un nuevo formalismo para dosimetria de referencia de pequeños campos

Medical Physics Letter
A new formalism for reference dosimetry of small and nonstandard fields
R. Alfonso
International Atomic Energy Agency, A-1400 Vienna, Austria and Instituto Nacional de Oncologia y
Radiobiologia, CP 10400 La Habana, Cuba
P. Andreo
International Atomic Energy Agency, A-1400 Vienna, Austria and University of Stockholm-Karolinska
Institutet, SE-17176 Stockholm, Sweden
R. Capote
International Atomic Energy Agency, A-1400 Vienna, Austria
M. Saiful Huq
University of Pittsburgh Cancer Institute, Pittsburgh, Pennsylvania 15232
W. Kilby
Accuray Inc., Sunnyvale, California 94089
P. Kjäll
Elekta Instrument AB, SE-10393 Stockholm, Sweden
T. R. Mackie
Department of Medical Physics, University of Wisconsin, Madison, Wisconsin 53706
H. Palmansa兲
National Physical Laboratory, Teddington, Middx TW11 0LW, United Kingdom and Slovenský Metrologický
Ústav, SK-84255 Bratislava, Slovakia
K. Rosser
Royal Marsden NHS Foundation Trust, Sutton, Surrey SM2 5PT, United Kingdom
J. Seuntjens
Medical Physics, McGill University, Montreal, Québec H3G 1A4, Canada
W. Ullrich
BrainLAB AG, D-85622 Feldkirchen, Germany
S. Vatnitsky
International Atomic Energy Agency, A-1400 Vienna, Austria
共Received 2 June 2008; revised 2 October 2008; accepted for publication 2 October 2008;
published 28 October 2008兲
The use of small fields in radiotherapy techniques has increased substantially, in particular in
stereotactic treatments and large uniform or nonuniform fields that are composed of small fields
such as for intensity modulated radiation therapy 共IMRT兲. This has been facilitated by the increased
availability of standard and add-on multileaf collimators and a variety of new treatment units. For
these fields, dosimetric errors have become considerably larger than in conventional beams mostly
due to two reasons; 共i兲 the reference conditions recommended by conventional Codes of Practice
共CoPs兲 cannot be established in some machines and 共ii兲 the measurement of absorbed dose to water
in composite fields is not standardized. In order to develop standardized recommendations for
dosimetry procedures and detectors, an international working group on reference dosimetry of small
and nonstandard fields has been established by the International Atomic Energy Agency 共IAEA兲 in
cooperation with the American Association of Physicists in Medicine 共AAPM兲 Therapy Physics
Committee. This paper outlines a new formalism for the dosimetry of small and composite fields
with the intention to extend recommendations given in conventional CoPs for clinical reference
dosimetry based on absorbed dose to water. This formalism introduces the concept of two new
intermediate calibration fields: 共i兲 a static machine-specific reference field for those modalities that
cannot establish conventional reference conditions and 共ii兲 a plan-class specific reference field
closer to the patient-specific clinical fields thereby facilitating standardization of composite field
dosimetry. Prior to progressing with developing a CoP or other form of recommendation, the
members of this IAEA working group welcome comments from the international medical physics
community on the formalism presented here. © 2008 American Association of Physicists in Medicine. 关DOI: 10.1118/1.3005481兴
5179
Med. Phys. 35 „11…, November 2008
0094-2405/2008/35„11…/5179/8/$23.00
© 2008 Am. Assoc. Phys. Med.
5179
5180
Alfonso et al.: Reference dosimetry of small and nonstandard fields
5180
Key words: small field, composite field, reference dosimetry, formalism
I. INTRODUCTION
Recent developments in radiotherapy delivery techniques
have substantially increased the use of small fields for stereotaxy and of larger uniform or nonuniform fields that are
composed of small fields. This is the case for all treatment
beam types including high-energy photon beams, electron
beams, proton beams, and light-ion beams. A small field is
defined as a field with a size smaller than the lateral range of
charged particles. Nonstandard fields are either made of
small fields or whenever nonequilibrium conditions exist;
this occurs, for example, when the size of the penumbrae is
similar to the field size.1 The recent advances have become
possible due to technological changes in conventional accelerators that have improved mechanical accuracy, stability
and dosimetric control. At the same time, there has been an
increasing availability in the clinic of standard-, mini- and
micro-multileaf collimators 共MLCs兲 on conventional accelerators 共e.g., BrainLAB—m3®High-Resolution Multileaf
Collimator, BrainLAB AG, Feldkirchen, Germany兲, as well
as the introduction of treatment units specifically designed
for stereotaxy 关GammaKnife 共Leksell GammaKnife®, Elekta
Instrument AB, Stockholm, Sweden兲, CyberKnife® Robotic
Radiosurgery System, Accuray Inc., Sunnyvale, CA兴 or intensity modulated treatments 共TomoTherapy® Hi-Art®, TomoTherapy Inc., Madison, WI, USA兲.
These developments have increased the uncertainty of
clinical dosimetry and its link to reference dosimetry based
on Codes of Practice 共CoPs兲 or dosimetry protocols. At the
same time, dosimetry errors have become considerably larger
than in conventional beams as is illustrated by various references, including Das et al.,1 Verhaegen et al.,2 and the discrepancies between Monte Carlo calculated and measured
outputs by different detectors shown in Fig. 1.3 Moreover, in
some of the treatment units mentioned above, the reference
conditions specified in current dosimetry protocols for beam
calibration cannot be realized.
Ionization chambers, which have been the “backbone” of
radiotherapy dosimetry, are not always suitable where situations of high dose gradients, time-dose variance, and nonuniform beam distributions are encountered. Volume averaging
and lack of electronic equilibrium, which requires a sufficiently large region of uniform beam intensity surrounding
the detector, complicate the use of ionization chambers for
the dosimetry of small photon beams. A large detector, such
as an ionization chamber, perturbs particle fluence in the medium. This implies that the conversion from ionization to
absorbed dose to water based on cavity theory and using the
currently available perturbation factors used in existing dosimetry CoPs or protocols such as IAEA TRS-3984 and
AAPM TG-515 is not accurate.6–8 Furthermore, spectra, and,
therefore, beam quality, may change as the field size
decreases.2
At some treatment units, the use of water phantoms for
reference dosimetry is possible but highly inconvenient, and,
therefore, plastic phantoms may be necessary. Today, pure
plastic materials such as polymethylmethacrylate 共PMMA,
Lucite兲 and bisphenol-A polycarbonate have well-controlled
densities, well defined atomic properties, and can be machined for accurate positioning of dosimeters.
Therefore, there is a clear need for a methodology that
complements the dosimetry in the reference conditions recommended in existing CoPs. The aim of this paper is to
present a methodology that has been developed at two meetings held at the IAEA in December 2007 and in May 2008.
This may form the basis of new international recommendations for the dosimetry of small and nonstandard treatment
fields and will provide a standardized framework for establishing traceable dosimetry for the wide variety of beams and
delivery techniques now available and others that will become available in the future.
II. FORMALISM
II.A. General formalism and definitions
The core of the formalism consists of two related routes to
determine the absorbed dose to water in external beam radiotherapy using ionization chambers in situations different
from the conventional reference conditions for which dosimetry CoPs apply, usually related to special or complex treatment deliveries. Both routes require the extension of the concept of reference field to incorporate small and nonstandard
fields as well as modified reference conditions such as phantom shape and material. These two routes are:
FIG. 1. Ratios of Monte Carlo calculated absorbed doses to water 共square
symbols; right hand vertical axis兲 and ratios of measured readings of different detectors 共other symbols; left hand vertical axis兲 as a function of field
size, both normalized to the corresponding quantity for a 10 cm⫻ 10 cm
field. The results shown are for a 6 MV beam at a depth of 5.0 cm in RW3
water equivalent plastic 共PTW, Freiburg兲 关data from Ref. 3兴.
Medical Physics, Vol. 35, No. 11, November 2008
1.
Small static-field dosimetry, traceable to a broad-beam
calibration, which introduces an intermediate step
through a machine-specific-reference field 共msr兲 for
treatment machines that cannot establish a conventional
reference field.
5181
2.
Alfonso et al.: Reference dosimetry of small and nonstandard fields
Composite-field dosimetry, traceable to a broad-beam
calibration, which can include an intermediate machinespecific-reference field if needed, as well as a so-called
plan-class specific reference field 共pcsr兲 as defined below. Note that a plan-class specific reference field can be
a 3D irradiated volume or a 4D delivery sequence. The
pcsr should be as close as possible to a class of clinical
plans of interest, and provide a uniform dose over a
region exceeding the dimensions of a reference detector.
The following calibration and delivery fields are introduced:
f ref denotes the conventional reference field in dosimetry
CoPs for which the calibration coefficient of an ionization
chamber in terms of absorbed dose to water has been provided by a standards laboratory.
f msr denotes a machine-specific reference field, for static
modalities or treatment machines that cannot establish the
conventional reference field. Examples of machine-specific
reference fields are the 6 cm diam collimator field in a CyberKnife, the 1.6/ 1.8 cm diam collimator field in the GammaKnife, or the 5 cm⫻ 20 cm static field in TomoTherapy
equipment. In general, the machine-specific reference field
should be as close as possible to the conventional reference
field in existing dosimetry CoPs.4,5
f pcsr denotes a plan-class specific reference field that represents a class of dynamic or step-and-shoot delivery fields,
or a combination of fields, such that full charged-particle
equilibrium 共CPE兲 is achieved in a time-average sense at the
position of the detector 共as opposed to transient chargedparticle equilibrium in conventional broad-beam dosimetry兲.
Different plan-class specific reference fields can be defined
for different treatment sites and can utilize a delivery protocol agreed upon with the manufacturer. Examples of possible
plan-class specific reference fields are a composed homogeneous dose distribution delivered to a 10 cm diam ⫻10 cm
long cylindrical irradiation volume in a 20 cm diam cylindrical water equivalent phantom in a tomotherapy or other
linac-based IMRT machine; a multiple-shot plan delivered
with a GammaKnife unit, a square IMRT field generated by
adding small square fields, and a 10 cm⫻ 10 cm square field
homogeneously spread out over a depth of 10 cm for a
scanned, range-modulated proton beam.
f clin denotes the clinical radiation field for which the absorbed dose to water needs to be determined.
The proposed formalism is applicable to both small static
fields and composite fields. For the sake of clarity, the formalism will first be described for small static fields, followed
by a discussion for composite fields.
II.A.1. Small static fields
f msr
The absorbed dose to water, Dw,Q
, at the reference
msr
depth in water, in a beam of quality Qmsr and reference field
f msr and in the absence of the chamber is given by
f msr
Dw,Q
msr
= M Qf msr · ND,w,Q0 · kQ,Q0 · kQf msr,f,Qref ,
msr
msr
where:
Medical Physics, Vol. 35, No. 11, November 2008
共1兲
5181
Q is the beam quality of the conventional reference field
f ref according to an established dosimetry CoP.
Qmsr is the beam quality of the machine-specific reference
field f msr. If the msr field is in the same machine as the ref
field, as will mostly be the case, the difference in beam quality is due purely to the difference in field size between f msr
and f ref 共and possibly other conditions of geometry and phantom material兲. If, in addition, the msr field is large enough to
preserve charged particle equilibrium, the beam quality will
be equal to Q and, strictly, a different notation is not required. However, the usage of a different notation is maintained here to indicate that the beam quality could be different.
M Qf msr is the reading of the dosimeter in the field f msr
msr
corrected for influence quantities, such as pressure, temperature, incomplete charge collection, and polarity effects.
ND,w,Q0 is the calibration coefficient in terms of absorbed
dose to water for an ionization chamber at a reference beam
quality Q0 共usually 60Co兲. ND,w,Q0 is measured at the standards laboratory for a reference field of size 10 cm⫻ 10 cm.
kQ,Q0 is the beam-quality correction factor, which corrects
for the differences between the reference beam quality Qo at
the standards laboratory and the beam quality Q of the conventional reference field f ref.
kQf msr,f,Qref is a factor that corrects for the differences between
msr
the conditions of field size, geometry, phantom material, and
beam quality of the conventional reference field f ref and the
machine-specific reference field f msr. This is a generalized
version of the classical beam-quality correction factor. If the
field size and all other conditions of geometry and phantom
material, i.e., water, are the same, it reduces to a conventional beam-quality correction factor. In that sense, the
beam-quality correction factor in TRS-398 can be regarded
as a special case of this newly introduced factor.
The factor kQf msr,f,Qref accounts for the difference between the
msr
responses of an ionization chamber in the fields f ref and f msr
and is defined as
kQf msr,f,Qref =
msr
f msr
Dw,Q
/M Qf msr
msr
msr
f ref
Dw,Q
/M Qfref
,
共2兲
and represents an extension from the established CoP. It is
expected that the change in beam quality between the conventional reference field and the machine-specific-reference
field is minor. Ideally, this factor would be obtained by a
direct calibration of the ionization chamber in the two fields
against a primary standard or against another dosimeter such
as alanine, radiochromic film, or ferrous sulphate dosimetry
whose calibration is traceable to a primary standard of absorbed dose to water and does not exhibit a substantial beam
quality dependence. Alternatively, it could be calculated by a
Monte Carlo simulation alone 共as for example explained in
Ref. 9兲 or it could be measured using a suitable detector
applying corrections obtained from a Monte Carlo simulation.
5182
Alfonso et al.: Reference dosimetry of small and nonstandard fields
Equations 共1兲 and 共2兲 describe the formalism for reference
dosimetry. For the purposes of relative dosimetry a field factor is introduced.
f clin
, at a reference point
The absorbed dose to water, Dw,Q
clin
in a phantom for a clinical field f clin of quality Qclin and in
the absence of the chamber is given by
f clin
Dw,Q
clin
f msr
= Dw,Q
msr
· ⍀Qf clin,f,Qmsr ,
clin
共3兲
msr
where Qclin is the beam quality of the clinical field f clin and
⍀Qf clin,f,Qmsr is a field factor that converts the absorbed dose to
clin msr
water for the machine-specific reference field f msr to the absorbed dose to water for the clinical field f clin. In relative
dosimetry of single static fields, this factor is conventionally
called a field output factor.
From Eq. 共3兲, it is clear that this field factor is defined as
a ratio of absorbed doses to water. It can be calculated directly as a ratio of absorbed doses to water using Monte
Carlo alone. Alternatively, the field factor can be measured
as a ratio of detector readings multiplied by a Monte Carlo
calculated correction factor kQf clin,f,Qmsr 共similar to factors calclin msr
culated in Refs. 6–8兲, which accounts for the difference between the detector response in the fields f clin and f msr according to:
⍀Qf clin,f,Qmsr =
clin msr
M Qf clin
clin
M Qf msr
msr
·
冋
f clin
Dw,Q
/M Qf clin
clin
clin
f msr
Dw,Q
/M Qf msr
msr
msr
册
,
共4a兲
or
⍀Qf clin,f,Qmsr =
clin msr
M Qf clin
clin
M Qf msr
·
kQf clin,f,Qmsr ,
clin msr
共4b兲
msr
showing that if the correction factor kQf clin,f,Qmsr is close to
clin msr
unity for a given detector, then the ratio of readings alone is
a sufficiently accurate approximation to the field factor. For
any detector not satisfying this condition, the correction factor in square brackets needs to be taken into account.
For some treatment delivery systems that utilize an
add-on beam modifier, such as a BrainLab mini multileaf
attached to a conventional linac, the conventional reference
field can be realized by removing the add on. In this case
there is, strictly, no need for the intermediate machinespecific reference field. Equation 共1兲 then reduces to
f ref
= M Qfref · ND,w,Q0 · kQ,Q0 ,
Dw,Q
共5兲
which is essentially the formalism for the determination of
absorbed dose to water in a reference field according to
IAEA TRS-398 or AAPM TG-51. The relative dosimetry
step then requires a field factor ⍀Qf clin,f,Qref, which converts the
clin
absorbed dose to water in the field f ref to the absorbed dose
to water in the field f clin. However, the concept of the
machine-specific reference field remains useful for machines
with an add on since it may not be practical to refer every
small field measurement to the conventional reference field.
The field factor above can then be regarded as the product of
two field factors:
Medical Physics, Vol. 35, No. 11, November 2008
5182
⍀Qf clin,f,Qref = ⍀Qf msr,f,Qref · ⍀Qf clin,f,Qmsr .
clin
msr
clin
共6兲
msr
For beams in which the add on cannot be removed or the
conventional reference field cannot be established in any
way, the conventional reference field is a hypothetical field.
The relation between the absorbed dose to water in the msr
field and the ref field may then have to be determined experimentally through the use of another treatment unit with similar beam quality.
As a summary, a graphical representation of the procedures of route 1 for dosimetry in small static fields is given
in Fig. 2.
II.A.2. Composite fields
The absorbed dose to water in a composite field, during a
step-and-shoot or a dynamic IMRT delivery, a TomoTherapy,
a GammaKnife, a CyberKnife, a scanned proton beam delivery, etc. can also be determined using Eq. 共1兲.
However, instead of a machine-specific reference field,
preference is given to use a different type of intermediate
field, which is a plan-class specific reference field 共pcsr兲.
This is a reference field for a class of dynamic or step-andshoot delivery fields, or a class of combinations of fields in a
configuration that is as close as possible to the final clinical
delivery scheme, but delivers a homogeneous absorbed dose
to an extended and geometrically simple target volume. In
this case, the msr field is replaced with the pcsr field and Eq.
共1兲 converts to:
f pcsr
Dw,Q
pcsr
= M Qf pcsr · ND,w,Q0 · kQ,Q0 · kQf pcsr,f,Qref ,
pcsr
pcsr
共7兲
and Eq. 共3兲 for relative dosimetry converts to:
f clin
Dw,Q
clin
f pcsr
= Dw,Q
pcsr
· ⍀Qf clin,f,Qpcsr .
clin
pcsr
共8兲
The determination of the correction factor kQf pcsr,f,Qref from the
pcsr
conventional reference field to the plan-class specific reference field in Eq. 共7兲, which might require different setups,
can be established with tight tolerance levels 共as in conventional reference dosimetry兲. On the other hand, the relative
conversion of absorbed dose to water from the plan-class
specific reference field to the clinical field in Eq. 共8兲, which
will have to be established for every treatment delivery, can
be performed in the same setup using the same phantom,
e.g., for patient specific QA in IMRT, the phantom should be
the same as for the dosimetry in the pcsr field with a crosscalibrated field chamber 共see below兲.
The factor kQf pcsr,f,Qref will generally be close to unity under
pcsr
the condition that the addition and geometrical matching of
fields in the homogeneous phantom compensates for the loss
of charged-particle equilibrium in the penumbrae of individual fields. Ideally, it would be obtained by a direct calibration of the ionization chamber in the conventional reference field and the plan-class specific reference field against a
primary standard or against another dosimeter such as alanine, radiochromic film, or ferrous sulphate dosimetry whose
calibration is traceable to a primary standard of absorbed
dose to water. Alternatively, it could be calculated by a
5183
Alfonso et al.: Reference dosimetry of small and nonstandard fields
5183
FIG. 2. Schematic overview of the dosimetry of small static fields with reference to a machine-specific reference field according to the formalism presented
in this paper.
Monte Carlo simulation alone or it could be measured using
a suitable detector applying corrections obtained from a
Monte Carlo simulation.
In this second route, the possibility also exists that a treatment unit cannot establish a conventional reference field and
that the dosimetry of the plan-class specific reference field is
referred to a machine-specific reference field. An additional
factor is then required to correct between the msr and the
pcsr fields, i.e.,
f pcsr
Dw,Q
pcsr
= M Qf pcsr · ND,w,Q0 · kQ,Q0 · kQf msr,f,Qref · kQf pcsr,f,Qmsr .
pcsr
msr
pcsr
msr
共9兲
As a summary, a graphical representation of the procedures
of route 2 for dosimetry in nonstandard composite fields is
given in Fig. 3.
In both cases, small and composite fields, it can be seen
that the formalism stays close to the one of conventional
CoPs in the sense that a calibration of a reference field is
performed followed by the application of output factors or an
equivalent type of factors for clinical fields, the main difference being the extension of the concept of reference field.
In line with this approach, it is possible that a standards
laboratory is able to provide a direct calibration coefficient
f msr
for an ionization chamber in the machine-specificND,w,Q
msr
f pcsr
reference field or ND,w,Q
for the plan-class specific referpcsr
ence field. From these, the factors kQf msr,f,Qref and kQf pcsr,f,Qref in Eqs.
msr
pcsr
共1兲 and 共7兲 can be derived. This opens the path for introducMedical Physics, Vol. 35, No. 11, November 2008
ing a cross-calibration procedure. If a detector is available,
which, based on a calibration in a standards laboratory for a
conventional reference field, can reliably measure absorbed
dose to water in the machine-specific-reference field or in the
plan-class specific reference field, it can be used to crosscalibrate a second detector for subsequent use as a reference
dosimeter in the clinical field.
In what follows two example applications of the dosimetry of small and nonstandard fields will be discussed to illustrate the use of the different correction and field factors. They
are not meant to be comprehensive and there are many more
examples from the literature that can be found. These will be
extensively reviewed in future activities of the IAEA working group.
III. EXAMPLES
III.A. Application to TomoTherapy
The TomoTherapy HiArt system delivers helical tomotherapy plans using a nominal 6 MV unflattened x-ray beam
and a binary multileaf collimator. For this system, the reference conditions for applying TRS-398 or TG-51, i.e., a
10 cm⫻ 10 cm homogeneous field, cannot be established. In
addition, the calibration procedure is integrated in the
fluence-based planning and delivery. Dosimetry according to
route 1 共small static fields兲 can form part of a QA system to
5184
Alfonso et al.: Reference dosimetry of small and nonstandard fields
5184
FIG. 3. Schematic overview of the dosimetry of nonstandard composite fields with reference to a plan-class specific reference field according to the formalism
presented in this paper 关note that in the route starting from the hypothetical field, Eq. 共9兲 applies instead of the equation in the figure兴.
verify, and if necessary, adjust the calibration of the monitor,
but has no relevance to the user regarding the calculated
output of a helical delivery. In this case, the relevant dosimetry methods for the user are described by route 2 共composite
fields兲.
A first issue with the TomoTherapy unit is beam quality
since high-energy x-ray dosimetry in TG-51 and TRS-398
has been developed mainly based on experience with photon
beams that contain flattening filters and, thus, have a harder
energy spectrum. On the other hand, it has been argued that
in the energy region of beams with nominal energies around
6 MV, the variation of the beam quality correction factors
kQ,Q0 with beam quality Q is small.10,11
Factors equivalent to kQf msr,f,Qref were calculated using the
msr
Monte Carlo method by Jeraj et al.10 for a generic
secondary-standard-level graphite-walled Farmer-type chamber 共cfr. Table I兲 and Thomas et al. for an Exradin A1SL
chamber. For a machine-specific reference field f msr of
5 cm⫻ 10 cm at 85 cm SSD in the TomoTherapy unit and a
hypothetical 10 cm⫻ 10 cm reference field f ref, the correction factor was unity to within 0.3% for both chamber
types.10,11
These findings were experimentally confirmed by Duane
et al.,12 who determined calibration coefficients for an
NE2611 and an Exradin A1SL ionization chamber. They
were obtained by comparison with the therapy-level alaninedosimetry service13 from the National Physical Laboratory
Medical Physics, Vol. 35, No. 11, November 2008
共NPL兲 in 5 cm⫻ 10 cm and 5 cm⫻ 20 cm static fields at
85 cm SSD in four different TomoTherapy HiArt units. The
calibration coefficients agreed to within 0.3% with those determined in NPL beams with the same values of TPR20,10
共ranging between 0.62 and 0.64兲.
Table I also shows the factor kQf pcsr,f,Qref derived from Dupcsr
ane’s alanine work12 in a TomoTherapy unit at the Cromwell
Hospital, London, UK. The calibration coefficient for two
ionization chambers 共one of type NE2611 and one of type
Exradin A1SL兲 derived by calibration against NPL’s alanine
TABLE I. Correction factors kQfmsr,f ref derived from a Monte Carlo study 共Ref.
msr,Q
10兲 for a generic graphite-walled Farmer type chamber 共second column兲 and
fpcsr,f ref
derived from an experimental study 共Ref. 12兲 for a NE2611
factors kQ
pcsr,Q
ionization chamber 共last column兲. Route 1 refers to the calibration in the msr
field of 5 cm⫻ 10 cm; route 2 refers to calibration in the pcsr field, which is
a helically delivered composite field of 8.0 cm diam and 10.0 cm length in
a cylindrical clear polystyrene phantom. The axial collimations 共fan widths兲
for the pcsr fields were 5.0, 2.5, and 1.0 cm.
kQf msr,f,Qref
kQf pcsr,f,Qref
msr
f msr 共route 1兲 共Ref. 10兲
5 cm⫻ 10 cm
2 cm⫻ 10 cm
2 cm⫻ 2 cm
0.997
0.993
0.990
pcsr
f pcsr 共route 2兲 共Ref. 12兲
Helical, 5 cm
Helical, 2.5 cm
Helical, 1 cm
1.000
1.000
0.997
5185
Alfonso et al.: Reference dosimetry of small and nonstandard fields
TABLE II. kQf clin,f,Qmsr factors derived from Monte Carlo simulations18 for the
clin msr
three smallest fixed circular collimators 共0.5, 0.75, and 1.0 cm diam兲 of a
CyberKnife unit with reference to the msr field of 6.0 cm diam. Values of
kQf clin,f,Qmsr are stated for a range of electron beam spatial full width at half
clin msr
maximum 共FWHM兲 values between 1.4 and 2.6 mm.
Field size
Detector
Exradin A16
PTW 31014 共PinPoint兲
PTW 60012 共diode兲
PTW 60003 共diamond兲
0.5 cm
0.75 cm
1.0 cm
1.067–1.112
1.082–1.124
0.940–0.957
1.066–1.123
1.017–1.027
1.024–1.037
0.966–0.967
1.001–1.012
1.007–1.012
1.013–1.017
0.978–0.978
0.999–1.001
indicates that kQf pcsr,f,Qref is unity within the experimental relapcsr
tive standard uncertainty 共which is reported to be 0.8% for
this specific example12兲.
III.B. Application to CyberKnife
The CyberKnife system cannot achieve a 10 cm⫻ 10 cm
square reference field. The machine specific reference field,
f msr, is defined by a fixed collimator that produces a beam
diam of 6 cm at the standard treatment distance of 80 cm.
This is the largest field size available.
TPR20,10 is measured in the field f msr using SCD= 80 cm
in water. The equivalent square field size S is 5.4 cm, based
on the BJR supplement 25 conversion method.14 TPR20,10共S兲
values for these treatment units are typically in the range
0.629–0.647; they can be converted to equivalent values for
f ref 共i.e., a 10 cm⫻ 10 cm square field兲 using data tables in
BJR supplement 2514 as described by Sharma et al.15 This
corrected TPR20,10 value is then used to select kQ,Q0 from a
conventional CoP 共alternatively, beam quality may be specified using a depth-dose ratio measured at 100 cm SSD. In
this case S = 6.75 cm. This PDD may be converted to a
10 cm⫻ 10 cm field using the method described here, and
this value used to obtain kQ,Q0兲.
As the field f msr is large enough to preserve chargedparticle equilibrium on the central axis, no significant difference in beam quality is expected between f msr and f ref and,
therefore, kQf msr,f,Qref is unity. The kQ,Q0 value determined using
msr
this method agrees to within 0.1%,16 and 0.4%17 with Monte
Carlo calculations of the product kQ,Q0 · kQf msr,f,Qref.
msr
Francescon et al.18 have reported a Monte Carlo simulation method for deriving correction factors equivalent to
kQf clin,f,Qmsr in Eq. 共4b兲. These values are given in Table II. The
clin msr
method was validated by deriving and applying the kQf clin,f,Qmsr
clin msr
to determine ⍀Qf clin,f,Qmsr from ratios of detector readings obclin msr
tained with all four detectors, resulting in excellent agreement between the four corrected measurement results obtained for each collimator 共relative standard deviation
艋 0.4%兲.
IV. CONCLUSIONS
A new formalism is presented for the dosimetry of small
and composite fields. The concept of two new intermediate
Medical Physics, Vol. 35, No. 11, November 2008
5185
calibration fields is introduced: 共i兲 a static machine-specific
reference field for those modalities that cannot establish conventional reference conditions and 共ii兲 a plan-class specific
reference field closer to the patient-specific clinical fields
thereby facilitating standardization of composite field dosimetry. This formalism may form the basis of new international
recommendations for the dosimetry of small and nonstandard
treatment fields and will provide a standardized framework
for establishing traceable dosimetry for the wide variety of
beams and delivery techniques currently available. It will
further harmonize research efforts in this field so that data
produced in line with this formalism become more consistent
and more widely usable.
We want to emphasize that this paper presents a framework and does not aim to present a code of practice or recommendation for its practical implementation. For enabling
the latter, a substantial research effort is still required. For
static field dosimetry, where a lot of data are available in the
literature, the working group will undertake a thorough review and determine areas where sufficient data are lacking.
For composite and dynamic field dosimetry, the main challenge is the definition of suitable pcsr fields. The working
group will propose procedures to come up with and test pcsr
fields based on a class of clinical plans and will actively
participate in testing these procedures. Prior to progressing
with developing a CoP or other form of recommendation, the
members of this IAEA working group welcome comments
from the international medical physics community on the
formalism presented here.
ACKNOWLEDGMENTS
The authors would like to thank the AAPM Therapy Physics Committee for endorsing this initiative. The authors are
further thankful for input and comments from Ahmed
Meghzifene 共IAEA兲, Joanna Izewska 共IAEA兲, Ken Shortt
共IAEA兲, Robert Jeraj 共University of Wisconsin, Madison WI,
USA兲, and Simon Duane 共National Physical Laboratory, Teddington, UK兲, and for comments on this manuscript by Maria
Mania Aspradakis 共TomoTherapy, Zurich, Switzerland兲, Lee
Walton 共Royal Hallamshire Hospital, Sheffield, UK兲, Stefaan
Vynckier 共Catholic University of Louvain, Belgium兲,
Stephen Seltzer 共National Institute of Standards and Technology, Gaithersburg, MD, USA兲, and David Rogers 共Carleton
University, Ottawa, Canada兲.
a兲
Author to whom correspondence should be addressed. Electronic mail:
[email protected]
I. J. Das, G. X. Ding, and A. Ahnesjö, “Small fields: Nonequilibrium
radiation dosimetry,” Med. Phys. 35, 206–215 共2008兲.
2
F. Verhaegen, I. J. Das, and H. Palmans, “Monte Carlo dosimetry study of
a 6 MV stereotactic radiosurgery unit,” Phys. Med. Biol. 43, 2755–2768
共1998兲.
3
F. Sánchez-Doblado, G. H. Hartmann, J. Pena, J. V. Roselló, G. Russiello,
and D. M. Gonzalez-Castaño, “A new method for output factor determination in MLC shaped narrow beams,” Phys. Medica 23, 58–66 共2007兲.
4
P. Andreo, D. T. Burns, K. Hohlfeld, M. S. Huq, T. Kanai, F. Laitano, V.
G. Smyth, and S. Vynckier, “Absorbed dose determination in external
beam radiotherapy,” International Atomic Energy Agency, Vienna, IAEA
Technical Report Series No. 398, 2000.
5
P. R. Almond, P. J. Biggs, B. M. Coursey, W. F. Hanson, M. S. Huq, R.
Nath, and D. W. Rogers, “AAPM’s TG51 protocol for clinical reference
1
5186
Alfonso et al.: Reference dosimetry of small and nonstandard fields
dosimetry of high-energy photon and electron beams,” Med. Phys. 26,
1847–1870 共1999兲.
6
P. Francescon, S. Cora, C. Cavedon, P. Scalchi, S. Reccanello, and F.
Colombo, “Use of a new type of radiochromic film, a new parallel-plate
micro-chamber, MOSFETs, and TLD 800 microcubes in the dosimetry of
small beams,” Med. Phys. 25, 503–511 共1998兲.
7
H. Bouchard and J. Seuntjens, “Ionization chamber-based reference dosimetry of intensity modulated radiation beams,” Med. Phys. 31, 2454–
2465 共2004兲.
8
R. Capote, F. Sánchez-Doblado, A. Leal, J. I. Lagares, R. Arrans, and G.
H. Hartmann, “An EGSnrc Monte Carlo study of the microionization
chamber for reference dosimetry of narrow irregular IMRT beamlets,”
Med. Phys. 31, 2416–2422 共2004兲.
9
J. Sempau, P. Andreo, J. Aldana, J. Mazurier, and F. Salvat, “Electron
beam quality correction factors for plane-parallel ionization chambers:
Monte Carlo calculations using the PENELOPE system,” Phys. Med.
Biol. 49, 4427–4444 共2004兲.
10
R. Jeraj, T. R. Mackie, J. Balog, and G. Olivera, “Dose calibration of
non-conventional treatment systems applied to helical tomotherapy,”
Med. Phys. 32, 570–577 共2005兲.
11
S. D. Thomas, M. Mackenzie, D. W. O. Rogers, and B. G. Fallone, “A
Monte Carlo derived TG-51 equivalent calibration for helical tomo-
Medical Physics, Vol. 35, No. 11, November 2008
5186
therapy,” Med. Phys. 32, 1346–1353 共2005兲.
S. Duane, D. Nicholas, H. Palmans, B. Schaeken, J. Sephton, P. Sharpe,
R. Thomas, M. Tomsej, D. Verellen, and S. Vynckier, “Dosimetry audit
for Tomotherapy using alanine/EPR,” Med. Phys. 33, 2093 共2006兲.
13
P. H. Sharpe, K. Rajendran, and J. P. Sephton, “Progress towards an
alanine/ESR therapy level reference dosimetry service at NPL,” Appl.
Radiat. Isot. 47, 1171–1175 共1996兲.
14
BJR, “Central axis depth dose data for use in radiotherapy,” Br. J. Radiol.,
Suppl. 25 共1996兲.
15
S. C. Sharma, J. T. Ott, and J. B. Williams, and D. Dickow, “Commissioning and acceptance testing of a CyberKnife linear accelerator,” J.
Appl. Clin. Med. Phys. 8, 119–125 共2007兲.
16
P. Francescon, S. Cora, C. Cavedon, P. Scalchi, and J. Stancanello, “CyberKnife dosimetric beam characteristics: comparison between experimental results and Monte Carlo simulation,” Robotic Radiosurgery, Vol. 1
共CyberKnife Society Press, Sunnyvale, CA, 2005兲, pp. 71–80.
17
F. Araki, “Monte Carlo study of a Cyberknife stereotactic radiosurgery
system,” Med. Phys. 33, 2955–2963 共2006兲.
18
P. Francescon, S. Cora, and C. Cavedon, “Total scatter factors of small
beams: A multidetector and Monte Carlo study,” Med. Phys. 35, 504–513
共2008兲.
12