Technological innovations

Transcription

1 Technological innovations Paul Scherrer Institute SWITZERLAND Eros Pedroni D. Meer, C. Bula, S, Safai, S. Zenklusen R. Kobler, S.Lin, PSI E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

2 Summary 1. Gantry layout 2. Fast scanning 3. Organ motion solutions 4. Competition or complement? ion therapy 5. Proton Radiography 6. Questions E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

3 1. Gantry layout E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

4 Integration of the whole system for optimized beam delivery Combined and optimized use of Cyclotron degrader beam line gantry nozzle patient handling All contributing together to the best possible (dynamic) beam delivery Medical cyclotron Degrader Gantry 2 Collimator E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

5 Easy access to the patient Gantry rotation limited to -30 to (instead of 360 ) Flexibility of beam delivery by rotating the table in the horizontal plane Analogy with longitude and latitude in the world-geography Expected advantages: Fixed walls for mounting supervision equipment - like Vision-RT Permanent fixed floor for a better access to the patient table Nobody (patient or personnel) falling in the gantry pit PSI Still River E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

6 In-room positioning with sliding-ct Within reach of the patient table Sliding CT of Siemens Use of time-resolved images before (and after) treatment Adapt dose field to the organ situation of the day (the body regions with soft tissues) Setup of respiration gating The first proton gantry of the world coupled with a sliding CT Adaptive system to new diagnostic equipment CT-PET? MRI? E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

7 BEV X-rays Retractable support with a flat panel X-ray tube on beam axis Shining through a hole in the yoke of the 90 bending magnet Simultaneous use of x-rays during proton beam delivery For controlling the position of moving targets during beam delivery Image guided Proton Therapy? E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

8 Option to use BEV X-ray - simultaneous to proton beam? Equivalent to portal imaging with KV photon Large field-of-view area (26 cm x 16 cm) not masked by equipment or collimators in the beam path For QA control of gating and tracking (scanning + pulsed X-rays) Fluoroscopy? Neutron damage to panel? Proton beam Sweepers Upstream scanning PSI X rays tube Yoke hole Bending magnet nozzle Patient Imager Sweeper or Scatterer Collimator a) compact gantry b) long throw gantry Downstream scanning IBA Hitachi Varian E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

9 Optimized nozzle design Vacuum up to the patient Sharp pencil beam - 3 mm sigma Two monitors and a strip monitor 2 mm strips (delivered by TERA collaboration) Removable pre-absorber IN and OUT of beam (motorized) For ranges below 4? -10? cm Telescopic motion of the nozzle To reduce air gap (keep patient at isocenter) Option to add collimator and compensators To shield OAR on top of scanning To simulate passive scattering with a scanning beam Collision protection to treat patients remotely (multiple fields in one go) Field extension with patient table E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

10 Nozzle: preliminary results small size of pencil beam Use of minimal material in the nozzle for keeping the beam size between 3 and 4 mm sigma at all energies The quality of the beam depends from the design of the beamline and of the nozzle mm Measured while removing piece by piece the materials in the nozzle E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

11 2. Advancing the scanning technology E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

12 Very fast changes of the beam energy Continuous choice of the beam energy Setting all elements of the whole beam line within a single command Constant beam transmission from COMET to the gantry Compensation of degrader losses from 100 to 200 MeV Fast energy changes Cyclotron (fixed energy) Fast degrader ahead of the gantry The beam line follows the energy variations in the degrader Shown 80 ms dead time for range steps of 5 mm Dynamic beam energy A factor of 10 better than other systems E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

13 Double parallel scanning Fast parallel lateral scanning T sweeper 2 cm/ms U sweeper 0.5 cm/ms Scan area of 12 cm by 20 cm Motion of patient table for treating larger field sizes Table used as sweeper-offset Experience with Gantry 1 Apparent source at the infinity Simplify treatment planning Easy field patching with table Avoid errors from compensators Simplify verification dosimetry T U E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

14 Well focused beam for all energies 100 MeV 120 MeV Achieved by connecting a short correction quadrupole in series with the U sweeper 140 MeV 160 MeV 180 MeV 200 MeV Parallelism Max deviation ~4 mrad (at edge of field) E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

15 Vertical deflector plate (COMET) Dynamic use of the modulation of the beam intensity Deflector plate and vertical collimators in the first beam turn after the ion source Time delay to extracted beam in the order of 100 us Examples Delivery of line segments with changing voltages on the deflector plate Pulsing beam E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

16 Flexible control system Steering file for combined delivery of Spots Spot scanning as the default (starting) mode Lines For maximum repainting number and simulated scattering Contours? For optimizing repainting and lateral fall-off (difference Gaussian to error-function) Including the dose control for the passive scattering system of Optis 2 E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

17 Tabulated dose delivery with FPGA Combined tabulated control of U-sweeper T-sweeper Beam intensity As a function of time Example 1 U T meander path T U Example 2 - Dose box with continuous scanning 494 energy layers (85 ms per layer) (6 x 8 cm) in less than 1 minute E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

18 Dose control of time driven lines scan Time driven devices U - T - I Dose control with feed-back loop Input: difference of Monitor 1 to required dose rate Output vertical deflector plate voltage Paths with variable speed and/or variable intensity Dose linearity of simple T-lines 10 cm in ms 1 time *0.5 Constant intensity Variable T speed 5 ms is the limit systematic errors!!! Delivered MUs ms Required Dose 10 cm in 5 ms 23 times Max T speed Variable intensity E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

19 Conformal volumetric repainting Painting of lines (contours?) With maximal possible velocity ~ 2 cm / ms, 0.5 cm/ms Dose shaping through Beam Intensity Modulation (BIM) Painting of energy iso-layers < 200 ms per plane (20 lines x 5 mm) Change of energy (100 ms - 5mm range) Repainting of iso-layers ~ 6 s per liter (20 energies at 5mm steps) Volumetric repainting capability (aiming at) 10 repaintings / liter in 1 or 2 minutes Analogy with a TELEVISION tube E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

20 3. Coping with the organ motion problem (of scanning) E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

21 (Uncontrolled) motions less than 5 mm repainting methods Scaled repainting Repeat same scan n times with dose divided by n Layered repainting Deliver partial scans with constant max. partial dose Revisit only those spots which need more dose Volumetric repainting Repeat volumes Requires fast energy changes Non volumetric repainting Repeat only iso-energy layers beam 0 % missing dose 25 % missing dose 50 % missing dose 75 % missing dose 100 % missing dose E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

22 How many repaintings do we need? Analysis of repainting strategies - thesis work of S. Zenklusen (on press) 15 repainting in 1.5 minutes Full parameter space: Frequency [1/min]: Start phase: 0-2π t max [ms]: 1, 2, 3, 4, 5, 7,10, 20, dose calculations on PSI Linux Cluster Standard error reported for each dose distribution as difference to - repainting within the target Motion amplitude 5 mm Volume 0.5 liter G1 volumetric scaled repaining G2 volumetric scaled repainting G2 volumetric iso-layer repainting G2 volumetric lines Volumetric repainting is important for the motion mitigation of scanning beams Breast, pancreas, cervix, E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

23 Respiration motion (> 5mm) Does gating need repainting? Only if residual motion error is small enough (we don t known yet) Which gating signal should we use? External signal (like thoracic strip?) 3-d stereo viewing of external marks on patient skin? A whole energy layer delivered within a gate? Needs fast iso-energy layer painting ( ms per plane) Synchronize energy change to next gate Beam No need for fast energy changes But treatment time can be very high E Target at times t i Plan1 Iso-energy layers E at t i E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

24 Respiration motion- is tracking a viable solution? Lateral corrections as a function of the target displacement (easy) Energy corrections as a function of the target displacement (Range shifter) Density heterogeneities in the beam path (example ribs in front of a lung tumor) Deformation model of the target (complex delivery and treatment planning) Energy corrections -> Slow painting -> Probably only one single painting Residual error precision? Are 1-2 mm required? Is the residual error of detecting motion small enough to avoid repainting? Most of the expertise in this direction is at GSI (C. Bert) Beam Lung Rib Nodule E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

25 Why not then Multi-gating? synchronized plans delivery Associate several time-gates with corresponding gated-ct plans Deliver multiple time-instances-treatment plans synchronized with breathing Energy layer of a plan is delivered within its own gate ( ms/plane) A better alternative to tracking? No need for energy corrections - No target motion-deformation model needed Repainted by definition Fast scanning - Good duty factor of using of the beam Plan3 change to the next energy when all iso-energy plans are completed Beam Plan1 Plan2 E Target at times t i Iso-energy layers E at t i t 1 t 2 t 3 E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

26 Respiration motion (> 5mm) Our preferred solution Whole painting within a breath hold cycle Small volume (<1/4 liter) Full volume treated within 4s with intensity modulated line scanning Needs fast energy changes High dose rate could be of advantage for reducing treatment time Especially in the context of a hypo-fractionation A breath hold length of 5 s would then be sufficient Repainted, high efficiency, static planning, easier QA Beam Breath hold Whole target E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

27 Part 4. Competition with carbon ions E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

28 The rationale of using carbon ions Arguments in favor of carbon Correct - The high LET of heavy ions could be of advantage for treating Radio-resistant tumors Poorly oxygenated tumors (OER oxygen enhancement ratio) Correct Better precision of the beam - Less MCS - Sharper Bragg peaks But with fractionation tails Large dose uncertainties due to the varying RBE Simplistic - Carbon is better because is more effective (higher RBE) What matters is the RBE difference of healthy to tumor cells the therapeutic ratio Possibly wrong - Carbon therapy needs only few fractions is therefore less expensive Protons can also be delivered with low fractionation (small tumors radio-surgery) At a high dose per fraction protons behave similar to high-let (example OPTIS) Arguments against carbon High LET can be a contra-indication (for example for pediatric treatments) Absence of repair of the healthy tissues (tolerance within the target) Risk of late effects (not convincing experience with neutrons and pions) Assessment of the dose is difficult - needs an own radiobiology research group E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

29 Carbon + proton facility: a new standard for the hospitals? Carbon ions The magnetic rigidity of carbon is 3 times higher than with protons The dimensions and costs of the facility are > doubled size x 2 3 Huge gantry or no gantry at all 70m C+P? or hypofractionated P? The answer will come in the future 37m Protons E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

30 5. Other DEVELOPMENT topics (QA related) E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

31 Proton radiography (tomography) Image of the range loss through the patient in-vivo Poor image contrast due to MCS QA tool for range control Predict proton stop (PSI development work in the 90s) The key for a new paradigm of working with the precision in range Distal fall-off is potentially sharper then lateral fall-off (by a factor of 2) Radiography Range E1 E2 Therapy x,y x,y Stopping the beam in front of OAR PR of Rando phantom E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

32 CONCLUSIONS Proton therapy must remain competitive with conventional photon therapy IMPT and scanning beams are needed to compete with IMRT PT will have to follow the development towards image-guided RT Scanning is most optimal for biological targeting (intentional non-homogeneous dose distributions) and adaptive beam delivery The experience of using scanning with organ motion is still missing Fast volumetric repainting and volume within a breath-hold could be an easy solution of this problem Initiatives to reduce the size and costs of the equipment (new accelerators) are welcome but difficult to achieve at equal performance Is high-let radiation a true clinical need? More clinical research with ions needed Superconducting cyclotron and gantries could help ion therapy E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School

33 THANK YOU E. Pedroni Center for Proton Radiation Therapy - Paul Scherrer Institute - Proton Therapy Winter School