gsbs presentation ptc-h overview - uthgsbsmedphys.org · fluence of short-range charged secondaries...
TRANSCRIPT
4/21/10
1
GSBS Presentation PTC-H Overview
Michael T. Gillin, Ph.D.
Professor, Deputy Chair,
Chief of Clinical Services
Department of Radiation Physics
PTC H Brief History • First scattering beam patient: May, 2006 • First scanning beam patient: May, 2008 • In < 4 years, > 1500 patients, including > 50 cranial
spinal patients, > 250 discrete spot scanning patients. • April, 2010, ~ 100 patients per day being treated • Same EMR (Mosaiq, V 1.6 O3) used for photons and
protons • 96% uptime • Major developmental effort: G3 Discrete spot scanning
IMPT SFO and IMPT MFO • Observation: Proton Planning System is less mature
than the proton delivery system – Physics time before releasing new TPS SW – 4 to 6 months
4/21/10
2
A proton beam and an electron beam with the same 90% distal dose
Protons stop – In water, the PTC H scattered beam system can spot the protons with 1 millimeter increments.
Clinical Experience with Protons Goitein
• “The physical rationale for proton beam therapy is unimpeachable….
• “All this having been said, it is important that the application of protons is not without its difficulties and some limitations… While many of these limitations can be overcome, nevertheless protons are not uncritically appropriate for all patients.
• “The reduced doses delivered by protons to normal tissues could significantly improve the tolerance of patients to concurrent chemo and radiation therapy and perhaps allow the use of a greater intensity of one or both of them.”
Radiation Oncology: A Physicist’s-Eye View. Michael Goitein
4/21/10
3
Clinical Experience with Protons Goitein
• “The physical rationale for proton beam therapy is unimpeachable….
• “All this having been said, it is important that the application of protons is not without its difficulties and some limitations… While many of these limitations can be overcome, nevertheless protons are not uncritically appropriate for all patients.
• “The reduced doses delivered by protons to normal tissues could significantly improve the tolerance of patients to concurrent chemo and radiation therapy and perhaps allow the use of a greater intensity of one or both of them.”
Radiation Oncology: A Physicist’s-Eye View. Michael Goitein
PTC H Current Staffing A Large Cast
• Radiation Oncologists – Wild estimate: ~ 8 FTEs • Anesthesia Staff - > 2 • Clinical Physicists – ~ 8 FTEs 6 days per week • Dosimetrists: ~ 9 • RTTs – 2 to 4 per beam line • Anderson engineers – 3 • Hitachi engineers – 7 • Machinists – 5 A production machine shop to make
apertures and compensators • RNs, Data Managers, etc. – not enough • Administrators – too many
4/21/10
4
Major Systems for Protons • CT scanners: GE at PTC and GE and Philips on
main campus CT number/stopping power • Hitachi Treatment Delivery ProBeat with 3
gantries, 2 fixed beam lines, and 1 experimental line
• Hitachi Medical System X-Ray and Hitachi PIAS • Varian Eclipse Planning, V 8.9 • Anderson MU Determination • IMPAC EMR Mosaiq V1.6 supports scattering
and scanning • Gibbscam for aperture and compensator
construction on Mazak milling machines.
A proton beam and an electron beam with the same 90% distal dose
WHY PROTONS? It is a positive experience.
‘Low’ entrance dose
Flat dose plateau
Very rapid falloff.
4/21/10
5
Attix Chapter 8
Charged-Particle Interactions in Matter
Charged Particles Interaction in Matter
• A charged particle, being surrounded by its Coulomb electric force field, interacts with one or more electrons or with the nucleus of practically every atom it passes.
• An individual photon or neutron incident upon a slab of matter may pass through it with no interactions at all, and consequently no energy loss.
4/21/10
6
Charged Particles Interaction in Matter
• Continuous slowly down approximation (CSDA) – most charged particle interactions individually transfer only minute fractions of the incident particle’s kinetic energy.
• It is convenient to think of the particle as losing its kinetic energy in a friction-like process.
Charged Particles Interaction in Matter
• A 1 MeV charged particle would typically undergo approximately 105 interactions before losing all of its kinetic energy.
• Charged particles can be roughly characterized by pathlength, which is traced out by most such particles of a given type in a specific medium.
• Range is the expectation value of pathlength, namely the mean value for a very large population of identical particles.
4/21/10
7
A – classic atomic radius b – classic impact parameter
Charged particle interactions
Types of Charged Particles Coulomb-Force Field Interactions
A. Soft Collisions (b>>a) Atom as a whole increases higher energy state very small energy transfer (a few eV)
Roughly half the energy transferred to the absorbing medium, as large values of b are clearly more probable than are near hits on individual atoms.
4/21/10
8
B. Hard (or “Knock-On”) Collisions (b a) incident particle interacting with a single atomic electron
Single atomic electron – ejected from the atom, a delta ray
Few in number, but energy transfer is generally the same as soft collisions
C. Coulomb-Force Interactions with the External Nuclear
Field (b<<a)
( particles – bremsstrahlung)
4/21/10
9
Nuclear Interactions by Heavy Charged Particles
A heavy charged particle having sufficiently high kinetic energy (100 MeV) and an impact parameter less than the nuclear radius may interact inelastically with the nucleus.
When one or more individual nucleons are struck, they may be driven out of the nucleus in an intra-nuclear cascade process, collimated strongly in the forward direction.
In solid organic materials for protons > 500 MeV, this process dominates the energy loss.
STOPPING POWER
The expectation value of the rate of energy loss per
unit of path length x by a charged particle of type Y
and kinetic energy T in a medium of atomic number
Z is called its stopping power
dT/dx)Y,T,Z MeV cm2/g
4/21/10
10
The mass collision stopping power: sum of soft and hard
collisions
The Bethe soft-collision formula C = 0.150Z/A cm2/g
4/21/10
11
Hard-Collision Term for Protons and other heavy particles
Comments
1. Dependence on stopping medium
decrease in mass collision stopping power as Z increases
Z/A decrease 20% is going from C to Pb
2. Dependence on particle velocity -2
Note inverse dependence
3. Dependence on charged particle Z2
4. Dependence on particle mass- there is no dependence on mass.
4/21/10
12
Energy Loss by Nuclear Interactions
• Elastic Nuclear Interactions between a proton with the atomic nucleus leads to an energy transfer to the nucleus and a change in flight direction of the proton.
• The interaction probability is quite low, 0.1% for a 1 MeV proton, and decreases with increasing energy.
• Elastic nuclear interactions can be neglected for E > 1 MeV.
Energy Loss by Nuclear Interactions
• Inelastic Nuclear Interactions between a proton with the atomic nucleus leads to the excitation of the nucleus, the formation of a compound nucleus, or to a direct interaction between the incoming proton and individual components of the nucleus.
• The result is the loss of many MeV per collision and by a significant change in the flight direction of the proton.
4/21/10
13
Energy Loss by Nuclear Interactions
• The probability for inelastic nuclear reactions increases rapidly with increasing proton energy – it dominates the energy loss of protons for E exceeding a few hundred MeV.
• The relative probability for an inelastic nuclear interaction in solid organic materials is about 1% for a 25 MeV proton, 25% for a 200 MeV proton, and dominates the energy loss for E > 500 MeV.
Energy Loss by Nuclear Interactions
• Inelastic nuclear interactions are localized predominantly in the region proximal to the Bragg peak, thus mainly affecting the entrance dose.
4/21/10
14
Proton Interactions • For protons with energies between 0.01 MeV
and 250 MeV, interactions with electrons are dominant.
• For tissue equivalent material, the probability that protons will undergo a nuclear interaction while traversing a path length of 1 g cm-2 is on the order of 1%.
• At a depth of 20 cm, approximately 1 in 4 protons will have undergone a nuclear interaction. This will contribute a background of nuclear interaction products.
Mass Electronic Stopping Power, S/ ρ
S/ ρ = 1/ ρ (dE/dx) MeV M2/kg
S/ ρ depends on the composition of the material and on the nature
and energy of the charged particle.
4/21/10
15
Proton Mass Electronic Stopping Power (MeV cm-2 g-1) ICRU 59
E(MeV) Water Air Bone Polystyrene
250 3.911 3.462 3.646 3.827
200 4.492 3.976 4.186 4.397
150 5.445 4.816 5.070 5.331
100 7.289 6.443 6.778 7.140
75 9.063 8.006 8.420 8.882
50 12.45 10.99 11.55 12.21
25 21.75 19.15 20.10 21.36
20 26.02 22.94 24.06 25.62
10 45.67 40.06 41.96 45.00
5 79.11 69.09 72.28 78.20
1 260.8 222.9 233.9 257.7
0.1 816.1 730.1 791.2 916.4
Energy loss per unit distance, dE/dx
Energy loss results in a reduction of speed
Proton Energy Velocity MeV M/s
200 2 x 106 2 2 x 103
4/21/10
16
Lateral Scattering
• While penetrating matter a parallel beam of protons (with energies in the range used for therapy) spreads insignificantly sideways due to excitation and ionization. The spreading is dominated by scattering.
• The lateral spread of the dose is negligible, considering that the maximum range of the secondary electrons ejected by protons is < 0.1 cm in solid organic substances.
Multiple Coulomb Scattering • As protons slow down, they also scatter by
interactions with atomic nuclei. • Most single deflections are very small. • Moliere in the 1940’s wrote 2 papers which are
considered to be the definitive work on this topic. • In his first paper, Moliere investigated single
scattering in the Coulomb field of the nucleus, which is screened by atomic electrons.
• In his second paper he applies this to plural scattering and multiple scattering.
4/21/10
17
The Bragg Peak
• One definition: Depth dose distribution in water of a mono-energetic proton beam.
• The characteristic shape of the Bragg Peak is caused by the increase of stopping power as the proton slows down, followed by a a steep drop to zero when the proton stops.
• Range straggling does occur and so the protons do stop at slightly different depths.
Pristine Bragg Peak TPS Input Data Monte Carlo Calculated and Measured Normalization Dose in Gy/MU versus range Normalization measurements: 2 cm depth, 8 cm chamber
94 Energies: Ranges from 4.0 cm to 30.6 cm Note different shape of high energy and low energy Bragg Peaks.
4/21/10
18
The Bragg Peak Straggling
• The distribution of endpoints is Gaussian with some σs that can be found from straggling theory.
• The depth dose of a large cohort of stopping protons is the convolution of S/ρ with a Gaussian describing the range straggling.
The Bragg Peak Straggling
• Range straggling expressed as a percentage of range falls slightly with energy, but absolute straggling increases strongly with energy.
MeV % Straggle g/cm2 Straggle
70 1.21 0.050 150 1.12 0.179 250 1.063 0.407
4/21/10
19
The Bragg Peak Nuclear Interactions
• As protons slow down, they suffer nuclear interactions at the rate of approximately 1.2%/(g/cm2).
• The primary protons acquire a companion fluence of short-range charged secondaries (which receive about 60% of the primary energy) and long range neutrons. The energy of the charged secondaries is deposited near the interaction point.
March 14, 2006 One of the first SOBPs measured
Protons stop in water. Distal fall off from 90% to 10% is 7 mm.
4/21/10
20
Fluence to Dose
• By definition: Dose = Energy/Mass
• Dose = (dE/dx) x Δx x N ρ x A x Δx
Dose = Φ S/ρ where Φ is the fluence (protons/cm2) , S is the stopping power, and ρ is the density
Fluence to Dose
• Let Φ = 109 p/cm2 or 1 gp/cm2 and S = 1 MeV/(g/cm2)
• Dose = 1 gp/cm2 x 1 MeV/(g/cm2) x 103 g/kg x 0.1602 x 10-12 J/MeV x 109 p/gp
• Dose = 0.1602 Φ S/ρ Gy ( gp MeV) -1 cm2 g/cm2
4/21/10
21
Fluence to Dose • S/ρ is typically 5 MeV/(gm cm2) and a
therapeutic dose is typically 1 Gy. • Thus therapy fluences should be on the order of
109 p/cm2
• Assume 1011 p/spill which is what come out of the synchrotron
• For the 10 cm x 10 cm, then there are approximately 109 protons/spill-cm2
• For the 25 cm x 25 cm, then there are approximately 1.6 x 108 protons/spill-cm2
Electron Linear Accelerators
• Electron linear accelerators are common. For < $3M, an institution can acquire a state-of-the-device, which offers 2 x-ray energies, typically 6 MV and 18 MV, and 5 or more electron energies from 5 MeV to 20 MeV.
• RF is amplified by a klystron, dumped into a wave guide, in which electrons are accelerated. The electrons strike a water cooled target and x-rays are produced.
4/21/10
22
Proton Accelerators • Proton accelerators are expensive , on the order
of $75M, to procure and expensive to maintain, $6M/year.
• The required uptime is > 96%. • Currently in the U.S., there are 7 systems which
routinely treat patients, namely LLUMC, MGH, University of Florida, Midwest Proton Center, U of Penn, Oklahoma City, and the University of Texas Proton Therapy Center.
• In the next several years, there will be other facilities opening up
Proton Systems – Brief History • Pioneering work on proton beam therapy was done in
the US (UC) Berkeley (1954) and Uppsala Sweden (1957). Synchrocyclotrons were used until a synchrotron began supplying protons for therapy purposes in 1969 (Moscow).
• The Harvard 160 MeV cyclotron was used beginning in 1961 to treat pituitary adenomas and AVMs and later uveal melonomas
• The first dedicated proton therapy facility was built at Loma Linda University Medical Center, using a 250 MeV synchrotron, designed and built by Fermilab. 1990
• The Northeast Proton Therapy Center, located across the street from MGH, utilizing a compact cyclotron, began treating patients in 2002.
4/21/10
23
Proton Specifications
• The maximum energy for proton beam therapy ranges from 230 MeV to 250 MeV, which in water corresponds to 32 + cm.
• Beam intensity specifications are 10 to 20 nA to deliver a dose of approximately 2 Gy into a target volume within a minute or so.
Synchrotrons
• The synchrotron is a technically mature product, with high extraction efficiency (reduced radiation shielding), inherent variable energy capability, and an acceptable dose rate. (0.5 to 2.0 G/min)
• It is capable of being used for both passive scattering and spot scanning. It is easy to change energy with a synchrotron.
4/21/10
24
Cyclotrons
• Cyclotron cw beams are acceptable for both passive scattering and sophisticated scanning systems. The fixed beam energy might not be a big disadvantage.
• The fixed energy does earn demerits of unwanted neutrons and smearing the distal edge by path length straggling in an additional energy absorber.
MDACC Synchrotron • The proton spill for passive beams: spill length 0.5
seconds with a time between the spills of 1.5 seconds. • The proton spill of discrete spot scanning: maximum spill
length 4.4 seconds with a time between the spills of 2.1 seconds.
• Approximately 1011 protons are accelerated. • 90% extraction using a slow extraction technique. • At the end of the spill, beams are decelerated. The
accelerator can be entered moments after the beam has stopped, i.e. no large radiation field in the room.
4/21/10
25
Linstar Synchrotron Injector Linac AccSys – now owned by Hitachi
• Designed to provide moderate energy protons (7 MeV) for injection into high energy proton synchrotrons. These systems consist of a radiofrequency quadrupole (RFQ) linac, a drift tube linac (DTL), a debuncher cavity, and an injector with an Eizel lens.
2 Stage Linear Accelerator
Ion Source
RFQ DTL Einzel lens
4/21/10
26
7 MeV Linac Injector • The 7 MeV linac injector has an H+ ion source
and two acceleration stages: a 3.5 MeV radio frequency quadrupole (RFQ) and a 3.5 MeV drift tube linac (DTL),
• The linac has a maximum pulse current of 15 mA, operates at a frequency 0f 0.1 to 0.5 Hz, and has a 10-100 micro-second pulse width.
• Hydrogen gas is ionized, passes through a pin hole collimator, is focused with the Einzel lens into the RFQ.
Synchrotron Accelerator
Circumferential
Length: 23 meters
Extraction
Energy: 70 to 250 MeV
Energy resolution:
0.4 MeV
Design value of the extracted particles: 0.75 x 1011 ppp at 250 MeV
Extraction Scheme
Transverse RF driven with Separatrix Constant Extraction
4/21/10
27
Major Components of Synchrotron
• Bending magnets – 6 each with a beam bending radius of 1.4 m and bending angle of 60 degrees, magnetic field strength 1.76 T.
• Focusing quadrupole magnets – 4 each with a magnetic pole length of 0.2 m.
• Defocusing quadrupole magnets – 6 each with a magnetic pole length of 0.1 m
• RF Cavity – one with a non-synchronized accelerator cavity, operating frequency 1.6 to 8.0 MHz, 40 mm acceleration gap.
Synchrotron Accelerator Basics
• The proton beam circulates within the ring repeatedly through the accelerating structure which comprises part of the ring.
• In order to keep the beam within the enclosed ring, the magnetic fields of the magnets must increase in strength in synchrony with the beam energy. (Thus the name.)
• Thus, the beam is contained within the ring at a constant trajectory, as its energy increases.
4/21/10
28
Synchrotron Accelerator Basics
• When the beam reaches the desired energy, it is extracted.
• The circulation time for one turn in the ring depends upon the energy of the protons. It is usually less than 1 µsecond.
• If a fast extraction technique was used, the pulse length would be less than 1 µsecond. For passive scattering the slow extraction technique extracts the beam over 500 milliseconds.
• The Hitachi synchrotron permits an arbitrary start of the acceleration cycle and is capable of extended extraction. Thus it can be synchronize with the patient’s respiration cycle.
Synchrotron 70 to 250 MeV
• Protons are accelerated by passing through a wide band radio-frequency (RF) cavity.
• The extraction of protons from the ring is accomplished through a wide band RF-drive slow extraction scheme.
• The beam is extracted by use of the RF signal with a constant separatix which has the features that it is easy to control (all parameters of the magnets are fixed during extraction); it has stable beam parameters (beam size and position); and has a short switching time (controlled by the extraction RF power).
• The energy spread (ΔE/E) is < 0.2%.
4/21/10
29
Synchrotron 70 to 250 MeV
• The operation cycle of the synchrotron is variable from approximately 2.5 to 6.5 seconds: 1 second is required for acceleration and 1 second is required for deceleration. The spill length is variable between 0.5 and 5 seconds for passive scattering techniques and between 0.5 and 4.4 seconds for beam scanning techniques.
• The extracted beam energy can be changed from pulse to pulse and approximately 1011 protons can be extracted per pulse. The energy resolution is 0.4 MeV and the reproducibility of the beam position in the extraction channel is less than + 0.5 mm.
Synchrotron Basic Operation Pattern
4/21/10
30
20 nC per pulse or 1011 protons per pulse
4/21/10
31
4/21/10
32
High Energy Beam Lines
• After the protons have been extracted, they enter the high energy beam line to be delivered to one of 6 different treatment nozzles (5 clinical and 1 experimental).
• 2 passive scattering gantries, 1 scanning beam gantry, 1 large field fixed, 1 small field fixed (the eye line), and 1 experimental line – NASA and animals
4/21/10
33
Gantry Beam Transport
Gantry Treatment Rooms • The gantry rings are 5.5 meters in
diameter, the rotating mass is 190 tons, and the gantry rotates 360 degrees ( + 180 degrees with 10 degrees over travel) around the patient.
• The maximum speed is 1 RPM. Emergency stop will occur within 4 degrees at maximum speed.
• The gantry mechanical isocenter is required to be contained within a sphere of < 1 mm diameter.
• A correction algorithm will correct for residual gantry errors at each gantry angle: this correction is made by repositioning the couch when a correction request is made on the treatment control pendant.
4/21/10
34
3 Gantries 190 tons
• Gantry rotation +/- 180 degrees at a speed of 1 revolution per minute.
• Gantries are supported by two concrete pillars.
Gantry from Underneath
• Counter-weight • Bending magnets. • Nominal TAD: 270 cm • Gantries hold
isocenter better than the typical Varian gantry
4/21/10
35
Some bending magnets on the gantry
• Beam optics are very complex.
• Gantry 3: 94 different beam energies which are steered over 360 degrees.
• Gantry 1 and 2: 8 different beam energies which are steered over 360 degrees.
Passive Scattering A well established method
• The proton beam is small (approximately 4 to 8 mm) when it leaves the beam line. It needs to be spread out laterally to cover field sizes of 10 cm x 10 cm (small snout), 18 cm x 18 cm (medium snout), or 25 cm x 25 cm (large snout).
• A double scattering system is used. The exact components depend upon the beam energy and the field size.
4/21/10
36
Treatment Rooms 3 Gantries + 1 Fixed Beam Room
with 2 beamlines
• Each treatment room contains 3 separate rooms, namely the treatment room itself, an imaging room, and a ‘behind the wall’ room, which contains many different electronics modules.
4/21/10
37
Passive Scattering A well established method
• The pristine Bragg Peak is spread out using a rotating (400 revolutions per minute) modulation wheel (RMW) to produce the Spread Out Bragg Peak.
• There are 3 peaks (6 modulating slopes) on the RMWs and SOBPs from 2 cm to 16 cm can be obtained.
• For high energy proton beams, the RMWs are made from Al, while a plastic is used for lower energy beams.
Passive Scattering Nozzle with Range Modulation Wheel
The aperture and compensator production shop – 4 talented mechanists.
4/21/10
38
For passive scattering the Pristine beam passes through various thicknesses of the RMW to create the Spread Out Bragg Peak (SOBP).
Passive Scattering A well established method
• Patient specific field shapes are made from bronze.
• An acrylic compensator is used to account for the heterogeneity of the human body.
• There are 8 different energies (100, 120, 140, 160, 180, 200, 225, and 250 MeV.)
• The range of protons in water can be controlled to within 1 millimeter from 4 cm to 32 cm using range shifting plates.
4/21/10
39
Ranges (cm) (d 90%)
Energy (MeV)
Small Snout
Medium Snout
Large Snout
250 32.4 28.5 25.0 225 26.9 23.6 20.6 200 21.8 19.0 16.5 180 16.9 16.1 13.7 160 13.4 13.0 11.0 140 10.2 10.0 8.4 120 6.9 6.4 6.3 100 4.9 4.3 4.3
4/21/10
40
Passive Scattering A well established method
• Patient specific field shapes are made from bronze.
• An acrylic compensator is used to account for the heterogeneity of the human body.
• There are 8 different energies (100, 120, 140, 160, 180, 200, 225, and 250 MeV.)
• The range of protons in water can be controlled to within 1 millimeter from 4 cm to 32 cm using range shifting plates. (Note to Physicists: Patients are not water tanks.)
Summary of Typical Penetration Uncertainties
standard energy (or range) 0.6 mm energy (or range) reproducibility 1.0 mm bolus WET 0.9 mm alignment devices* 1.0 mm
CT# accuracy (after scaling) 2.5% RLSP of tissues and devices 1.6% energy dependence of RLSP 1.0% CT# to RLSP (soft tissues only) 1.5%
bolus position relative to patient variable heterogeneity straggling variable patient motion variable
Range Uncert. 2 mm
CT Uncert. 3.5%
Planning bolus expansion multiple angles
Moyers PTCOG 2008
4/21/10
41
Range compensator - Smearing
• Range compensator to account for heterogeneities along the beam path.
• Smearing of the compensator: – Ensure distal coverage of the CTV when there are setup
uncertainty and shifts in the relative position of anatomy.
Urie et al 1983
Range compensator Example: No compensator smearing applied
Bone
Tumor
Bone
Tumor
Distal coverage not guaranteed as the position of the bone relative to the tumor changes.
Lung Lung
4/21/10
42
Range compensator Example: Compensator smearing applied
When appropriate smearing is applied, then distal coverage is assured, but as a consequence, the dose distribution becomes less conformal.
Bone
Tumor
Bone
Tumor
Lung Lung
Dose and Dose Equivalent
• Oncologists only speak in terms of dose equivalent.
• Dose distributions are interpreted in terms of dose equivalent.
• Physicists convert to physical dose by dividing dose equivalent by 1.1 to establish the number of MUs.
• This works.
4/21/10
43
Patient Specific QA 5 cm/10 cm water box Which is designed to
hold a Farmer type chamber
Solid water plates to obtain the desired depths.
Note brass apertures and no compensator
The Physics Miracle transforming treatment plans into treatment delivery parameters, including MUs. Generally physics spends 1 to 2 hours after planning is finished to review and prepare for treatment. 27 field cranial spinal patients may require 8 to 10 hours.
Image Guided RT No reticule – No Light Field Set up
• There are 3 x-ray tubes and flat panel detectors. The systems in the nozzle and the cage are used routinely.
• It is challenging to confirm the alignment of the proton beam and the x-ray beam to within 1 mm.
• The alignment of the x-ray system and the lasers are confirmed daily, together with the communication between the Patient Positioning Image and Analysis System (PIAS) and Mosaiq.
4/21/10
44
Proton Treatment Planning • Limited number of fields – 2 to 4 per target • Generally each field delivers a dose which covers the
CTV • An attempt is made to avoid aiming the beam at critical
structures, e.g. cord, due to uncertainties as to where the beam is actually stopping.
• There are no good tools to help evaluate the robustness of the treatment plan relative to changes in the patient’s anatomy or changes in the patient’s position.
• The penumbra depends upon the depth and at depth (25 cm) is on the order of 10 mm.
• At least 24 hours is needed to construct the apertures and compensators and to do the physics QA.
• The DVHs lead to DVH idolatry.
ICRU Report 78 Prescribing, Recording, and Reporting
Proton Beam Therapy, 2007 • Proton-specific issues regarding the PTV –
A Religious War in the Proton Community • “It is required that the dose distribution
within the ‘PTV’ be recorded and reported. This would be unworkable if there were a separate PTV for each beam employed, and impossible if separate lateral and depth margins were built into the computer’s beam design algorithm.
4/21/10
45
ICRU Report 78 Prescribing, Recording, and Reporting
Proton Beam Therapy, 2007 • Proton-specific issues regarding the PTV • “It is therefore proposed that, in proton therapy,
the PTV be defined relative to the CTV on the basis of lateral uncertainties alone. An adjustment must then be made within the beam-design algorithm to take into account the differences, if any, between the margins needed to account for uncertainties along the beam direction (i.e. range uncertainties) and those included in the so-defined PTV (i.e. based on lateral uncertainties).
Cranial spinal patient supine
> 50 cranial spinal patients treated in 3 years. Each new field requires new compensators. Spinal fields change each week.
4/21/10
46
120 MeV, Medium snout Range: 4.4 to 6.4 cm
16 month old
Retinoblastoma
45 CGE (40.9 Gy)
Note:
Limited Penetration of beam
1 YO Alveolar Rhabdomyosarcoma
45 CGE Single Field Entrance Dose 37.5 CGE
4/21/10
47
Prostate Dose distribution 80% to 20% penumbra 250 MeV
Surface 0.50 cm
12.5 cm 0.66 cm
23.5 cm 0.92 cm
28.5 cm 0.93 cm
Treatment Process Time in F2 (Prostate, Two Fields)
Average in Room Time: 22 minutes
K. Suzuki, PhD
4/21/10
48
Protons have a limited range, which should limit toxicity
Patient with T2, N0, MX
COPD
87.5 CGE
Limited dose to the non-involved lung
Note: Penetration through lung
3 fields,
Lateral and 2
Posterior
obliques
Standard fractionation
Original Proton Plan Dose recalculated on the new anatomy
Bucci/Dong et al. ASTRO Abstract, 2007
Impact of Tumor Shrinkage on Proton Dose Distribution
4/21/10
49
87.5 CGE Early Stage Lung Cancer
87.5 CGE DVH DVH Idolatry
Mean Dose Lt Lung 34 Gy, Rt Lung < 1 cGy, ICTV 91 Gy
4/21/10
50
140 MeV Protons and 50 MeV Electrons (U of Michigan)
Protons: Flat peak, sharp drop off. Range controlled to within 1 mm in water. What clinical sites benefit from these characteristics?
Basic Proton References
• ICRU Report 59 (1998): Clinical Proton Dosimetry Part 1: Beam Production, Beam Delivery and Measurement of Absorbed Dose.
• ICRU 78 (2007): Prescribing, Recording, and Reporting Proton-Beam Therapy.
• Proton and Charged Particle Radiotherapy, Thomas F. DeLaney and Hanne M. Kooy, 2008