Principal knowledge for commissioning of Linac using

FFF

Francisco J. Hernández Flores

T
franciscohernandez [email protected]

AF
MMP Student 2015 - 2016

March 3, 2016
DR

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 1 / 22
1 Purpose 6 Penumbra

T
2 Introduction 7 Slope
Aspect of commissioning 8 The peak Position

AF
3
4 Profile normalization 9 Conclusion
5 Dosimetric field size 10 References
DR

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 2 / 22
Purpose

To investigate dosimetric characteristics of a new linear accelerator designed to

deliver flattened, as well as flattening filter-free (FFF), beams. To evaluate the
accuracy of beam modeling under physical conditions.[4]

T
AF
DR

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 3 / 22
Purpose

To investigate dosimetric characteristics of a new linear accelerator designed to

deliver flattened, as well as flattening filter-free (FFF), beams. To evaluate the
accuracy of beam modeling under physical conditions.[4]

T
To kow the possible definitions and suggestions for some dosimetric parameters for

AF
use in quality assurance of FFF beams generated by medical linacs in radiation
therapy. DR

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 3 / 22
Purpose

To investigate dosimetric characteristics of a new linear accelerator designed to

deliver flattened, as well as flattening filter-free (FFF), beams. To evaluate the
accuracy of beam modeling under physical conditions.[4]

T
To kow the possible definitions and suggestions for some dosimetric parameters for

AF
use in quality assurance of FFF beams generated by medical linacs in radiation
therapy. DR
To compare the dosimetric accuracy of advanced dose calculation algorithms for
flattened (FF) and unflattened (FFF) photon beams.[5]

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 3 / 22
Purpose

To investigate dosimetric characteristics of a new linear accelerator designed to

deliver flattened, as well as flattening filter-free (FFF), beams. To evaluate the
accuracy of beam modeling under physical conditions.[4]

T
To kow the possible definitions and suggestions for some dosimetric parameters for

AF
use in quality assurance of FFF beams generated by medical linacs in radiation
therapy. DR
To compare the dosimetric accuracy of advanced dose calculation algorithms for
flattened (FF) and unflattened (FFF) photon beams.[5]

We must to know the Dose calculation accuracy using Flattening filter free (FFF)
in Advanced treatment techniques, such as IMRT, VMAT and SBRT.

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 3 / 22
Introduction

In recent years, the clinical use of flattening filter free (FFF) beams is growing fast.

T
Among the reasons, the very high dose rate achieved (up to four times the dose rate of
the standard flattened (FF) beams plays a decisive role.

AF
This allowed for stereotactic radiotherapy deliveries of very high dose per fraction (as 20
to 25 Gy) in very short treatment times, comparable with the conventional fractionation
time slots.
DR
FFF beams have been extensively investigated and characterized before their introduction
in the clinical practice.[1]

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 4 / 22
Introduction

Flattening Filter
Conventional medical linear accelerators delivering photon beams are equipped

T
with a flattening filter (FF) in order to allow delivery of homogeneous dose
distributions with broad beams.

AF
The differences between FFF and FF in terms of quality assurance is mainly
related to beam dosimetry, and not to mechanical characteristics of the linear
accelerator, for which the standard quality assurance procedures still hold.
DR
Flattening Filter contribute to scattered, reduce dose rate, leakage from the
treatment head, beam hardening and also neutron fluence for high energy of X ray
used in Linac.[2]

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 5 / 22
Flattening Filter free benefits
Increase the dose rate and reduce treatment time small treatment time less patient
movement.

T
Reduce leakage from the treatment head, they have 50% to 60% reduced
collimator and treatment head scatter.

AF
reduced ”out of field” dose obserbed to be less than 10% at 2 cm for a 6MeV
beams FFF, up to 20% reduced neutron contamination for 18MeV Beams [2]
DR

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 6 / 22
Flattening Filter free benefits
Increase the dose rate and reduce treatment time small treatment time less patient
movement.

T
Reduce leakage from the treatment head, they have 50% to 60% reduced
collimator and treatment head scatter.

AF
reduced ”out of field” dose obserbed to be less than 10% at 2 cm for a 6MeV
beams FFF, up to 20% reduced neutron contamination for 18MeV Beams [2]
DR
Flattening Filter free Problem
Ion chamber and EPID saturation.
Inter leaf leakage, very high dose per pulse,
FFF can deposit dose of 1 Gy in 2.5 second inadvertent dose to critical structures
can be dangerous in extremely short time so therapist and patient must be
educated (24 Gy/min)

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 6 / 22
Quality Assurance of FFF beams

Flattening Filter Free (FFF)

FFF beams are used in the linac without FF in place of carousel. FFF delivered with

T
conventional medical linear accelerator have the conical flattening filter removed and
replaced by a thin foil.[2]

AF
This foil is introduced for two reason:
For safety. It will stop the electron beam reaching the patient if the target
collapses.
DR
Producing enough signal in the ion chamber by producing electrons.[1]

The main advantages of removing the flattening filter are an increased dose rate,
reduced scatter, reduced leakage and reduced out-of-field doses.[1]

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 7 / 22
Spectrum of two energy using FF and FFF

T
AF
DR
(a) spectrum of 6 MeV (b) spectrum of 10 MeV
Figure: Photon spectra for 6 MV FFF and 6 MV (left), and 10 MV FFF and 10
MV (Right).

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 8 / 22
Profile normalization

The inflection point

Pnisch et al [3] suggested the use of the inflection point at the field edge to
renormalize a FFF beam to the same dose level of a FF beam. From this

T
renormalized profile it is then possible to evaluate penumbra and the field size.
The correct evaluation of the inflection point position is critical, being located by

AF
definition at the point of the highest dose gradient. [1]

The re-normalization value

DR
The use of a renormalization factor, compared to the inflection point procedure,
allows for a location of the normalization point in a less critical position, at the
profile shoulder, where the FF and the corresponding FFF profiles start to differ,
and it is located at the second maximum point of the third derivative of the profile.
a + b ∗ FS + c ∗ depth
Renorm Factor = (1)
1 + d ∗ FS + e ∗ depth

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 9 / 22
Profile normalization
Based on the fact that FFF beams deliver higher dose to the central axis, FFF and FF
beams should be mutually renormalized to superimpose the profile fall-off. Two methods
can be followed: the inflection point or the renormalization value. Both methods hold
only for symmetric beams.

T
AF
DR
(a) Renormalization Point (b) Infection Point
Figure: Renormalization point obtained through the profile third derivative [1]
Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara
Radiation
Italy)Therapy March 3, 2016 10 / 22
Dosimetric field size

T
Once the FFF beams are renormalized as above, the concept of dosimetric field size as
the distance between the 50% dose levels can be used for FFF beams, as for FF beams

AF
[generally the full width half maximum (FWHM) is used for standard FF beams normalized
to 100% at central beam axis].
DR
Alternatively, as suggested by Pnisch et al.[3] the distance between the left and right
inflection points could be used

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 11 / 22
Penumbra
For conventional flattened-beam profiles, the definition of the penumbra is based on the
80-20% dose values. This is not applicable in the unflattened case. Therefore, the
penumbra of the unflattened profile were derived from the spatial distance between the
positions where the doses were 20 and 80 % of the normalized dose Dn

T
Du
Dn = ∗ DCAX (2)
Df

AF
DR
Figure: Normalization of an unflattened profile of a measured 6-MV photon
beam. [3]
Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara
Radiation
Italy)Therapy March 3, 2016 12 / 22
Flattened region and field region
The flatness is based on the flattened region definition, and should be applied to a ”field
region” in a way that it could be used for both beam modalities. Once renormalized as

T
above, the ”field region” can be defined as the region within a certain defined percentage
of the field. The percentage could be the same for all field sizes, or it can be changed.

AF
Flatness and Unflatness
Unflatness is the parameter relative to FFF beams corresponding to flatness for FF beams.
DR
Unflatness can be defined as the ratio between the dose level at the beam central axis and
the dose level at a predefined distance from the central axis as a function of field size, or
at the edge of the field region.
Dosecentral axis
Unflatness = (3)
Doseoff axis

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 13 / 22
Slope
The peak shape of the FFF profile can be defined by the slope parameter describing the
left and right inclinations of the profiles. Because the FFF profile depends on the energy,
with different shapes in terms of concavity or convexity of the slopes.
(x1 − x2 ) ∗ (y1 − y2 )
Slope = (4)

T
(x1 − x2 )2

AF
DR

Figure: slope of profile FFF

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara
Radiation
Italy)Therapy March 3, 2016 14 / 22
The peak Position

The peak of the FFF profile is the indication of the forward direction of the beam.

T
Intuitively this peak should be located on the beam central axis. The peak position
parameter is defined as the off-axis position of the intersection point of the left

AF
and right slopes, as follows:
IL − IR
peakposition = (5)
SR − Sl
DR
where IL and IR are the left and right intercepts and SL and SR are the left and
right slopes,
(x1 − x2 ) ∗ (y1 − y2 )
I = y2 − x2 ∗ (6)
(x1 − x2 )2

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 15 / 22
Symmetry

Symmetry, as a parameter checking the equality level between left and right sides of a
profile, can be defined as usual for standard FF beams, with the only difference that the
evaluation area should be within the field region for FFF beams instead of the flattened

T
region commonly used in FF beams.[1]

Dx

AF
The maximum dose ratio : (7)
D−x max

The maximum Variation : (Dx − D−x )max (8)

DR
where Dx and D−x are the doses at x and -x positions (symmetric relative to central axis).

LeftIntegral − RightIntegral
The area ratio :
(9)
LeftIntegral + RightIntegral
where LeftIntegral and RightIntegral are the areas bounded by the profile on the left and
right of the beam central axis.

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 16 / 22
Energy spectrum and quality index

T
FFF beams present an energy spectrum significantly different from FF beams since
the thick conical shaped attenuator is removed.[1]

AF
Despite the differences in the FFF spectrum with respect to the corresponding FF
beam, there is no reason to change quality index definitions that can be.[1]
D20 cm
DR QI = 1.2661 ∗ − 0.0595. (10)
D10 cm

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 17 / 22
Surface Dose
Due to different electron contamination and lower photon energy spectrum, surface doses
of FFF are expected to be different from FF beams.18 The surface dose parameter Ds is
defined here as the relative dose at d = 0.5 mm with respect to the dose at dmax .Due to

T
different electron contamination and lower photon energy spectrum, surface doses of FFF
are expected to be different from FF beams. The surface dose parameter Ds is defined

AF
here as the relative dose at d = 0.5 mm with respect to the dose at dmax.[1]

Output Factor
DR
The head scatter component of a FFF beam relative to the corresponding FF
beam is markedly different. Variation in output factors is then less pronounced for
FFF beams due to the head scatter component.[1]
Output factor definitions are kept identical for both FFF and FF beams.[1]
In both setup conditions the output factors of FFF fields are less spread, in
particular for in air evaluation, confirming the lower head scatter component for
such fields.[1]

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 18 / 22
Dose Rate
In FFF beams the dose rate increase in two to four times higher than standard
beams.
The common check on dose rate dependence has to be performed on the entire

T
dose rate range, keeping the consolidated experience in use for FF beams.
For FFF beams particular attention has to be paid in the dosimetry system choice:

AF
the collection efficiency of ionization chambers, the possible saturation are just
examples to consider for correct measurements.
DR

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 19 / 22
Dose Rate
In FFF beams the dose rate increase in two to four times higher than standard
beams.
The common check on dose rate dependence has to be performed on the entire

T
dose rate range, keeping the consolidated experience in use for FF beams.
For FFF beams particular attention has to be paid in the dosimetry system choice:

AF
the collection efficiency of ionization chambers, the possible saturation are just
examples to consider for correct measurements.
DR
Absolute dose calibration
Absolute calibration of the beam output shall follow dedicated protocols.
There is no reason to change the reference conditions for calibration, but there is a
need for a re-evaluation of the beam quality factor (kQ ) values for FFF beams in
relation to beam quality indices, as they are not listed as clinical used beams.
To note that the recombination factor ks changes slightly between FFF and FF
beams, but this difference is systematic.

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 19 / 22
Conclusion

Removing the flattening filter improved the characteristics of the accelerator in terms
of smaller penumbras especially for the 18-MV mode, reduced MLC leakage, and less
variation in the total scatter factors.[3]

T
AF
Although the FFF beams provide much high dose rate at the treatment target, the ion
recombination effect of the Farmer, PinPoint, and plane-parallel chamber in the FFF
photons is not significantly different from the flattened photons. These ion chambers are
suitable in the quality assurance and exposure measurement for the FFF beams regarding
DR
their negligible ion recombination and sufficient collection efficiency.

We have presented ideas regarding the quality controls (QC) that have to be considered
during the establishment of a quality assurance program (QA) when introducing FFF
beams into a clinical setting.

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 20 / 22
References

A.Fogliata et al, Definition of parameters for quality assurance of flattening

filter free (FFF) photon beams in radiation therapy, Oncology Institute of Southern
Switzerland, 3 October 2012

T
Dhuruvan Viswanathan, flattening Filter Free LINAC, RPD 331 Hales , 17
August 2014

AF
Falk Pnisch, Properties of unflattened photon beams shaped by a multileaf
collimator, Houston Texas 77030, 08 April 2006

JAN HRBACEK, Commissioning of the photon beams of flattening filter-free

DR
linear accelerator and the accuracy of beam modeling using an anisotropic
analytical algorithm, Department of Radiation Oncology, University Hospital
Zurich, Zurich, Switzerland, Elsevier, 2011.

Gabriele Kragl et al, Radiation therapy with unflattened photon beams:

Dosimetric accuracy of advanced dose calculation algorithms,Article from
Department of Radiotherapy Medical University of Vienna, 23 July 2011

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 21 / 22
on
nti
a tte
d
kin

T
r
ou

AF
y
for
n ks
DR
y tha
a n
M

Francisco J. Hernández Flores (Arcispedale Santa Anna, Ferrara

Radiation
Italy)Therapy March 3, 2016 22 / 22