Particle Beam Technology and Delivery
AAPM Particle Beam Therapy Symposium
Flanz 2013; AAPM Symposium
Types of Accelerator Systems Cyclotron Synchrotron Laser Linac
FFAG
Isochronous Cyclotron
Synchro- Cyclotron
CycLinac Strong Focusing
Rapid Cycling
Weak Focusing
CycFFAG
DWA
Rf Linac
2 Flanz 2013; AAPM Symposium
Vladimir Derenchuk
Jay Flanz
Rock Mackie
3
Particle Beam Technology & Delivery
Synchrotrons
J. Flanz, Ph.D. Massachusetts General Hospital
Harvard Medical School
Flanz 2013; AAPM Symposium
Ongoing Themes in Field of Particle Therapy Examples for CONTEXT of Technology and Delivery:
1. Pencil Beam Scanning (PBS) – Impact on: Beam Parameters from Accelerator + Delivery – Scanning “type”; Beam Size; etc.
2. Image Guided Therapy (IGRT) – Impact on: Imaging; Beam Alignment – PROTON Radiography/Tomography ????
3. Organ Motion – Impact on: Beam Parameter timing; Beam Tracking
4. End of Range ® - Proton Range vs. HU ? 5. Field Directions(θ,ϕ,ψ): How to treat specific sites? 6. Lower Capital Costs $$$ 7. Increased Throughput
– Positioning, Aligning, Field-to-field time, Beam time
4
These may not be new concepts, but they are the current foci owing to the fact that the ‘first’ round of system specs have been satisfied. (i.e. the Berkeley/MGH report of 20 years ago.)
X-ray Simulation Proton Radiography Simulation
Joao Seco, MGH
Flanz 2013; AAPM Symposium
Separated Themes may not lead to system solutions! Need System Solutions
• How to deliver a Rx non-uniform dose distribution to a moving target with a desired conformance? vs. e.g. Continuous Beam Scanning: with 3mm Sigma – This involves beam parameter TIMING issues, beam trajectory
and range manipulation, Apertures (or not), etc. • How to deliver quality treatment to the appropriate number
of patients at an affordable cost? vs. e.g. Out of room setup; Energy Change in 2 seconds – This involves automated remote operations
• Imaging and Analysis and Correction of Position • Go from Field to Field without delays (e.g. Moving things remotely)
5
It’s not just the accelerator. It’s the other components and how the accelerator works with the other components !
Flanz 2013; AAPM Symposium
6
Goal of Particle Radiotherapy • Deliver required dose to a target in a reasonable time • Deliver the correct dose distribution to the correct location
– Minimize the dose outside the target
A B
C
D E
Dosimetric Regions of Interest
Depth Direction
Transverse Direction
A B
C
D E
Dosimetric Regions of Interest
Depth Direction
Transverse Direction
• C&D = desired dose • A+B+E = unwanted dose
• E, B&part of A= desired dose • Some of A+(C-E) = unwanted
dose
E
(Scattering)
Flanz 2013; AAPM Symposium
Use this as the building blocks to deliver dose
Spread out Transverse Dose with Scanned beam spots Y0
0
0.2
0.4
0.6
0.8
1
1.2
-8 -6 -4 -2 0 2 4 6 8
7
Y0
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
0 10 20 30 40 50 60 70 80
Y0
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
0 10 20 30 40 50 60 70 80
Transverse spreading using superposition of unmodified beams.
Uniform dose scenario.
Gaussians are Magic
Spac
ing:
2 si
gma
1
sigm
a
Asymmetric Sharp Edged
Flanz 2013; AAPM Symposium
Scanning – The “Penumbra” of a scanned beam
PSI/Berkeley
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
8 Flanz 2013; AAPM Symposium
Small beam sigma vs. Cut off edges.
In this case the Technology required has to deliver a dose modulated distribution
In this case the Technology required has to produce a different beam profile (where it counts).
9
Materials in the Beam Path – Including the Patient
Stefan Schmidt
Ran
ge
Shift
er
4.5
18
Flanz 2013; AAPM Symposium
The Raw beam shape is NOT the only contributing factor
Dose / Dose Rate • Power = Joules/sec = Energy * Current
– e.g. 150 MeV * 1 nA = 0.15 Watts • Dose = Joules/kg ≡ Gray (Gy)
– Dose = (Power* seconds) /kg – e.g. 150 MeV * 1 nA * 60 sec = 9 Joules
• Water 1kg/1000cc = 1kg/liter • Dose = 9 Joules /1kg (in a liter) • 150 MeV, 1nA == 9 Gy in 1 liter in 1 minute • But not all energy goes into the target (see Bragg peak) 3-6
Gy in 1 liter in 1 minute
• Therefore, for 1Gy in 1 liter we need ~ 120 GigaProtons – (120 GP/min ~ 0.3nA)
10 Flanz 2013; AAPM Symposium
What is a Synchrotron?
11
Ener
gy
# Pr
oton
s
Time Flanz 2013; AAPM Symposium
Ring Orbit – Alternating Focusing
Unperturbed closed orbit
Perturbed a little - closed orbit
Perturbed a lot - closed orbit
12
Focusing elements
Flanz 2013; AAPM Symposium
Synchrotron Beam Dynamics
13 Flanz 2013; AAPM Symposium
Resonant Extraction (one method)
14
0
1
2
3
4
5
6
0 0.5 1 1.5 2
Fiel
d St
reng
th
Horizontal Displacement
Linear
Octupole
• The process is partly stochastic (uncorrected time structure is not smooth)and • The extracted beam phase space is NOT Gaussian in the extraction plane (depending on the type of extraction). x
θ Extraction Septum
Flanz 2013; AAPM Symposium
Beam size from the last magnet to Isocenter?
15
0
2
4
6
8
10
12
2 4 6 8 10
Dip
ole
Pow
er/W
eigh
t
1 sigma Isocenter Beam Size
Gantry Dipole Power / Weight
PS Current ~ Gap (2.1 sigma) Magnet weight ~ gap 2 Power ~ Current 2 ~ gap 2
Gantry/Beamline Dipole
Isocenter
Flanz 2013; AAPM Symposium
Treatment Time Contributions Synchrotron
• Time to Inject • Time to Accelerate • Time needed to wait until the patient is
ready for particles (e.g. gating) • Time needed to extract particles
– Instrumentation will only allow a finite number of particles per unit time
• Time needed to Decelerate
16
If m
ore
parti
cles
are
nee
ded
or c
hang
e en
ergy
Additional ‘cycle’ times are needed if there are not enough protons in the ring to deliver the required dose at a given range.
Flanz 2013; AAPM Symposium
To
Tb
Breath
Variable Cycle Synchrotron Beam Available
Beam Available
Beam Available
Fixed Cycle Synchrotron or Cyclotron
CW Cyclotron
Okay to Deliver Beam
Time needed to wait until the patient is ready for particles
17
Comparison of Beam Utilization for Treatment Requiring Synchronization
Effective Dose Rate may depend upon accelerator time structure
Flanz 2013; AAPM Symposium
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+
+ +
+ +
+
+ + +
+ +
+ +
+ + +
+
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ + +
+ +
+ +
+ +
+
+ +
+ + +
+ +
+ +
+ + + +
+
+ +
+ +
+ +
+
+ +
+ +
+ +
+ +
Charges ⇒ repulsion Parallel currents ⇒ attractive
Are additional cycle times needed? Accelerator Physics: Space Charge effects
How many protons can be stuffed into a Ring?
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+
+ +
+ +
+
+ + +
+ +
+ +
+ + +
+
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ + +
+ +
+ +
+ + +
+ +
+ + +
+ +
+ +
+ + + +
+
+ +
+ +
+ +
+
+ +
+ +
+ +
+
x
x
Linear
Non-linear
Defocusing force
Gaussian density
distribution
y
18
Coulomb
Flanz 2013; AAPM Symposium
0.00E+00
2.00E+10
4.00E+10
6.00E+10
8.00E+10
1.00E+11
1.20E+11
1.40E+11
1.60E+11
0 2 4 6 8 10 12
Ext
ract
ed C
urre
nt (n
a)
Proton Injection Energy (MeV)
Proton Limit in Ring due to Space Charge Effects
Assumptions: A finite cycle time High Efficiency Extraction Beam Size
How many protons can be stuffed in a ring? How many are needed?
LLUMC MDA
Also 1Gy/min in a liter 113GigaProtons/min <3Gp/cycle
McLaren
19 Flanz 2013; AAPM Symposium
20
First Proton Therapy Synchrotrons
LLUMC, Tsukuba, MD Anderson, Shizuoka, Wakasa, HIMAC, Hyogo, McLaren, Mayo …
Hitachi Extraction
20 Flanz 2013; AAPM Symposium
Rapid Cycling Medical Synchrotron the second generation
Synchrotron accelerator
Gantries
Synchrotron Extremes Faster/Bigger Smaller / Cheaper
• Current (~<nA?). • Beam Time Structure (maybe) - Scanning (not line?) / Organ Motion • Lower Emittance – Good for beamlines and Gantries • Energy Change time linked to acceleration time (Faster?) • At 30 Hz in 120 sec ==> 3600 pulses ! (What can be done ?)
• 10x10x10cm (@3mm spot) ~ 10,000 spots (ONE pulse/spot)
“Compact” Medical Synchrotron
Accelerator Physics VS. ANTI-Accelerator Physics?
21 Flanz 2013; AAPM Symposium
NIRS Japan
Heavy Ion Accelerators and Facilities - Conventional
Remember: 400MeV/nucleon: Could be big and Expensive! But they are shrinking also !
22
Heidleberg Flanz 2013; AAPM Symposium
What might someone do with a Blank Slate? • How to deliver a Rx non-uniform dose distribution to a moving
target with a desired conformance? – Scanning beams with adjustably sharp edges – Accelerator with timing consistent with Scanning and organ motion
(Flexible timing) – Appropriate on-line imaging
• How to deliver quality treatment to the appropriate number of patients at an affordable cost
– Accelerator Design: • Make it INEXPENSIVE, Simple
– Design only what’s needed » What Energy is needed? (Treatment, Tomography, … ) » What Current is needed? (Scattering, Scanning, Losses …)
• Simplify the Optics – Minimize Building costs; e.g. Some Shielding Requirements
• No ‘designed’ sources of BEAM LOSS – Chose a Beam delivery like Scanning
23 Flanz 2013; AAPM Symposium
24
Gratuitous Comments: • Treating with particles requires a system approach. • The various subsystems interact with each other and
depend upon each others capabilities. • Trade-offs include size, speed, intensity, of
everything, (equipment, beam etc.) • More and more demand for affordable particle
therapy solutions is apparent. • New Approaches are being fueled by both
accelerator interests, and by the more and more demanding requirements of particle therapy.
25
The Francis H. Burr Proton Therapy Center
Thank You ! 25