1

Functional and Molecular Imaging for Radiation Therapy Guidance

Department of Radiation OncologyStanford University School of Medicine

L Xing, T Li, Y Yang, E. Schreibmann, B Thorndyke, D. Spielman

4D modeling

Gated tx planning

4D planning

Adaptive therapy (imaging, planning, delivery)

Day 0 Day 14 Day 24

3D/4D CBCT

4D imagingBiological imaging

Imaging 3D modeling Treatment planningPt setup and treatment delivery

Targeting in current radiation oncology

Inter-fraction organ movement

Intra-fraction organ movement, in particularly, respiratory motion

Target volume definition & localization

PET/CTPET/CT

Where is the tumor?

21 September 2000

19 July 2000 18 July 2000

21 August 2000

Is tumor responding to therapy?Is tumor responding to therapy?

• Tissue Characterization and Classification

• Staging

• Restaging

• Prognosis

• Monitoring Treatment

• Tissue Characterization and Classification

• Staging

• Restaging

• Prognosis

• Monitoring Treatment

Anatomic MolecularImaging Imaging

A. Quon

Treatment plan for prostate patient with nodal risk>15%

Anterior Field Lateral Field

100%90%

50%

Axial Field

Treatment plan for prostate patient with nodal risk <15%

Pre-treatment staging & planning

2

Shortcomings of Current Radiation Therapy

• Staging and treatment decision-making are based largely on anatomical imaging.

• Tumor volume defined on CT/MRI may often be too small or too large.

• Spatial biology distribution is not considered.

• The whole Rx course takes 5-6 weeks and there is no effective tools to distinguish responders from un-responders.

• Detection of recurrence is problematic.

Where is the tumor?

What is the biology distribution?

Where is the boundary of the tumor?

Imaging 3D modeling Treatment planningPt setup and treatment delivery

4D modeling

4D BCRT planning

Adaptive therapy (imaging, planning, delivery)

Day 0 Day 14 Day 24

3D/4D CBCT4D imagingBiological imaging

IMRT provides an unprecedented means to produce customized 3D-dose distributions with sub-cm resolution.

Integration of radiological imaging techniques allows better patient positioning and dose delivery.

Functional/metabolic imaging modalities are available fornoninvasively providing critical needed metabolic and physiological data.

Molecular imaging techniques are emerging.

Integration of biological and functional Image.-- Biologically Conformal Radiation Therapy (BCRT)

----to truly individualize radiation therapy

The Current Imaging Toolbox

400,000400,0001~2 mm1~2 mmCTCT

100010000.02 mm0.02 mmOpticsOptics100,000100,000<1 mm<1 mmHFUSHFUS400,000400,0003~5 mm3~5 mmPETPET600,000600,0004~6 mm4~6 mmSPECTSPECT

1,000,0001,000,0007 mm (~3mm at 3T)7 mm (~3mm at 3T)MRSIMRSI400,000400,0001~2 mm1~2 mmMRIMRI

Minimum Detected Minimum Detected Cells (n)Cells (n)

Minimum DetectableMinimum DetectableSize (Size (φφ))MethodMethod

3

day1 day4 day8

P Maxim, S Gambhir, C Contag, and L Xing, Molecular Imaging.

Day11 after inoculation

GenomeGenome

Luciferase--Transcriptional Reporter

hνhν

Luciferase

Mg2+

O2

Luciferase

Mg2+

O2

LuciferinLuciferin

OxyluciferinOxyluciferin

ATPATP

ADP + PiADP + Pi

PromoterPromoter

LuciferaseLuciferase

LuciferaseLuciferase GenomeGenome

Project in Cellular & Molecular ImagingStanford University Medical Center (C. Contag)

Substrate& energySubstrate& energy

ExcitationExcitationEmiss

ion

Emission

Emission

EmissionFluorescenceFluorescence

BioluminescenceBioluminescence

Bench-top Small Animal Imaging System: Macroscope

Clinically available molecular/Functional imaging tools

MR Spectroscopic Imaging (MRSI)

Functional MRI, DWI/DTI

PET, SPECT

Endorectal Coil-Based 3T MR Spectroscopic Imaging for Radiotherapy

3T Endorectal Probe

MRSI: monitor the molecular properties of a tumor ER Coil----high SNR MRI/MRSIER Coil----severely distorts the prostate

Topic Example

Characterization of Tissue Solitary Pulmonary Nodules

Staging Lung Cancer

Restaging/Prognosis Lymphoma/Breast Cancer

Monitoring Treatment GIST, Breast CA, Lymphoma

Emerging Applications XRT planning

• Appropriate management decisions starts with accurate staging

• PET changes overall management in more than 30% of cases

Gambhir, et al 2001

PET and PET/CT

Dizendorf et al, JNM 44, 24-29, 2003

4

PET/CT for Radiation Therapy

Pre-treatmentPET/CT/MRI

Post-treatmentPET/CT/MRI

Planning& treatment

Registration& analysis

PET/CT/MRI-based tumor delineation

• delineate target volume • assess therapeutic response• distinguish recurrent tumor and radiation necrosis

Commonly used positron emitters

1.71.7GeneratorGenerator1.31.38282RbRb

1.71.7GeneratorGenerator1.1 hr1.1 hr6868GaGa

FDG, FLT FDG, FLT 1919F(p,pn)F(p,pn)1818FF1818O(p,n) O(p,n) 1818FF

1.01.0CyclotronCyclotron1101101818FF

1515O water O water 1414N(d,n)N(d,n)1515OO1.51.5CyclotronCyclotron221515OO

1313N ammonia N ammonia 1313C(p,n)C(p,n)1313NN1.41.4CyclotronCyclotron10101313NN

1111CC--CholineCholine 1111B(p,n)B(p,n)1111CC1.11.1CyclotronCyclotron20201111CC

Application ProductionApplication ProductionMax. range Max. range in water in water (mm)(mm)

Production Production methodmethod

HalfHalf--life life (min.)(min.)

IsotopeIsotope

FLT, 11C-choline, 18F-choline, Acetate, F-Dopa, Cu-ATSM, F-MISO, F-EF5…

Receptors and peptide-based tracers.

Proliferation imaging

Buck et al., J NuclMed, 2003

Post-treatment

Pre-treatment Post-treatment

Functional Image-Guided IMRT

• What needs to be done?

Integration---network transfer of functional image files, file format conversion, image fusion/deformable image registration….

Image fusion.

Inverse treatment planning.

Reliable imaging tools.

Clinical study of the efficacy.

Quality assurance .

5

Example: 4D PET

Patientsetup

4Dplanning

Reconselected

gates

CTAC(best at

expiration)GE gated

PET

RPMsystem

If bin size = ∆t, then in eachrespiratory cycle:

Bin 1 accumulates events from [0, ∆t]Bin 2 accumulates events from [∆t, 2∆t]Bin 3 accumulates events from [2∆t, 3∆t]...Bin N accumulates events from

[(N-1)∆t, N∆t]

A Liver Cancer patientA Liver Cancer patient

GE 4D PET Conventional 3D PETStanford new 4D

PET technique1 cm

0 5 10 15 20 25 30 35 40 45 50-2

0

2

4

6

8

10

12

14

16

craniocaudal location (mm)

CNR

UngatedGated at Insp with 1.25 secRetro. Stacked at Insp

CN

R

3D PET --- the lesion in the ungated image is elongated, and mislocalized superiorly by ~1 cm.

GET PET – location is right but signal is week.

RS 4D PET – location and signal are great☺.3D PET give wrong location and wrong volume

RS 4D PET imaging technique

coronalsagittal

Stanford 4D PET image at end-inspiration. The lesion has clearly emerged from the background activity at the end-inspiration location.

(B. Thorndyke, E. Schreibmann, A. Koogn, and L. Xing, Med Phys, 2006).

Functional Image-Guided IMRT

• What needs to be done?

Image fusion.

Inverse treatment planning.

Reliable imaging tools.

Quality assurance .

Schreibmann and Xing, Med. Phys., 2006.

Image RegistrationAutomated registration based on control volume(s). CT

scanRaw

CT data

PET scan

Raw PET data

Scatter and attenuation correction

CT images

PET images

20 min timesaving

~ 8 min per coach position

Hardware Fusion

6

Head and neckCT, PET, and IMRT treatment plan

PET

PET

CT

CT

Deformable image registrationDeformable image registration

E. Schreibmann and L Xing, IJROBP, 2006

4D Image Registration

7

Functional Image-Guided IMRT

• What needs to be done?

Image fusion.

Inverse treatment planning.

Reliable imaging tools.

Quality assurance .

Biological image based inverse treatment planning

PET/MRI/MRSI Post-scan data processing

3 SD5 SD7 SD

Quantify metabolism

Inverse planning

Prescription

Beam orientation/profile optimization

Fine-tune regional doses

Plan review

MLC leaf sequencing

IMRT Plan Incorporating MRSI data

SD=3SD=5

SD=720% 40%

40%

50%

95%

Target 3 (SD=3)

TissueTarget 1 (SD=7)

Target 2 (SD=5)

Dose (Gy)

SD=3SD=5

SD=720% 40%

40%

50%

95%

Target 3 (SD=3)

TissueTarget 1 (SD=7)

Target 2 (SD=5)

SD=3SD=5

SD=7

SD=3SD=5

SD=7

SD=3SD=5

SD=720% 40%

40%

50%

95%

Target 3 (SD=3)

TissueTarget 1 (SD=7)

Target 2 (SD=5)

Dose (Gy)

L. Xing et al, PMB 47, 3567-3578, 2002 .

Spatial distribution of biological parameters

Biological imaging

Spatially non-uniform conformal dose distribution

Biologically Conformal Radiation Therapy

Prescription for molecular/functional image guided IMRT

)]exp(exp[ 0 TDvTCP iiiiii ∆+−−= γαρ

−∆−−=

ii

rr

iir

ir

i

ri TDD

0

0ln1)(1ραρα

αγγ

ααα

Yang Y and Xing L, Med. Phys. 2005.

Functional Image-Guided IMRT

• What needs to be done?

Image fusion.

Inverse treatment planning.

Reliable imaging tools.

Quality assurance .

8

S. Hunjan, D. Kim, A. Adalsteinsson, D. Spielman, L. Xing, IJROBP 57, 1159-1173, 2003

Quality Assurance Relationship between the Cho/NAA ratios of the calibration solutionsobtained at 9.4 T versus the calibration-solution-filled vials inside the phantom obtained at 1.5 T using a 2D PRESS sequence.

S. Hunjan, D. Kim, A. Adalsteinsson, D. Spielman, L. Xing, IJROBP 57, 1159-1173, 2003

3D PET

4D PET (after post-acquisition data processing using our new algorithm)

Motion direction

-10 0 10 20 30 40 50 60 70 80-5

0

5

10

15

20

25

30

superioinferior location (mm)

CNR

UngatedGated with 0.4 secGated with 1.0 secRetro. Stacked

CN

R

Motion direction

Intensity profiles along the motion direction

4D PET Image Enhancement - Phantom Validation

(B. Thorndyke, E. Schreibmann, A. Koogn, and L. Xing, IJROBP).

B. Thorndyke, E. Schreibmann, A. Koong, L Xing, Med. Phys. 2006

Functional Image-Guided IMRT

• What needs to be done?

Image fusion.

Inverse treatment planning.

Reliable imaging tools.

Quality assurance .

A lot of research!

P Maxim, S Gambhir, C Contag, and L Xing, Molecular Imaging., 2004

Day11 after inoculation

day1day4 day8

Summary

Next Generation Radiation TherapyIntegration of functional & molecular imagingBiologically Conformal IMRT

----to truly individualize Rx

Reliable imaging tool, registration, inverse planning, QA, research.

Functional/molecular imaging is emerging.

9

Former postdocs- S. Hunjan， J. Lian, …

Clinical faculty – A. Koong, K. Goodman, Q. Le, S. Hancock, A. Quon …

Physicist – G. Luxton, T. Pawlicki, P. Maxim,….

ACKNOWLEDGEMENTS

Funding agency –NIH (1R01 CA98523 & 1R01CA104205) DOD (PC040282 )