to tumor therapy?” Phys. Rev. ST Accel. Beams 10, 094801 (2007).
465
466
12
S. Bulanov and V. Khoroshkov, “Feasibility of using laser ion accelerators in proton therapy,” Plasma Physics Reports 28, 453–456 (2002).
467
468
U. Linz and J. Alonso, “What will it take for laser driven proton accelerators to be applied
13
G. A. P. Cirrone, D. Margarone, M. Maggiore, A. Anzalone, M. Borghesi, S. B. Jia, S. S.
469
Bulanov, S. Bulanov, M. Carpinelli, S. Cavallaro, M. Cutroneo, G. Cuttone, M. Favetta,
470
S. Gammino, O. Klimo, L. Manti, G. Korn, G. La Malfa, J. Limpouch, A. Musumarra,
471
I. Petrovic, J. Prokupek, J. Psikal, A. Ristic-Fira, M. Renis, F. P. Romano, F. Romano,
472
G. Schettino, F. Schillaci, V. Scuderi, C. Stancampiano, A. Tramontana, S. Ter-Avetisyan,
473
B. Tomasello, L. Torrisi, S. Tudisco, and A. Velyhan, “ELIMED: a new hadron therapy
474
concept based on laser driven ion beams,” (2013).
475
14
http://www.srim.org.
476
477
15
A. Macchi, M. Borghesi, and M. Passoni, “Ion acceleration by superintense laser-plasma
interaction,” Rev. Mod. Phys. 85, 751–793 (2013).
478
479
J. Ziegler, J. Biersack, and M. Ziegler, “SRIM - the stopping and range of ions in matter,”
16
F. Nürnberg, M. Schollmeier, E. Brambrink, A. Blažević, D. C. Carroll, K. Flippo, D. C.
480
Gautier, M. Geißel, K. Harres, B. M. Hegelich, O. Lundh, K. Markey, P. McKenna,
481
D. Neely, J. Schreiber, and M. Roth, “Radiochromic film imaging spectroscopy of laser-
482
accelerated proton beams,” Review of Scientific Instruments 80, 033301 (2009).
483
17
T. Fujii, Y. Oishi, T. Nayuki, Y. Takizawa, K. Nemoto, T. Kayoiji, K. Horioka, Y. Okano,
484
Y. Hironaka, K. G. Nakamura, and K.-i. Kondo, “MeV-order proton and carbon ion
485
acceleration by irradiation of 60 fs TW laser pulses on thin copper tape,” Applied Physics
486
Letters 83, 1524–1526 (2003).
487
18
L. Willingale, G. M. Petrov, A. Maksimchuk, J. Davis, R. R. Freeman, T. Matsuoka,
22
488
C. D. Murphy, V. M. Ovchinnikov, L. V. Woerkom, and K. Krushelnick, “Front versus
489
rear side light-ion acceleration from high-intensity laser-solid interactions,” Plasma Physics
490
and Controlled Fusion 53, 014011 (2011).
491
19
B. M. Hegelich, B. J. Albright, J. Cobble, K. Flippo, S. Letzring, M. Paffett, H. Ruhl,
492
J. Schreiber, R. K. Schulze,
493
monoenergetic MeV ion beams,” Nature 439, 441 (2006).
494
20
and J. C. Fernández, “Laser acceleration of quasi-
M. Hegelich, S. Karsch, G. Pretzler, D. Habs, K. Witte, W. Guenther, M. Allen, A. Blaze-
495
vic, J. Fuchs, J. C. Gauthier, M. Geissel, P. Audebert, T. Cowan, and M. Roth, “MeV ion
496
jets from short-pulse-laser interaction with thin foils,” Phys. Rev. Lett. 89, 085002 (2002).
497
21
B. M. Hegelich, B. Albright, P. Audebert, A. Blazevic, E. Brambrink, J. Cobble, T. Cowan,
498
J. Fuchs, J. C. Gauthier, C. Gautier, M. Geissel, D. Habs, R. Johnson, S. Karsch, A. Kemp,
499
S. Letzring, M. Roth, U. Schramm, J. Schreiber, K. J. Witte, and J. C. Fernández,
500
“Spectral properties of laser-accelerated mid-Z MeV/u ion beams,” Physics of Plasmas
501
12, 056314 (2005).
502
22
S. Kar, K. F. Kakolee, B. Qiao, A. Macchi, M. Cerchez, D. Doria, M. Geissler, P. McKenna,
503
D. Neely, J. Osterholz, R. Prasad, K. Quinn, B. Ramakrishna, G. Sarri, O. Willi, X. Y.
504
Yuan, M. Zepf, and M. Borghesi, “Ion acceleration in multispecies targets driven by
505
intense laser radiation pressure,” Phys. Rev. Lett. 109, 185006 (2012).
506
23
P. McKenna, K. W. D. Ledingham, J. M. Yang, L. Robson, T. McCanny, S. Shimizu,
507
R. J. Clarke, D. Neely, K. Spohr, R. Chapman, R. P. Singhal, K. Krushelnick, M. S. Wei,
508
and P. A. Norreys, “Characterization of proton and heavier ion acceleration in ultrahigh-
509
intensity laser interactions with heated target foils,” Phys. Rev. E 70, 036405 (2004).
510
24
Y. Fukuda, A. Y. Faenov, M. Tampo, T. A. Pikuz, T. Nakamura, M. Kando, Y. Hayashi,
511
A. Yogo, H. Sakaki, T. Kameshima, A. S. Pirozhkov, K. Ogura, M. Mori, T. Z. Esirkepov,
512
J. Koga, A. S. Boldarev, V. A. Gasilov, A. I. Magunov, T. Yamauchi, R. Kodama, P. R.
513
Bolton, Y. Kato, T. Tajima, H. Daido, and S. V. Bulanov, “Energy increase in multi-MeV
514
ion acceleration in the interaction of a short pulse laser with a cluster-gas target,” Phys.
515
Rev. Lett. 103, 165002 (2009).
516
25
A. Henig, S. Steinke, M. Schnürer, T. Sokollik, R. Hörlein, D. Kiefer, D. Jung, J. Schreiber,
517
B. M. Hegelich, X. Q. Yan, J. Meyer-ter Vehn, T. Tajima, P. V. Nickles, W. Sandner, and
518
D. Habs, “Radiation-pressure acceleration of ion beams driven by circularly polarized laser
519
pulses,” Phys. Rev. Lett. 103, 245003 (2009).
23
520
26
Y. Fukuda, H. Sakaki, M. Kanasaki, A. Yogo, S. Jinno, M. Tampo, A. Faenov, T. Pikuz,
521
Y. Hayashi, M. Kando, A. Pirozhkov, T. Shimomura, H. Kiriyama, S. Kurashima,
522
T. Kamiya, K. Oda, T. Yamauchi, K. Kondo, and S. Bulanov, “Identification of high
523
energy ions using backscattered particles in laser-driven ion acceleration with cluster-gas
524
targets,” Radiation Measurements 50, 92 – 96 (2013).
525
27
D. C. Carroll, O. Tresca, R. Prasad, L. Romagnani, P. S. Foster, P. Gallegos, S. Ter-
526
Avetisyan, J. S. Green, M. J. V. Streeter, N. Dover, C. A. J. Palmer, C. M. Brenner,
527
F. H. Cameron, K. E. Quinn, J. Schreiber, A. P. L. Robinson, T. Baeva, M. N. Quinn,
528
X. H. Yuan, Z. Najmudin, M. Zepf, D. Neely, M. Borghesi, and P. McKenna, “Carbon
529
ion acceleration from thin foil targets irradiated by ultrahigh-contrast, ultraintense laser
530
pulses,” New Journal of Physics 12, 045020 (2010).
531
28
A. Henig, D. Kiefer, K. Markey, D. C. Gautier, K. A. Flippo, S. Letzring, R. P. Johnson,
532
T. Shimada, L. Yin, B. J. Albright, K. J. Bowers, J. C. Fernández, S. G. Rykovanov, H.-C.
533
Wu, M. Zepf, D. Jung, V. K. Liechtenstein, J. Schreiber, D. Habs, and B. M. Hegelich,
534
“Enhanced laser-driven ion acceleration in the relativistic transparency regime,” Phys.
535
Rev. Lett. 103, 045002 (2009).
536
29
D. Jung, B. J. Albright, L. Yin, D. C. Gautier, R. Shah, S. Palaniyappan, S. Letzring,
537
B. Dromey, H.-C. Wu, T. Shimada, R. P. Johnson, M. Roth, J. C. Fernández, D. Habs, and
538
B. M. Hegelich, “Beam profiles of proton and carbon ions in the relativistic transparency
539
regime,” New Journal of Physics 15, 123035 (2013).
540
30
B. M. Hegelich, I. Pomerantz, L. Yin, H. C. Wu, D. Jung, B. J. Albright, D. C. Gautier,
541
S. Letzring, S. Palaniyappan, R. Shah, K. Allinger, R. Hörlein, J. Schreiber, D. Habs,
542
J. Blakeney, G. Dyer, L. Fuller, E. Gaul, E. Mccary, A. R. Meadows, C. Wang, T. Dit-
543
mire, and J. C. Fernández, “Laser-driven ion acceleration from relativistically transparent
544
nanotargets,” New Journal of Physics 15, 085015 (2013).
545
31
L. Willingale, S. P. D. Mangles, P. M. Nilson, R. J. Clarke, A. E. Dangor, M. C. Kaluza,
546
S. Karsch, K. L. Lancaster, W. B. Mori, Z. Najmudin, J. Schreiber, A. G. R. Thomas,
547
M. S. Wei, and K. Krushelnick, “Collimated multi-MeV ion beams from high-intensity
548
laser interactions with underdense plasma,” Phys. Rev. Lett. 96, 245002 (2006).
549
32
duced by pulsed laser deposition,” Carbon 56, 358 – 365 (2013).
550
551
A. Zani, D. Dellasega, V. Russo, and M. Passoni, “Ultra-low density carbon foams pro-
33
S. Jan, D. Benoit, E. Becheva, T. Carlier, F. Cassol, P. Descourt, T. Frisson, L. Grevillot,
24
552
L. Guigues, L. Maigne, C. Morel, Y. Perrot, N. Rehfeld, D. Sarrut, D. R. Schaart, S. Stute,
553
U. Pietrzyk, D. Visvikis, N. Zahra, and I. Buvat, “GATE V6: a major enhancement of
554
the GATE simulation platform enabling modelling of CT and radiotherapy,” Phys. Med.
555
Biol. 56, 881 (2011).
556
34
of Scientific Instruments 78, 096104 (2007).
557
558
S. Y. Lee, W. M. Tam, and Z. Liu, “Strip injection for carbon ion synchrotrons,” Review
35
F. Pönisch, K. Parodi, B. G. Hasch, and W. Enghardt, “The modelling of positron emitter
559
production and PET imaging during carbon ion therapy,” Phys. Med. Biol 49, 5217–32
560
(2004).
561
36
(OECD NEA Data Bank, 2012).
562
563
564
N. Soppera, E. Dupont, and M. Bossant, JANIS Book of proton-induced cross-sections
37
A. Sgattoni, P. Londrillo, A. Macchi, and M. Passoni, “Laser ion acceleration using a
solid target coupled with a low-density layer,” Phys. Rev. E 85, 036405 (2012).
25