Introduction
The clinical delivery of electron beam process
is complex than photon beam as it involves: (i) additional
devices like cones, fabrication of customized
block and differential transmission bolus and
(ii) issues related with extended SSD, additional
shielding for collimation at skin level, oblique incidence
and contour irregularity. Some of the above
issues are modeled and simulation in the Treatment
Planning System (TPS). Therefore, TPS is a crucial
component of clinical radiotherapy process.
In recent years, complexity of TPS has increased
significantly, especially with the advancement of
image based on three dimension (3-D) conformal
radiotherapy. This has led to need for a comprehensive
quality assurance (QA) guidelines. Increased
need has been paid to quality assurance of treatment
planning systems by several national and
international organizations that include Van Dyk
et. al.[] in 1993, Shaw et. al.[] in 1996, SSRPM
report[] in 1997, Fraass et. al.[] in 1998, Mayles
et. al.[] in 1999, ESTRO report[] in 2004 and NCS
report[] in 2006.
In the past, lack of complete TPS QA and
quality control of treatment machine procedures
led to some serious accidents (such as incorrect repair
of accelerator (Spain),[] accelerator software
problems (USA and Canada)[]). So, QA in the radiotherapy
treatment planning process is essential
for determination of accuracy in the radiotherapy
process and avoidance treatment errors.[]
A number of task groups[,,] over the past several
years have developed guidelines and protocols
for systematic QA of 3D radiotherapy treatment
planning systems (TPSs) that including specific
QA aspects of a TPS, such as anatomical description,
beam description, dose calculations, and data
output and transfer. Many studies have been performed
in which specific problems associated with
treatment planning and dose calculation procedures
were addressed.[-] Some studies were confined
to the performance evaluation of the vendor
specific TPS.[-]
The general need of QA of TPS in radiotherapy
has already been discussed in the literature.[,,]
Some reports[-,] have been published for help to
physicist in QA program. TRS-430 report[] that includes multiple steps is comprehensive report
of IAEA for QA. These steps are acceptance tests,
commissioning, periodic QA program and patient
specific QA. Acceptance tests perform to verify
functionality and agreement with determined specification
by manufacturer. Commissioning can be
divided into two groups that including non-dosimetric
and dosimetric tests. Non-dosimetric tests
perform to verify the functionality of the tools of
TPS. Dosimetric tests perform to verify the performance
of the dose calculation generated by the
TPS with the measured dose. Periodic QA program
perform to verify reproducibility of planning in accordance
with that established in commissioning.
Patient specific QA perform to verify the treatments
process as a whole.
A number of author as Jamema et. al.[] and Camargo
et. al.,[] Murugan et. al.,[] Kragl et. al.[]
implemented QA procedure into TPS for photon
beams with the guidance of IAEA TRS 430 report.
But there is not found article relevant to TPS QA
for electron beams with the guidance of IAEA TRS
430 at literature.
The purpose of the present study carry out application
of the IAEA TRS-430 QA procedures of
TPS for electron energies. As a result of this, the
local dose deviation values and their confidence
limit values (including systematic and random errors)
have been obtained. In addition, the trends of
the deviations found in the conducted tests have
been determined.
Methods
The commissioning procedure of IAEA TRS-
430 for clinical electron beams was implemented
for Generalized Gaussian Pencil Beam (GGPB)
algorithm of EclipseTM v7.5 TPS (Varian Medical
Systems, Palo Alto, CA, USA). The beam data
measurements of TPS have been carried out RFA-
300 3D radiation field analysis system (Wellhöfer
Dosimetrie GmbH, Schwarzenbruck, Germany)
controlled by OmniPro-Accept v6.5 software and
silicon semiconductor diode detectors (Wellhöfer
Dosimetrie GmbH, Schwarzenbruck, Germany).
Clinac DHX 2300 CD (Varian Medical Systems,
Palo Alto, CA, USA) linear accelerator is generated five electron energy beams that becomes 6 MeV,
9 MeV, 12 MeV, 16 MeV, 20 MeV (respectively
R50= 2.4 g cm-2, R50= 3.6 g cm-2, R50= 5.0 g cm-2,
R50= 6.7 g cm-2, R50= 8.4 g cm-2).
Electron Beam Commissioning
This stage including dosimetric test aimed to
compare the measurement dose and the calculated
dose of TPS. The IAEA TRS-430 tests were implemented
into electron beams of TPS. Calculation
grid size of TPS for all test was preferred 2.5 mm
because of clinically relevant general use.
The central axis percentage depth dose and
beam profile measurements were made using the
RFA-300 3D radiation field analysis system (Water
Phantom System) controlled by OmniPro-Accept
v6.5 software and EFD3G Diode. In addition, the
QA tests were applied on solid water phantom
and specially formed phantoms. The absolute
dose measurements were performed with in 0.65
cm3 FC65-G farmer type ion chamber and PPC05
parallel plane chamber connected to DOSE1 electrometer.
Film dosimetry measurements were made
using Gafchromic EBT2 films (International Speciality
Products, Wayne, New Jersey) and VIDAR
Dosimetry PRO Advantage Film Digitizer (Vidar
Systems Corporation, Hendon, Virginia).
Evaluation of Tests
For TPS QA, in principle there are two areas
with a homogenous dose, well inside or far outside
the beam. In between we have the penumbra and
build-up regions with a high dose gradient. Figure
1 show the various regions that can be defined in
terms of dose and dose gradient in a photon beam,
incident on a homogeneous phantom. Venselaar et.
al.[] have defined a set of criteria of acceptability
based on different tolerances for δ based on the
knowledge that dose calculation algorithms provide
better accuracy in some regions of the beam
than in others. At AAPM TG 53,[] Van Dyk et. al.[]
have defined such regions of different criteria of
acceptability. According to Venselaar et. al.,[] different
tolerances for δ are proposed for different
regions in the beam which can be distinguished,
analogous to the paper of Van Dyk et. al.[] and the
report of AAPM TG 53.[] According to report of NCS,[] different tolerances are proposed for the
various regions in an electron beam shown in Figure
2, such as δ1, δ2, δ3, δ4, δR85 and RW50. These
include the following:[]
Fig 1: Definition of different regions in a radiation beam,
based on the magnitude of the dose and dose gradient
(Adapted, from ESTRO report[]).
Fig 2: Different tolerances are proposed for the various regions
in a electron beam; (a) depth-dose curve: (b)
beam profile (Adapted, from ESTRO report[] and
NCS report[]).
• δ1: for points on the central beam axis between
a depth of 2 mm and R95, with dose gradients less
than 3% per mm (i.e. excluding the surface dose
points up to a depth of 2 mm): the high dose and
small dose gradient region.
• δ2: for points in regions with a high dose gradient,
such as on the central beam axis between R95
and R10, the penumbra, regions close to interfaces
of inhomogeneities: the high dose and large dose
gradient regions. The dose gradient is in general
larger than 3% per mm. The tolerance criterion is
preferably expressed as a shift of isodose lines (in
mm).
• δ3: for points with a high dose but off the central
beam axis and points describing the surface
dose: this region is also a high dose and small dose
gradient region.
• δ4: for points outside the geometrical beam
edges; this region is a low dose and small dose gradient
region, for instance below 7% of the central
beam axis normalization dose.
• δRW50: for deviations in the radiological width,
defined as the width of a profile measured at the
50% points.
•δR85 and δRp: for deviations in the therapeutical
range and the practical range of the electron beam,
respectively.
TPS performance was investigated the difference
between calculated and measured dose values
as a percentage of the dose measured locally.
Deviations between results of calculations and
measurements can be expressed as a percentage
deviation of the local dose according to Venselaar
et. al.,[]
δ = 100% × (Dcal - Dmeas) / Dmeas (1)
where Dcal and Dmeas are calculated dose at
particular point in the phantom and measured dose
at same point in the phantom, respectively. In low
dose regions where the points were outside the
penumbra or under a block, an alternative comparison
accordingly to Venselaar et. al.,[]
δ = 100% × (Dcal - Dmeas) / Dmeas,cax (2)
where Dmeas,cax is dose measured at a point at
same depth on the central axis of the open beam.
The deviations, δ, described above refer to comparisons
of individual calculated and measured
points. Although this is not strictly correct. Because
a study consisting of many points is evaluated,
some of these points may exceed or may not
the tolerance.
If a study consisting of many points is evaluated,
in this case some statistical assessment can be
performed on the calculation points and the measurement
points. For this purpose, the concept of
confidence limit was defined by Venselaar et. al.[]
Accordingly, confidence limit, Δ, as follow,
Δ = | average deviation | + 1.5×SD (3)
where SD is the standard deviation. According
to complexity of geometry, the tolerance as defined
in Table 2 can be applied to the confidence limit
rather than to individual points. At equation (3), the
factor 1.5 is chosen rather arbitrarily, but Venselaar
et. al.[] and Welleweerd et. al.[] showed to be
useful for this purpose in clinical practice. If a factor
greater than 1.5 was used in equation (3), this
would have emphasized the random errors, while a
factor smaller than 1.5 would increase the relative
importance of systematic deviations.[]
All tests of Electron Beam Commissioning
were simulated in the TPS and the performed calculations
were compared against that measured
on the treatment unit. As a result of this, the local
dose deviation values and their confidence limit
values (including systematic and random errors)
have been obtained. In addition, the trends of the
deviations found in the conducted tests have been
determined.
Results
Electron Beam Commissioning
Electron beam commissioning tests were given
in Table 1 and those tests were applied to confirm
the performance and limitations of systems. Results
of implementation were given in Table 3 in
detail. At Table 3, results were given separately for
each energy and confidence limits of individual
measurements type (%DD, profile, point dose) in
detail.
Table 1: Detail of dosimetric tests performed on TPS in the present study
Table 2: According to complexity of geometry, proposed values of the tolerance for percentage deviation of dose at different local
(Adapted, ESTRO report[] and NCS report[])
Table 3: Results of electron beam commissioning performed on TPS
While all %DD is used to calculate of confidence
limit value, confidence limit values of profiles
is separated two groups that is including all
profiles (Δall) and including profiles without Rp
depth (Δwithout Rp). Because all values of profiles of
Rp depth include high SD and this value causes to
increase Δ value.
Many results for square field test were satisfactory
found. At depth dose, the confidence limit values
of δ1 and δ3 was found outside tolerances for
low energies. For profiles, Δwithout Rp are found
within tolerances but Δall are found outside tolerances
because of high value of Rp depths. A no table point, all confidence limit value of profiles
was detected that it was increased depending on
expanding depth. But each confidence limit value
of profiles was found to show different change
depending on expanding field size. For example; while δ3 value of profiles increased depending on
expanding field size, δ2 value decreased depending
on expanding field size.
As results of shaped field test, it was found to
same results of the square field test. Results of slab bolus test were found within the tolerance limits
given in the ESTRO report[] and NCS report.[]
Results of CT based inhomogeneity corrections
and complex surface shapes were found outside
tolerance limits given in the ESTRO report[] and
NCS report,[] especially δ3 and shown in Figure 3.
At CT based inhomogeneity corrections test, maximum
deviations were found 7.9% for 16 MeV and
4.2% for 20 MeV electron beam.
Fig 3: Comparison between measured and calculated dose
profiles for 16 MeV electron beam (CT based Inhomogeneity
corrections tests With Gafchromic EBT2
film).
Discussion
In this study, we had commissioned Varian
EclipseTM v7.5 TPS in accordance with procedure
of IAEA TRS-430 for clinical electron beams. Result
of commissioning was investigated and the
trends of the deviations found in the tests conducted
have been determined. All confidence limit value
of profiles was detected that it was increased depending
on expanding depth. But each confidence
limit value of profiles was found to show different
change depending on expanding field size.
According to results of CT based inhomogeneity
corrections tests, values of δ3 were found outside
tolerance limits given in the ESTRO report[6]
and NCS report.[] Deviations between measurement
and TPS calculation has defined by technical
specifications of VARIAN Eclips GGPB algorithm.
According to these technical specifications,
deviation values can find about 2% for homogenous
media and 5% for non-homogenous media.
There isn't found article about apply QA procedure into TPS for electron beams with the guidance
of IAEA TRS 430 and other guidelines at literature.
According to Hogstrom et. al.,[] despite the
significant progress in calculating dose, treatmentplanning
systems currently fail the practice of radiation
therapy and the treatment of patients with
electron beam therapy by being unable to model
actual treatments. Treatment-planning tools, such
as skin collimation, internal collimation and bolus,
are modelled inadequately or not at all.
At study relevant comparison of electron beam
dose calculation of pencil beam and Monte Carlo
algorithm by Ding et. al.,[] the comparison has
demonstrated some serious limitations of the pencil
beam algorithm implemented in CADPLAN to
accurately predict hot and cold spots for 3D inhomogeneous
phantoms. The pencil beam model is
unable to predict sharp high- and low- dose variations
(10%) for simple 3D inhomogeneities and a
complex 3D inhomogeneous phantom consisting
of overlying both low-(air) and high-(bone) density
materials, even when the calculation resolution
is much smaller than the size of high- and low-dose
regions. The Monte Carlo results generally have
much better agreement with measurements, especially
in predicting sharp increases or decreases in
absorbed dose caused by the perturbation of adjacent
3D inhomogeneities.[]
Generally, there are differences between measurements
and calculations. It should not be forgotten
that the factors affecting discrepancies between
measurement and calculation include;
i. TPS beam data input,
ii. Beam model fitting,
iii. Dose calculation algorithm,
iv. The computation of the number of MUs,
v. Verification measurement set-up.
In case the TPS fails to meet these accuracy requirements,
NCS report[] suggests the following:
i. Check the basic beam data entered in the TPS
and the test beam data set.
ii. Adjust the model parameters.
iii. Restrict the clinical use of the TPS to geometries
that passed the test.
iv. Inform the vendor about the findings.
According to this study, it does not need application
to above suggestions for our EclipseTM TPS.
Only the errors found in our study for the parameters
used in patient treatment planning has to be
considered.
Conclusion
At commissioning of EclipseTM TPS, it has been
observed that the conducted test is generally within
tolerance and is outside of tolerances in some cases.
In addition the trends of the deviations found
in the conducted tests have been determined. Only
the errors found in this study for the parameters
used in patient treatment planning has to be considered.
This procedure must perform entirely after
upgrade of TPS.
This study has ensured the correctness of the
beam data entered in the TPS during the commissioning.
With commissioning tests, it was identified
as a baseline data for an ongoing QA program.
Acknowledgements
This study was supported by the Scientific Research
Project Fund of Erciyes University (Project
Code: TSY-09-1047).
References
Van Dyk J, Barnett RB, Cygler JE, Shragge PC.
Commissioning and quality assurance of treatment
planning computers. Int J Radiat Oncol Biol Phys
1993;26(2):261-73. CrossRef
Shaw JE. A guide to commissioning and quality control
of treatment planning systems. Report No. 68: IPEMB,
York; 1996.
Swiss Society for Radiobiology and Medical Physics.
Report 7: Quality control of treatment planning systems
for teletherapy: SSRMP; 1997.
Fraass B, Doppke K, Hunt M, Kutcher G, Starkschall
G, Stern R, et al. American Association of Physicists
in Medicine Radiation Therapy Committee Task Group
53: quality assurance for clinical radiotherapy treatment
planning. Med Phys 1998;25(10):1773-829. CrossRef
Mayles WPM, Lake R, McKenzie A, Macauly EM,
Morgan HM, Jordan TJ, et al. Physics aspects of quality
control in radiotherapy. IPEM Report No.81.York: IPEMB; 1999.
European Society of Therapeutic Radiation Oncology.
Quality assurance of treatment planning systems, practical
examples for non-IMRT photon beams: ESTRO;
2004.
Netherlands Commission on Radiation Dosimetry.
Quality assurance of 3-D treatment planning systems
for external photon and electron beams: Practical
guidelines for initial verification and periodic quality
control of radiation therapy treatment planning systems:
NCS, Delft; 2006.
Leveson NG, Turner CS. An investigation of the Therac-
25 accidents. IEEE Computer 1993;26:18-41. CrossRef
Spanish Society of Medical Physics, The accident of
the linear accelerator in the “Hospital Clínico de Zaragoza”.
Spanish Society of Medical Physics, 1991.
International Atomic Energy Agency. Technical Report
Series 430 Commissioning and quality assurance of
computerized planning systems for radiation treatment
of cancer. IAEA, Vienna; 2005.
Nisbet A, Beange I, Vollmar HS, Irvine C, Morgan A,
Thwaites DI. Dosimetric verification of a commercial
collapsed cone algorithm in simulated clinical situations.
Radiother Oncol 2004;73(1):79-88. CrossRef
Cheng CW, Das IJ, Tang W, Chang S, Tsai JS, Ceberg
C, et al. Dosimetric comparison of treatment planning
systems in irradiation of breast with tangential fields.
Int J Radiat Oncol Biol Phys 1997;38(4):835-42. CrossRef
Davidson SE, Ibbott GS, Prado KL, Dong L, Liao Z,
Followill DS. Accuracy of two heterogeneity dose
calculation algorithms for IMRT in treatment plans
designed using an anthropomorphic thorax phantom.
Med Phys 2007;34(5):1850-7. CrossRef
Paelinck L, Reynaert N, Thierens H, De Neve W, De
Wagter C. Experimental verification of lung dose with
radiochromic film: comparison with Monte Carlo simulations
and commercially available treatment planning
systems. Phys Med Biol 2005;50(9):2055-69. CrossRef
Miften M, Wiesmeyer M, Kapur A, Ma CM. Comparison
of RTP dose distributions in heterogeneous phantoms
with the BEAM Monte Carlo simulation system.
J Appl Clin Med Phys 2001;2(1):21-31. CrossRef
Van Esch A, Tillikainen L, Pyykkonen J, Tenhunen M,
Helminen H, Siljamäki S, et al. Testing of the analytical
anisotropic algorithm for photon dose calculation. Med
Phys. 2006;33(11):4130-48. CrossRef
Knöös T, Ceberg C, Weber L, Nilsson P. The dosimetric
verification of a pencil beam based treatment planning
system. Phys Med Biol 1994;39(10):1609-28. CrossRef
Aspradakis MM, Morrison RH, Richmond ND, Steele
A. Experimental verification of convolution/superposition
photon dose calculations for radiotherapy treatment planning. Phys Med Biol 2003;48(17):2873-93.
Jamema SV, Upreti RR, Sharma S, Deshpande DD.
Commissioning and comprehensive quality assurance
of commercial 3D treatment planning system using
IAEA Technical Report Series-430. Australas Phys Eng
Sci Med 2008;31(3):207-15. CrossRef
Camargo PR, Rodrigues LN, Furnari L, Rubo RA.
Implementation of a quality assurance program for
computerized treatment planning systems. Med Phys
2007;34(7):2827-36. CrossRef
Murugan A, Valas XS, Thayalan K, Ramasubramanian
V. Dosimetric evaluation of a three-dimensional treatment
planning system. J Med Phys 2011;36(1):15-21.
Kragl G, Albrich D, Georg D. Radiation therapy with
unflattened photon beams: dosimetric accuracy of advanced
dose calculation algorithms. Radiother Oncol 2011;100(3):417-23. CrossRef
Venselaar J, Welleweerd H, Mijnheer B. Tolerances
for the accuracy of photon beam dose calculations
of treatment planning systems. Radiother Oncol
2001;60(2):191-201. CrossRef
Welleweerd J, van der Zee W. Dose calculations for
asymmetric fields using Plato version 2.01. In: Proc.
17th Annual ESTRO Meeting. Radiother Oncol 1998;48. p. 134.
Hogstrom KR, Almond PR. Review of electron beam
therapy physics. Phys Med Biol 2006;51(13):R455-89.
Ding GX, Cygler JE, Yu CW, Kalach NI, Daskalov G.
A comparison of electron beam dose calculation accuracy
between treatment planning systems using either a
pencil beam or a Monte Carlo algorithm. Int J Radiat
Oncol Biol Phys 2005;63(2):622-33. CrossRef
