IN VIVO IMAGING
Proton Beam Range Verification
With PET/CT
Antje-Christin Knopf 1/3
K Parodi 2
, H Paganetti 1
, T Bortfel...
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
Why do we want
to make that
effort?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
Optimal treatment
Protons have the superior advantage of a finite range,
but uncertainties compromise this advantage.
MOTI...
Optimal treatment
Since we often don’t know the uncertainties we often don’t apply the optimal
treatment.
Uncertainties ca...
What is the
idea?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
METHOD
Procedure
1.
Proton Treatment at the F. H.
Burr Proton Therapy Center
2.
Walk the patient to the
PET/CT scanner
3.
...
METHOD
Nuclear reactions
In this approach we do not use any radioactive tracers but positron emitters, which
are produced ...
METHOD
Data
DOSE planned Dose
PET
ACTIVITY
measured PET
- dyn.
- FB
- IT
CT planning CT PET CT
METHOD
Data
DOSE planned Dose MC Dose
PET
ACTIVITY
measured PET
- dyn.
- FB
- IT
MC PET
CT planning CT PET CT
The detailed...
What do we want
to achieve with
this data?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
GOAL
Dose verification
difficult because:
no unique correlation between dose and
activity distribution
patient and tissue ...
GOAL
Dose verification
difficult because:
no unique correlation between dose and
activity distribution
patient and tissue ...
GOAL
Range verification
DOSE planned Dose MC Dose
PET
ACTIVITY
measured PET
- dyn.
- FB
- IT
MC PET
CT planning CT PET CT
GOAL
measured PET
planned dose
Range verification
match within x mm
match within y mm
range was correct within
(x+y) mm
MC...
GOAL
Range verification
normalize
GOAL
Range verification
pointwise
20%: - sensitive to
smoothing of MC
profiles
- sensitive to back-
ground noise
50%: - se...
GOAL
Range verification
shift
more robust strategy for
range verifications than a
pointwise comparison
Is that technical
and physical
feasible?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
RESULTS
Phantom
1.) Homogeneous phantom and simple slab phantom
Parodi et al “PET/CT imaging for treatment verification af...
RESULTS
Phantom
1.) Homogeneous phantom and simple slab phantom
Beam Parameter: Slab phantom: one field, 16cm range, 2Gy t...
RESULTS
Phantom
1.) Homogeneous phantom and simple slab phantom
Results:
Activity composition: Main fraction from 11
C, mi...
RESULTS
Knopf et al “Quantitative assessment of the physical-potential of proton beam range verification
with PET/CT”, sub...
RESULTS
Phantom
2.) Complex inhomogenous phantom with different angled tissue
interfaces
Beam Parameters: One field, 15 cm...
RESULTS
Phantom
2.) Complex inhomogenous phantom with different angled tissue
interfaces
Results:
Physical feasibilities: ...
How does it
look in clinical
reality?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
RESULTS
Patients
# of patients # of patients
that received
1 field
# of patients
that received
2 fields
dose
per field
[Gy...
RESULTS
Patients
1.) Head and neck tumor sites
Parodi et al “Patient study on in–vivo verification of beam delivery and ra...
RESULTS
Patients
1.) Head and neck tumor sites
Advantages:
few patient motion
-> the same immobilization as during the tre...
RESULTS
Patients
1.) Head and neck tumor sites
Data analysis:
At positions where the beam stopped in bone
Parodi et al “Pa...
RESULTS
Patients
1.) Head and neck tumor sites
Data analysis:
At positions where the beam stopped shortly behind in bone
P...
RESULTS
Patients
1.) Head and neck tumor sites
Data analysis:
At positions where the beam stopped in soft tissue
Parodi et...
RESULTS
Patients
1.) Head and neck tumor sites
Results:
In soft tissue biological washout effects degrade the measured act...
RESULTS
Patients
2.) Abdominopelvic tumor sites
RESULTS
Patients
2.) Abdominopelvic tumor sites
Challenges:
motion
-> breathing and organ motion results
in a blurring of ...
RESULTS
Patients
2.) Abdominopelvic tumor sites
Challenges:
distal beam end in soft tissue
opposed beams
prostate patients...
RESULTS
Patients
2.) Abdominopelvic tumor sites
Results:
for abdominal tumor sites, lateral blurring due to motion was fou...
How can we get
further to reach
the goal?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
OUTLOOK
Better biological wash-out models
scanning of high dose patients (>3Gy in a single session)
high dose translates i...
OUTLOOK
In room / online imaging
Shorter / no delay between irradiation and PET imaging
Shorter data acquisition
less wash...
OUTLOOK
In room / online imaging
Online
Parodi et al “Comparison between in-beam and offline PET imaging of proton and car...
OUTLOOK
In room / online imaging
Online
Minimal delay
Geometry compromises
efficiency
Detectors are exposed to
scattered r...
OUTLOOK
In room / online imaging
In room
Parodi et al “Comparison between in-beam and offline PET imaging of proton and ca...
OUTLOOK
In room / online imaging
In room
Small delay
Patient throughput is
compromised
…
+
-
Parodi et al “Comparison betw...
Is it worth it?
MOTIVATION
METHOD goal
RESULTS phantom
patients
OUTLOOK
CONCLUSION
CONCLUSION
Proton Therapy seems to be the “standard” treatment of
the future
1993
1996
2000
2006
2007
2008
“Is it possible...
Thank you for your attention!
Upcoming SlideShare
Loading in …5
×

Outlook on a doctoral thesis at the Massachusetts General ...

852 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
852
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
15
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Outlook on a doctoral thesis at the Massachusetts General ...

  1. 1. IN VIVO IMAGING Proton Beam Range Verification With PET/CT Antje-Christin Knopf 1/3 K Parodi 2 , H Paganetti 1 , T Bortfeld 1 Siemens Medical Solutions Supports This Project 1 Department of Radiation Oncology, MGH and Harvard Medical School, Boston, MA 02114 2 Heidelberg Ion Therapy Center, Heidelberg, Germany 3 Department of Medical Physics, DKFZ Heidelberg, Germany
  2. 2. MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  3. 3. Why do we want to make that effort? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  4. 4. Optimal treatment Protons have the superior advantage of a finite range, but uncertainties compromise this advantage. MOTIVATION Tumor OAR Beam
  5. 5. Optimal treatment Since we often don’t know the uncertainties we often don’t apply the optimal treatment. Uncertainties can be up to 10 mm. To take full advantage of the superior characteristics of proton beams mm- accurate tools to monitor and control these uncertainties are needed. MOTIVATION Tumor OAR Beam Patched Beams
  6. 6. What is the idea? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  7. 7. METHOD Procedure 1. Proton Treatment at the F. H. Burr Proton Therapy Center 2. Walk the patient to the PET/CT scanner 3. PET/CT scan at a Siemens Biograph 64 PET/CT scanner
  8. 8. METHOD Nuclear reactions In this approach we do not use any radioactive tracers but positron emitters, which are produced as a by-product of irradiation with protons. Proton Neutron Positron Electron Photon
  9. 9. METHOD Data DOSE planned Dose PET ACTIVITY measured PET - dyn. - FB - IT CT planning CT PET CT
  10. 10. METHOD Data DOSE planned Dose MC Dose PET ACTIVITY measured PET - dyn. - FB - IT MC PET CT planning CT PET CT The detailed simulations of the PET signal are based on Geant4 and FLUKA Monte Carlo (MC) code.
  11. 11. What do we want to achieve with this data? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  12. 12. GOAL Dose verification difficult because: no unique correlation between dose and activity distribution patient and tissue specific activity wash-out Oelfke et al., PMB 1995
  13. 13. GOAL Dose verification difficult because: no unique correlation between dose and activity distribution patient and tissue specific activity wash-out Range verification promising because: unique correlation between dose and activity range robust range determination through gradient analysis Parodi et al., PMB 2006 Oelfke et al., PMB 1995
  14. 14. GOAL Range verification DOSE planned Dose MC Dose PET ACTIVITY measured PET - dyn. - FB - IT MC PET CT planning CT PET CT
  15. 15. GOAL measured PET planned dose Range verification match within x mm match within y mm range was correct within (x+y) mm MC dose MC PET
  16. 16. GOAL Range verification normalize
  17. 17. GOAL Range verification pointwise 20%: - sensitive to smoothing of MC profiles - sensitive to back- ground noise 50%: - sensitive to noise in the data
  18. 18. GOAL Range verification shift more robust strategy for range verifications than a pointwise comparison
  19. 19. Is that technical and physical feasible? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  20. 20. RESULTS Phantom 1.) Homogeneous phantom and simple slab phantom Parodi et al “PET/CT imaging for treatment verification after proton therapy- a study with plastic phantoms and metallic implants”, Medical Physics 2007: 34, 319-435 PMMA bone equivalent lung equivalent
  21. 21. RESULTS Phantom 1.) Homogeneous phantom and simple slab phantom Beam Parameter: Slab phantom: one field, 16cm range, 2Gy total dose Cylinder: two perpendicular fields, 15cm / 16cm range, 8Gy total dose To study: The composition and the total yield of activity that can be expected after a proton treatment PMMA bone equivalent lung equivalent Parodi et al “PET/CT imaging for treatment verification after proton therapy- a study with plastic phantoms and metallic implants”, Medical Physics 2007: 34, 319-435
  22. 22. RESULTS Phantom 1.) Homogeneous phantom and simple slab phantom Results: Activity composition: Main fraction from 11 C, minor traces from 13 N and 15 O Imaging protocol: For a usual treatment fraction (1-3 Gy) and a delay of about 15 min between treatment and PET imaging 30 min of data acquisition should be sufficient for a mm accurate range monitoring. Parodi et al “PET/CT imaging for treatment verification after proton therapy- a study with plastic phantoms and metallic implants”, Medical Physics 2007: 34, 319-435
  23. 23. RESULTS Knopf et al “Quantitative assessment of the physical-potential of proton beam range verification with PET/CT”, submitted Phantom 2.) Complex inhomogenous phantom with different angled tissue interfaces Interfaces: 6° bone/air 0° bone/air 6° bone/lung 0° lung/air 6° lung/air 0° bone/lung PMMA bone equivalent lung equivalent
  24. 24. RESULTS Phantom 2.) Complex inhomogenous phantom with different angled tissue interfaces Beam Parameters: One field, 15 cm range, 8 Gy total dose same routine as for patients was performed To study: The reproducibility of the method The consistency of the method The sensibility of the method Interfaces: 6° bone/air 0° bone/air 6° bone/lung 0° lung/air 6° lung/air 0° bone/lung PMMA bone equivalent lung equivalent Knopf et al “Quantitative assessment of the physical-potential of proton beam range verification with PET/CT”, submitted
  25. 25. RESULTS Phantom 2.) Complex inhomogenous phantom with different angled tissue interfaces Results: Physical feasibilities: Reproducibility of range values within 1mm standard deviation Consistent range determination within 1 mm standard deviation PET measurements are sensitive enough to detect millimeter range changes induced by small tissue inhomogeneities. Knopf et al “Quantitative assessment of the physical-potential of proton beam range verification with PET/CT”, submitted
  26. 26. How does it look in clinical reality? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  27. 27. RESULTS Patients # of patients # of patients that received 1 field # of patients that received 2 fields dose per field [GyE] head 11 3 8 0.9-3 eye 1 1 10 C-spine 1 1 1 T-spine 2 2 0.6-1.8 L-spine 2 2 2 sacrum 2 1 1 1-2 prostate 2 2 2 TOTAL 21 9 12 0.6-10
  28. 28. RESULTS Patients 1.) Head and neck tumor sites Parodi et al “Patient study on in–vivo verification of beam delivery and range using PET/CT imaging after proton therapy” Int. Journal of Radiation Oncology, Biology, Physics 2007
  29. 29. RESULTS Patients 1.) Head and neck tumor sites Advantages: few patient motion -> the same immobilization as during the treatment is used rigid target geometry -> small differences in the positioning are taken into account by coregisting planning and PET CT few different tissues -> tissues can be resolved by means of CT numbers -> tissue specific elemental compositions and biol. washout parameters can be assigned in the simulation Planning CT PET CT fat brain bone Parodi et al “Patient study on in–vivo verification of beam delivery and range using PET/CT imaging after proton therapy” Int. Journal of Radiation Oncology, Biology, Physics 2007
  30. 30. RESULTS Patients 1.) Head and neck tumor sites Data analysis: At positions where the beam stopped in bone Parodi et al “Patient study on in–vivo verification of beam delivery and range using PET/CT imaging after proton therapy” Int. Journal of Radiation Oncology, Biology, Physics 2007
  31. 31. RESULTS Patients 1.) Head and neck tumor sites Data analysis: At positions where the beam stopped shortly behind in bone Parodi et al “Patient study on in–vivo verification of beam delivery and range using PET/CT imaging after proton therapy” Int. Journal of Radiation Oncology, Biology, Physics 2007
  32. 32. RESULTS Patients 1.) Head and neck tumor sites Data analysis: At positions where the beam stopped in soft tissue Parodi et al “Patient study on in–vivo verification of beam delivery and range using PET/CT imaging after proton therapy” Int. Journal of Radiation Oncology, Biology, Physics 2007
  33. 33. RESULTS Patients 1.) Head and neck tumor sites Results: In soft tissue biological washout effects degrade the measured activity distribution and therefore prevent mm-accurate offline PET/CT range verification. However offline PET/CT scans permit mm-accurate range verification in well-coregistered bony structures. Number of profiles Mean agreement between measured and simulated range [mm] pointwise verification shift verification 20 % 50% Bone 25 2.5 1.2 2.4 Bone/soft tissue 15 3.8 8.6 2.4 Soft tissue 30 6.8 3.9 4.3
  34. 34. RESULTS Patients 2.) Abdominopelvic tumor sites
  35. 35. RESULTS Patients 2.) Abdominopelvic tumor sites Challenges: motion -> breathing and organ motion results in a blurring of the measured activity distribution demanding positioning complex tissue heterogeneities -> tissues like bladder, bone marrow and muscle with very different elemental compositions and washout characteristics can not be resolved by CT numbers bladder bone marrow muscle
  36. 36. RESULTS Patients 2.) Abdominopelvic tumor sites Challenges: distal beam end in soft tissue opposed beams prostate patients need to void their bladder between treatment and imaging
  37. 37. RESULTS Patients 2.) Abdominopelvic tumor sites Results: for abdominal tumor sites, lateral blurring due to motion was fount to be up to 25mm where as the lateral conformity for head and neck tumor sides was within 5mm For opposed treatment beams range verification was found to be not practicable. In abdominal tumor sites, mm-accurate offline PET/CT range verification is not feasible primarily due to patient motion and the position of the distal beam edge in soft tissue.
  38. 38. How can we get further to reach the goal? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  39. 39. OUTLOOK Better biological wash-out models scanning of high dose patients (>3Gy in a single session) high dose translates into an enhanced positron emission enables a time analysis of the PET distribution over the 30 min of data acquisition improved biological wash-out models estimate of the improvement of the image quality for an in room PET/CT scanner Measured activity averaged over First 2 min First 10min First 20 min
  40. 40. OUTLOOK In room / online imaging Shorter / no delay between irradiation and PET imaging Shorter data acquisition less wash-out better statistics less motion online In room Parodi et al “Comparison between in-beam and offline PET imaging of proton and carbon ion therapeutic irradiation at cyclotron and synchrotron-based facilities, in press
  41. 41. OUTLOOK In room / online imaging Online Parodi et al “Comparison between in-beam and offline PET imaging of proton and carbon ion therapeutic irradiation at cyclotron and synchrotron-based facilities, in press
  42. 42. OUTLOOK In room / online imaging Online Minimal delay Geometry compromises efficiency Detectors are exposed to scattered radiation Patient throughput is compromised … + - - - Parodi et al “Comparison between in-beam and offline PET imaging of proton and carbon ion therapeutic irradiation at cyclotron and synchrotron-based facilities, in press
  43. 43. OUTLOOK In room / online imaging In room Parodi et al “Comparison between in-beam and offline PET imaging of proton and carbon ion therapeutic irradiation at cyclotron and synchrotron-based facilities, in press
  44. 44. OUTLOOK In room / online imaging In room Small delay Patient throughput is compromised … + - Parodi et al “Comparison between in-beam and offline PET imaging of proton and carbon ion therapeutic irradiation at cyclotron and synchrotron-based facilities, in press
  45. 45. Is it worth it? MOTIVATION METHOD goal RESULTS phantom patients OUTLOOK CONCLUSION
  46. 46. CONCLUSION Proton Therapy seems to be the “standard” treatment of the future 1993 1996 2000 2006 2007 2008 “Is it possible to verify directly a proton-treatment plan using positron emission tomography?” UCL-Cliniques Universitaires St-Luc, Brussels, Belgium “Proton dose monitoring with PET: quantitative studies in Lucite” TRIUMF, Batho Biomedical Facility, Vancouver, Canada “Potential application of PET in quality assurance of proton therapy” Forschungszentrum Rossendorf, Dresden, Germany “Dose-volume delivery guided proton therapy using beam on-line PET system” National Cancer Center, Kashiwa, Japan “Patient study of in vivo verification of beam delivery and range, using positron emission tomography and computed tomography imaging after proton therapy” Department of Radiation Oncology, MGH, Boston, USA “Experimental validation of the filtering approach for dose monitoring in proton therapy at low energy’ Department of Physics, University of Pisa, Italy
  47. 47. Thank you for your attention!

×