AAPM
th
49
Annual Meeting
CE - Therapy
PROTON THERAPY
Alfred Smith, Ph.D.
M. D. Anderson Cancer Center
Houston, TX
Harald Paganetti, Ph.D.
Massachusetts General Hospital,
Boston, MA
Topics Covered
• History of Proton Therapy
• Worldwide Facilities
• Rationale for Proton Therapy
• Physics of Proton Beams
• Treatment Delivery Techniques
• Proton Treatment Technology
• Clinical Commissioning
• Treatment Planning
History and Current Status
of
Proton Therapy
Abbreviated History of Protons
•
•
•
•
•
•
1919
Rutherford proposed existence of protons
1930
E. O. Lawrence built first cyclotron
1946
Robert Wilson proposed proton therapy
1955
Tobias et al treated patients at LBL
1961
Kjellberg et al treated patients at HCL
1972
MGH received first NCI grant for proton
studies at HCL
• 1991
• 2006
First hospital-based proton facility at LLUMC
28 facilities worldwide treating patients;
over 55,000 patients treated with protons.
Charged Particle Therapy Facilities
25 facilities worldwide are treating patients.
• 6 in Japan (4 hospital-based)
• Chiba, NCC East-Kashiwa, HIBMC-Hyogo, PMRCTsukuba, WERC-Wakasa, Shizuoka Cancer Center
• 6 in the United States ( 4 hospital-based)
LLUMC, MGH, MDACC, Univ. of Florida, Univ. of
Indiana, UC Davis
• 9 in Europe/Russia
• 4 Additional facilities (UK, China, Korea, South Africa)
• 15 additional institutions worldwide are either building a
new facility or are seriously planning to have a particle
therapy facility.
Particle Therapy Facilities Worldwide
In operation
In preparation
Considering
Treating: Protons 25, Carbon ions 3
Courtesy of Takashi Ogino, MD, PhD, NCC Kashiwa, Japan
Rationale for Proton Therapy
Need for Improved Local Control
in Cancer Treatment (selected sites)
(all numbers are estimates)
Tumor Site
Head/Neck
Gastrointestinal
Gynecologic
Genitourinary
Lung
Breast
Lymphoma
Skin, Bone, Soft Tissue
Brain
Total
Deaths/
year
Deaths due to
Local Failure
22,000
135,000
28,000
55,000
160,000
41,000
20,000
15,000
12,000
488,000
13,200 (60%)
54,000 (40%)
14,000 (50%)
27,500 (50%)
40,000 (25%)
4,920 (12%)
2,400 (12%)
5,000 (33%)
10,800 (90%)
171,820 (35%)
Over 1,350,000 new cancer patients per year in the US
Photons
Fundamental Things
to Remember about
Protons
Protons
Photons
•
Protons Stop!
•
Photons don’t stop.
•
Proton dose at depth
(target) is greater than
dose at surface.
•
Photon dose at depth
(target) is less than
Ideal dose
distribution
Protons
dose at dmax.
Rhabdomyosarcoma of Paranasal Sinus (7 y old boy)
6 MV
Photons
(3 field)
160 MeV
Protons
(2 field)
Photon
IMXT
(9 field)
Proton
IMRT
(9 field)
Proton Physics
• The physics of proton beams
• Passive scattering systems
• Pencil beam scanning systems
Electromagnetic energy loss of protons
p
p
e
120%
100%
Mass Electronic Stopping Power is the
mean energy lost by protons in
electronic collisions in traversing the
distance dx in a material of density ρ.
Relative Dose [%]
80%
60%
S/ρ = 1/ρ[dE/dx] ∝ 1/v2
40%
20%
Where v = proton velocity
0%
0
5
10
15
Depth in Water[cm]
20
25
30
Normalized (at peak) Bragg Curves for Various Proton Incident Energies
Range Straggling will cause the Bragg peak to widen with depth of penetration
1.2
Relative Dose
1
0.8
100 MeV
130 MeV
160 MeV
180 MeV
200 MeV
225 MeV
250 MeV
0.6
0.4
0.2
0
0
5
10
15
20
25
30
35
40
Depth in Water (cm)
Normalized (at entrance) Bragg Curves for Various Proton Incident Energies
50
Relative Dose
40
100 MeV
130 MeV
30
160 MeV
20
180 MeV
200 MeV
10
225 MeV
250 MeV
0
-10
0
10
20
30
Depth in Water (cm)
40
50
Dose depositions in water from 160 MeV protons.
Beam slit delimiters with width W cm.
Depth forUniform
160particle
MeVdistributions.
Protons for
200 MeV Beam - PDD vs Aperture size
F2 data - snout at 45 cm
various field sizes
120
30
100
W = 0.1 cm
W = 0.16 cm
W = 0.24 cm
W = 0.5 cm
W = 1 cm
W = 2 cm
20
80
2 cm x 2 cm
PDD
D o se (M eV /cm )
25
15
5 cm x 5 cm
60
10 cm x 10 cm
40
18 cm x 18 cm
10
20
5
0
0
0
0
5
10
15
50
100
150
200
250
Depth (mm)
Depth (cm)
Loss of in-scattering (charged particle equilibrium) results
in deterioration of Bragg peak and non uniformity of SOBP.
George Ciangaru
Narayan Sahoo
p’
Multiple Coulomb Scattering
θ
p
• Protons are deflected frequently in the electric field of the nuclei
• Beam broadening can be approximated by a Gaussian distribution
Lateral dose fall-off: Protons vs. Photons
80/20 Penumbra Comparison
Protons
80 - 20 % D istan ce (cm )
1.00
0.75
15 MV Photons
≈ 17 cm
0.50
0.25
0.00
15 cm
20 cm
25 cm
Norm alization Depth
Paganetti
Urie
Nuclear interactions of protons
p’
p’
p
p
γ, n
• A certain fraction of protons undergo nuclear
interactions, mainly on 16O
• Nuclear interactions lead to secondary particles
and thus to local and non-local dose deposition
(neutrons!)
• In passive scattering systems neutrons are
produced in the first and second scatterers,
modulation wheel, aperture, range compensator
in addition to those produced in the patient.
Pedroni et al PMB, 50, 541-561, 2005
Effect on the lateral dose
distribution
Primary
fluence
Secondaries from
nuclear interactions
Relative Dose [%]
Effect on Depth Dose
110
100
90
80
70
60
50
40
30
20
10
0
230 MeV protons
Total Absorbed Dose
'Primary' Dose
0
5
10
'Secondary' Dose
15
20
Depth in Water [cm]
25
30
35
Proton Therapy
Beam Delivery Technology
Physics of the Passive Scattering
Mode of Proton Beam Delivery
Goitein, Lomax, Pedroni - Med. Phys.
Passive Scattering Nozzle with Range
Modulation Wheel
How a Spread Out Bragg Peak (SOBP) is formed.
•
Modulation wheel rotates in the
beam.
•
Pull-back (energy shift)
determined by height of step.
•
Weight determined by width of
step.
•
Multiple SOBPs can be obtained
by gating beam.
Deficiencies of Proton Passive
Scattering Techniques
• Uniform SOBP - excess
normal tissue dose.
• Requires custom aperture
and compensator
• Inefficient - high proton
loss produces activation
and neutron production.
Chu, Ludewigt, Renner - Rev. Sci. Instrum.
The Pencil Beam Scanning Mode
of Proton Beam Delivery
Goitein, Lomax, Pedroni - Med. Phys.
Pencil Beam Scanning Nozzle
Beam
3.2m
Profile
Monitor
Scanning
Magnets
Ceramic
Ceramic
Vacuum
Chamber
Chamber
Spot Position
Monitor
Dose Monitor 1, 2
Iso
Center
Energy Filter
Energy Absorber
Performance
Range
4 – 36 g/cm2
Adjustability
0.1 g/cm2
Max. field size
30 x 30 cm
Beam size in air
6 – 10mm σ
SAD
> 2.5 m
Dose compliance
+/- 3% (2 σ)
Irradiation time < 2 min to
deliver 2 Gy to 1 liter
Proton Accelerators
Isocentric Gantries
Typical Facility
Hitachi 250 MeV synchrotron ring
7 MeV Linac injector
Typical Accelerators
used in proton
therapy facilities
ACCEL Superconducting Cyclotron
IBA 230 MeV Cyclotron
250 MeV
M. D. Anderson Gantry
Proton and
Carbon Ion
Gantries
190 tons
Hitachi
600 tons
Siemens
120 tons
Heidelberg
Proton Therapy Center - Houston
PTC-H
3 Rotating Gantries
1 Fixed Port
1 Eye Port
1 Experimental Port
Pencil Beam Scanning Port
Passive Scattering Port
Experimental Port
Accelerator System
(slow cycle synchrotron)
Large Fixed Port
Eye Port
Clinical Commissioning
• Tests for system functionality and safety
• Treatment Planning System commissioning
– Collection of data for input to planning system
– Validation of planning system output
• Beam calibration
– Calibration of transmission ionization chambers
– Measurement of dependence of dose on range
modulation, field size, etc.
• Patient treatment and machine QA
Pristine Bragg Peaks
250 MeV
G2_250MeV_RMW88_range25.0 cm_largeSnout@5cm
PTCH-G2, Pristine Bragg Peaks
120.0
140MeV_Range 10 cm
120
120MeV_Range 6.4 cm
100MeV_Range 4.3 cm
100.0
100
250MeV_Range 28.5 cm
225MeV_Range 23.6 cm
200MeV_Range 19.0 cm
80.0
80
60
SOBP 4 cm, Measured 3.9 cm
SOBP 6 cm, Measured 5.9 cm
SOBP 8 cm, Measured 7.9 cm
SOBP 10 cm, Measured 10.0 cm
SOBP 12 cm, Measured 12.2 cm
SOBP 14 cm, Measured 14.3 cm
SOBP 16 cm, Measured 16.6 cm
40
40.0
20
20.0
0
0.0
0
50
100
150
200
250
300
0
350
50
100
150
200
250
300
350
Depth (mm)
Depth (mm)
120 MeV
G2_160MeV_RMW
76_range13.0cm_mediumsnout@5cm
160 MeV
G2- NZL AT2.2 Modula t ion Te st( A- 3) UF10E120 R a nge 6.9[ c mH2O]
120
110%
100%
100
90%
80
80%
PDD
PDD
160MeV_Range 13 cm
60.0
PDD
180MeV_Range 16.1 cm
SOBP10cm,Measured10.6cm
SOBP8cm,Measured8.2cm
SOBP6cm,Measured6.1cm
SOBP4cm,Measured4.0cm
60
40
20
0
0
20
40
60
80
100
70%
60%
50%
40%
SOBP4cm ? 3.8cm T0521R
SOBP3cm ? 2.9cm T0522L+M+P
20%SOBP2cm ? 1.9cm T0521T
30%
SOBP1cm ? 1cm T0521U
120
140
160
180
10%
0%
PTCOG 46
Depth(mm)
0
10
Educational Workshop
20
30
40
50
De pt h in Wa t e r [ mm H2O]
60
70
80
90
100
The CT scanner used to acquire treatment
planning images should be calibrated.
Comparison of measurements with treatment planning calculations
250 MeV, Range 28.5 cm, SOBP Width 16 cm
120.0
Relative Stopping Power & Calibration Curve
1.8
100.0
Relative Stopping Power
1.6
1.4
80.0
PDD
1.2
1.0
Meas. 16.1 cm
Calc7551 16.1 cm
60.0
0.8
40.0
0.6
ICRP tissues
0.4
20.0
0.2
0.0
0.0
-1000
-500
0
500
1000
1500
0
HU
Educational Workshop
PTCOG 46
Al Smith
50
100
150
200
250
300
350
Distance (mm)
Educational Workshop
PTCOG 46
ComparisonBone-Water
of Measurements
and Treatment
Planning Calculations
Interface
Profiles
Comparison of measurements with treatment planning calculations
Bone-Water
Profile,
250 MeV,
G1,Interface
250 MeV,
d=17.9
cmDepth = 17.9 cm
G2, RMW
MeV,
d=cm,
23.3
cm,
Cross
250 91,
Mev,250
depth
= 23.3
Cross
Plane
ScanPlane
120
120
100
SOBP04 Calc
SOBP04 Meas.
SOBP10 Calc
SOBP10 Meas.
SOBP16 Calc
SOBP16 Meas.
80
60
Relative Dose
Relative Dose
100
40
80
Measured
Eclipse
60
40
20
20
0
-15
-10
-5
0
Off-Axis Distance (cm)
PTCOG 46
Educational Workshop
5
10
0
15
-10
PTCOG 46
-5
0
Off-axis
Distance
Educational
Workshop(cm)
5
10
Prostate Patient Treatment Plan
QA of Prostate Treatment using patient treatment
parameters/appliances and EBT film in water
phantom.
Treatment plan on CT anatomy converted to
dose distribution in water phantom.
CTV
Rectum
Bladder
Femoral heads
PTCOG 46
Educational Workshop
Al Smith
Measurements in water
phantom using EBT film,
patient aperture, and
range compensator
Cross Field Profile A to P through Isocenter
Educational Workshop
PTCOG 46
Patient Treatment QA – Measurements compared with treatment
planning calculations converted to water phantom.
Data measured in water phantom using Pin-Point ion chamber.
Treatment aperture and range compensator were both inserted.
EBT film m easurement
1
Eclipse
Eclipse vs. Measured - Crossplane
Dose (arb units)
0.8
0.6
100
0.4
0.2
80
0
-65
-45
-25
-5
15
35
55
Lateral Distance (mm)
60
Cross Field Profile F to H through isocenter
40
1
Dose (arb units)
0.8
20
0.6
EBT film measurement
Eclipse
0.4
0
0
0.2
2
4
6
8
10
12
0
-70
-50
-30
-10
10
30
50
Eclips e
70
Measured
Lateral Distance (mm)
PTCOG 46
Educational Workshop
Al Smith
PTCOG 46
Educational Workshop
Al Smith
Treatment Planning
• Acquisition of imaging data (CT, MRI)
• Conversion of CT values into stopping power
(not electron density)
• Delineation of regions of interest
• Selection of proton beam directions
• Optimization of the plan
Treatment Planning
•
•
•
•
Passive scattered proton beams
Scanned Proton Beams
Intensity modulated proton beams
Comparative Treatment Plans
Dose shaping: Range compensator
High-Density
Structure
Target
Volume
Beam
Double scattering
system
Aperture
Critical
Structure
Body
Surface
© Hanne Kooy, MGH
SOBP Modulation
High-Density
Structure
Target
Volume
Beam
Critical
Structure
Body
Surface
Aperture
© Hanne Kooy, MGH
Aperture and Range Compensator
+
=
To be ‘designed’ by the planning system !
© Hanne Kooy, MGH
Compensator smearing to account for uncertainties
BEAM
© Hanne Kooy, MGH
Dosimetry and QA for SOBP proton fields
Dose [%]
120
100
80
60
40
Beam range: 17.19 cm
Modulation width: 6.78 cm
20
0
0
50
100
150
200
Depth [mm]
Dose [%]
120
100
80
60
40
Beam range: 13.47 cm
Modulation width: 8.65 cm
20
0
0
20
40
60
80
100
120
140
160
Depth [mm]
Dose [%]
120
100
80
60
40
Beam range: 12.0 cm
Modulation width: 4.0 cm
20
0
0
20
40
60
80
Depth [mm]
100
120
140
Field dependent (!) absolute dosimetry
Volume for absolute dosimetry
Output − Factor
≅
D cal
iic
⎡cGy
⎤
MU ⎥⎦
⎢⎣
EXAMPLE 1: Para-spinal case using 3 fields
10 Gy
15 Gy
20 Gy
25 Gy
30 Gy
35 Gy
40 Gy
45 Gy
46 Gy
CTV
brainstem
EXAMPLE 2: Para-spinal case using 6 fields
(involves metal CT artifacts)
15 Gy
20 Gy
25 Gy
30 Gy
35 Gy
40 Gy
45 Gy
50 Gy
51 Gy
CTV
spinal cord
Field Patching
Urie M. M. et al (1986), Med Phys. 13, 734.
• Abutting the distal
dose edge of one field to
the dose boundary of
other field(s).
• Useful if target is close
to critical structures
• Not necessarily
homogeneous dose to
the target for each beam
(IM!)
• Range an penumbra
uncertainties need to be
considered
Field Patching
“THROUGH” Field A
followed by
“PATCH” field B,
followed by
“PATCH” field C
PTV
A
50
50
Critical
Structure
B
Match at 50% isodose,
lateral + distal, levels
C
© Hanne Kooy, MGH
© Marc Bussiere, MGH
EXAMPLE 3: Nasopharynx case using 14 fields
(plus additional photon fields to the lower neck)
CTV-2
25 Gy
30 Gy
35 Gy
40 Gy
45 Gy
50 Gy
55 Gy
60 Gy
67 Gy
CTV-2
CTV-1
Parotid
Brainstem
Spinal Cord
• GTV 76 Gy
– CTV1 60-66 Gy
– CTV2 60 Gy
• Nodes 54 Gy
Treating moving targets
with protons
Effect of respiration on dose
Range fluctuations due to respiration in the lung
© Shinishiro Mori, MGH
Effect of respiration on dose
Range Fluctuation on Cardiac MRI
© Shinishiro Mori, MGH
Burr Proton Therapy Center (2001-)
Patient Population
•
•
•
•
•
•
Brain
Spine
Prostate
Skull Base
Head & Neck
Trunk/Extremity
Sarcomas
• Gastrointestinal
• Lung
32%
23%
12%
12%
7%
6%
6%
1%
In general, 1-3 fields / day
© Thomas DeLaney, MGH
Treatment Planning
•
•
•
•
Passive scattered proton beams
Scanned Proton Beams
Intensity modulated proton beams
Comparative Treatment Plans
Depth Dose
Typical Spot
Beam in Water
Beam Profile
© G. Ciangaru, MDACC
Beam Scanning
No compensator, no aperture, no scattering system required
© Eros Pedroni, PSI
A major problem with spot scanning:
The target can move !
Remedies:
• Rescanning
• Beam Gating
• Real time tumor tracking with markers
© Alfred Smith, MDACC
1. Evenly spaced/weighted spots
1
to achieve uniform field
2. 1mm spot error due to delivery
error or patient motion.
3. Optimum spacing/weighting to
achieve sharper penumbra
2
3
Pedroni
© Eros Pedroni, PSI
Dosimetry and QA of pencil beams
• Energy/Range
• large number of energies required.
• energy spacing must provide dose uniformity
over all depths
• Spot size and shape
• spot size/shape dependence on energy
• spot orientation as a function of gantry angle
• Spot position accuracy
• Validation of treatment planning calculations
• Validation of treatment delivery
• Measurements require methods for rapid collection
large amounts of data
• Real-time beam information
© Alfred Smith, MDACC
Ionization Chamber Array
Water column with 26 small
ionization chambers of 0.1 cm3
Dose box
PTCOG 46
Educational Workshop
Pedroni, PSI, Switzerland
Pedroni, PSI, Switzerland
Orthogonal IC array measurements performed at different water
depths using a computer controlled water column and
compared with calculations.
‘Beam’s-eye-view’ of dose
in water
U axis profile
T axis profile
© Eros Pedroni, PSI
Mirror
M D Anderson Cancer Center
CCD
Scintillating Plate, Mirror and
Camera
CCD Camera used for pencil
beam scanning QA.
© Martin Bues, MDACC
Scintillating Plate
Beam
Spot Pattern Test
Uniform Field Scanning Test
Scintillating screen viewed with a
CCD through a 45° mirror
WER
6.65
CM
W= 6.65 cm
WE
R
7.82
CM
Measurement vs.
PTCOG 46
– ideal for non homogeneous
dose distributions
W= 7.82 cm
Calculation
Educational Workshop
Pedroni, PSI, Switzerland
© Eros Pedroni, PSI
Treatment Planning
•
•
•
•
Passive scattered proton beams
Scanned Proton Beams
Intensity modulated proton beams
Comparative Treatment Plans
Intensity-Modulated Proton Therapy–IMPT
© Alex Trofimov, MGH
IMPT Treatment Planning
• Bragg peaks of pencil beams are distributed
throughout the planning volume
• Pencil beam weights are optimized for several beam
directions simultaneously (inverse planning)
© Alex Trofimov, MGH
IMPT Delivery
Spot scanning at PSI (Switzerland)
IMPT Delivery
• Built-in magnets for IMPT
• Two (red) “dipole” magnets to
deflect the beam in X and Y
respectively
• Two (yellow) “quadrupole”
magnets to focus the beam in X
and Y respectively
© Hanne Kooy, MGH
IMPT in clinical practice
+
+
+
+
great flexibility and variability
no need for apertures/compensators
easy delivery of large fields
beamlet optimization routines
-
requires very high degree of precision
Treatment Planning
•
•
•
•
Passive scattered proton beams
Scanned Proton Beams
Intensity modulated proton beams
Comparative Treatment Plans
Example
(IM protons vs. IM photons)
Photon IMRT
Proton IMPT
© Jan Wilkens, DKFZ
Example
(protons vs. IM photons)
Nasopharynx
(case shown earlier)
A Composite plan
(14 proton fields, 4 photon fields)
• proton fields
CTV to 59.4 GyE (33 x 1.8 Gy)
GTV to 70.2 GyE
(+ 6 x 1.8 Gy)
• Photon fields
lower neck, nodes to 60 Gy
G
N
N
G
© Alex Trofimov, MGH
B IMXT plan
(7 coplanar photon beams)
© Alex Trofimov, MGH
C IMPT plan
(4 coplanar photon beams)
© Alex Trofimov, MGH
DVH for target structures
Comparable target coverage
© Alex Trofimov, MGH
• Critical normal structures (always outlined):
– brain stem, spinal cord, optic structures
– parotid glands, cochlea
•‘Extra’ structures were outlined on 3 data sets
– esophagus, base of tongue, larynx
– minor salivary, sublingual and submand. glands
– mastication and suprahyoid muscles
© Alex Trofimov, MGH
DVH for some critical structures
© Alex Trofimov, MGH
DVH for some critical structures
© Alex Trofimov, MGH
Example
(standard protons vs. photons)
Medulloblastoma
Medulloblastoma
• Irradiation of the whole cranium and spinal axis
– Low Risk – 23.4 CGE
– High Risk – 36.0 CGE
– Spine
• Include Vertebral Body
– Cranium
• Constrain Auditory System < 40 CGE
• Pituitary / Hypothalamus ~ 45 CGE
• Posterior Cranial Fossa Boost
– Total 54 CGE
© Hanne Kooy, MGH
Protons
Photons
Copyright© MGH/NPTC 2003
Example
(protons vs. IM photons)
Prostate
IMRT
(a)
Prostate carcinoma:
(GTV + 5mm) to 79.2 Gy
(CTV + 5mm) to 50.4 Gy
Dose [Gy]
3D CPT
(b)
Dose [CGE]
IMPT
(c)
Dose [CGE]
© Alex Trofimov, MGH
© Alex Trofimov, MGH
Summary
• Proton planning offers more options in terms
of beam directions and field shaping than
photon planning
• For specific sites IMRT and protons can be
comparable in terms of dose conformality
• Protons are able to reduce the dose to most
critical structures compared to photons
• Proton therapy is able to reduce the integral
dose compared to photons by up to a
factor of 3
• IMPT is the method of choice
Thanks to
Hanne Kooy (MGH, Boston)
Alex Trofimov (MGH, Boston)
George Chen (MGH, Boston)
Martin Bues (MDACC, Houston)
George Ciangaru (MDACC, Houston)
for providing slides and figures