Post on 15-Jun-2020
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
1The screen versions of these slides have full details of copyright and acknowledgements
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
1
David A. Mankoff, MD, PhDSeattle Cancer Care Alliance
University of WashingtonSeattle, WA
Molecular imaging to guide therapy: outline
• Clinical/biological questions and methods to address them
• Cancer imaging to guide therapy
2
Cancer imaging to guide therapy
– Choosing the right patients
– Choosing the right drug
– Getting the right result
– Predicting outcome
Anatomic vs. functional imaging
• Anatomic imaging
– Relies on tumor size, shape, density
• e.g., mammography, CT
– Measures response by changes in size
3
• Functional/molecular imaging
– Relies on in vivo tumor biology: perfusion, metabolism, molecular features
• e.g., MRI, PET
– Measures response by changes in functional/molecular processes
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
2The screen versions of these slides have full details of copyright and acknowledgements
Functional imaging modalities
• Magnetic Resonance (MR)
– Magnetic Resonance Imaging (MRI)
– Magnetic Resonance Spectroscopy (MRS)
• Radionuclide imaging
4
g g
– Positron Emission Tomography (PET)
– Single-Photon Emission Computed Tomography (SPECT)
• Ultrasound (U/S)
• Optical imaging
Positron Emission Tomography (PET)
5
PET/CTCombines molecular and anatomical imaging
6Alessio, Rad Clin N. Amer 42:1017, 2004
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
3The screen versions of these slides have full details of copyright and acknowledgements
Why radiotracer imaging?Answer: To achieve tracer conditions
• Example: Estrogen receptor imaging
– Tracer specific activity 1000 mCi/umol
– Injected activity dose: 5 mCi
– Injected molar dose: 5 nmol
7
– Peak blood concentration: 1 nM
(Typical estradiol blood concentration is nM)
• Radiographic, MR, or optical agents require mM (factor of 106 difference!)
• Therefore - can image biochemical processes without disturbing them
Existing cancer imaging paradigm:Targets for detecting tumor cells
Higher in tumor than normal tissue
Protein synthesis
Amino acids
DNA synthesisThymidine & analogs
Energy metabolism
FDG, acetate
8
Membrane synthesisCholine, acetate
Blood flow
Water, sestamibi
FDG PET detects internal mammary nodal metastases in locally advanced breast cancer
FDG PET Emission
CT
9
IM nodeAxillary node
Bellon, Am J Clin Oncol 27:407,2004
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
4The screen versions of these slides have full details of copyright and acknowledgements
A new paradigm for cancer imaging:Help direct cancer treatment
• New role for imaging:
– Guide cancer treatment selection
10
– Evaluate early treatment response
Imaging and targeted therapyHelp match therapy to tumor biology
• Goals in cancer treatment
– Characterize tumor biology pre-Rx
– Individualized, specific therapy
Static response may be acceptable
11
– Static response may be acceptable
• The implied needs for cancer imaging
– Characterize in vivo tumor biology - predict behavior
– Identify targets, predict response
– Identify resistance mechanisms
– Measure tumor response (early!)
Emerging cancer imaging paradigm:Measure factors affecting response
Variable levels in tumor
Surface receptorsOctreotide
Proliferative rateThymidine & analogs
Glycolytic rateFDG
Hypoxia
12
ypFMISO, ATSM
Drug transportMIBI, Verapamil,
F-Paclitaxel
Nuclear receptors
FES, FDHT
AngiogenesisWater
RGD peptides
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
5The screen versions of these slides have full details of copyright and acknowledgements
Imaging requirement for biomarker imaging:Simultaneously localize and characterize disease sites
PET/CT
FESFDG
Functional/anatomic imaging
Functional imaging combinations
13
FDG PET
PET/CT fusion
Glucose metabolism
Estradiol binding
Imaging requirement for biomarker imaging:Image acquisition and quantitative analysis
Time
Dynamic imaging Region-of-interest analysis
Time-activity curves
TumorVentricle
• Dynamic protocols– Allows kinetic modeling
– Full range of analysis options
14
Blood
TissueParameter estimates
Kinetic modeling
Inject tracer
Static imageStatic
Uptake Measure
(SUV)
p
– But… not for everyone
• Static protocols– Clinically feasible,
robust
– But… only simple quantification possible
Imaging and cancer therapy Novel approaches to biomarker imaging
• Choosing the right patients
– Is the therapeutic target present?
• Choosing the right drug
Does the drug reach the target?
15
– Does the drug reach the target?
• Getting the right result
– Is there a pharmacodynamic response?
• Predicting the outcome
– Will response lead to better patient survival?
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
6The screen versions of these slides have full details of copyright and acknowledgements
ER expression and breast cancer endocrine therapy
• Endocrine therapy for breast cancer
– SERMs - e.g., tamoxifen
– Aromatase inhibitors - e.g., letrozole
16
– SERDs - e.g., fulvestrant
• ER as predictive assay
– ER-: Response rate < 5%
– ER+: Response rate 50%
• ER+/PR+: Response rate 75%
[F-18]-fluoroestradiol (FES):PET estrogen receptor (ER) imaging
FES Estradiol
*
17Kieswetter, J Nucl Med 25:1212, 1984 Tewson, Nucl Med Biol 26:905, 1999
Relative binding (FES vs. estradiol)
ER 0.9
SHBG 0.2 - 0.8
Validation: ER+ vs. ER- tumorsFDG FES
ER-
axial
coronal
FDGFES
18
ER+Liver
Glucose metabolism ER expression
axial
coronal FES
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
7The screen versions of these slides have full details of copyright and acknowledgements
vs. IHC
18F-Fluoroestradiol (FES):PET estrogen receptor (ER) imaging
Provides a quantitative estimate of ER expression
*
19Peterson, J Nucl Med 49:
367, 2008
Kieswetter, J Nucl Med 25: 1212, 1984
• In a study by Mintun, Radiology 169: 45, 1988 it was shown that the radioligand binding technique measured by PET imaging correlated with assays in vivo
FES uptake predicts breast cancer response to hormonal therapy
Pre-Rx Post-Rx• Recurrent
sternal lesion• ER+ primary• Recurrent Dz
Excellent responseafter 6 wks Letrozole
Example 1
20
FES FDG FDG• Newly Dx’d
met breast CA• ER+ primary• FES-negative
bone mets
No response to several different
hormonal Rx’s
University of Washington
strongly FES+
Example 2
Linden, J Clin Onc 24:2793, 2006
FES uptake predicts response of advanced breast cancer to hormonal therapy
LABC or metastatic Br CAPrimary tamoxifen Rx
Recurrent or metastatic Br CAAromatase inhibitor Rx
21Mortimer, J Clin Oncol 19:
2797, 2001(P < .01 for both)
Non-respondersResponders
FES
SUV
Linden, J Clin Oncol 24: 2793, 2006
FES
SUV
Responders Non-responders
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
8The screen versions of these slides have full details of copyright and acknowledgements
Imaging hypoxia as the accumulation of a radiopharmaceutical
22University of Washington KA Krohn 22
MRI FMISO PET
Hypoxic
FMISO PET predicts outcome for GBM patientsSpence, Clin Cancer Res 14:2623, 2008
ng
23
Not Hypoxic
University of Washington
(FMISO not hot)
(FMISO hot)Frac
tion
surv
ivin
Hypoxia as a target for radiotherapy planningHead/neck CA treatment planning including FMISO PET
Rajendran, Eur J Nucl Med Mol Imaging 33: S44, 2006
24
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
9The screen versions of these slides have full details of copyright and acknowledgements
Imaging to direct hypoxia-specific treatmentRischin, J Clin Oncol 24:2098, 2006
• Advanced H & N Ca• Randomized to
– XRT + Cisplatin/5-FU– XRT + Cisplatin/Tirapazamine (TPZ)
• FMISO PET (observational only)
Time-to-locoregional failure
FMISO+/TPZ
25
FDG PET
FMISO PET
FMISO+/5FU
Loca
l fai
lure
free
(%)
Years from random assignment
Cis-FU/noCis-FU/yes
Cis-Tpz/yesCis-Tpz/no
Imaging and cancer therapy Novel approaches to biomarker imaging
• Choosing the right patients
– Is the therapeutic target present?
• Choosing the right drug
26
– Does the drug reach the target?
• Getting the right result
– Is there a pharmacodynamic response?
• Predicting the outcome
– Will response lead to better patient survival?
Resistance due to altered drug transport:11C-Verapamil PET to measure P-gp drug transport
P-gp
Hypotheses:• P-gp limits drug transport into the brain• Inhibiting P-gp will increase brain transport
27
P gpP-gp susceptible drug
11C-Verapamil
P-gpP-gp susceptible drug
11C-Verapamil xinhibitorHendrikse, Cancer Res 24: 2411, 1999
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
10The screen versions of these slides have full details of copyright and acknowledgements
Imaging P-gp activity in vivo in humans[11C]-Verapamil images pre- and post-cyclosporine (CSA)
P-gp P-gpx
Verapamil PET
Infuse cyclosporineVerapamil PET
Pre-CsA Post-CsAMRI
28Sasongko, Clin Pharm Ther 77: 503, 2005 Link/Unadkat Labs
Pre CsA Post CsA
88% +/- 20% increase in verapamil AUC (N= 12, P < .001)
MRI
Liver
Tumor
Pre-Rx
Tamoxifen Letrozole Fulvestrant
Serial FES PET measures endocrine therapy pharmacodynamics
29
Post-Rx
Linden, SABCS 2007
Imaging and cancer therapy Novel approaches to biomarker imaging
• Choosing the right patients
– Is the therapeutic target present?
• Choosing the right drug
30
– Does the drug reach the target?
• Getting the right result
– Is there a pharmacodynamic response?
• Predicting the outcome
– Will response lead to better patient survival?
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
11The screen versions of these slides have full details of copyright and acknowledgements
Biologic events in response to successful cancer therapy
Rationale for measuring early response by cell proliferation imaging
Cellular proliferationor
Rx
DNA synthesis
31
orCell death
Viable cell number
Tumor size
Thymidine incorporation pathwaysImaging tumor proliferation
32Mankoff & Eary, Clin Cancer Res14: 7159, 2008
Thymidine(proliferation)
Post-RxPre-Rx
Small cell lung cancer:PET imaging pre- and post one cycle of Rx
33
FDG(glucose
metabolism)
(p )Marrow (with mets)
Tumor
7 daysShields, J Nucl Med 39: 1757, 1998
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
12The screen versions of these slides have full details of copyright and acknowledgements
Thymidine analogs for PET cell proliferation imaging
Clinically feasible isotope and imaging protocol
18F-fluoro-L-thymidine (FLT)
FLT PET images of lung cancer
34Shields AF, from Mankoff, Shields and Krohn
Rad Clin N Amer 43:153, 2005Grierson, Nucl Med Biol
27:143, 2000
Early response measured by 18F-fluorothymidine (FLT) PET
Breast CA, ChemoRx(Kenny, EJNMMI 34: 1339, 2007)
Lung CA, Genfitinib Rx(Sohn, Clin Cancer Res 14: 7423, 2008)
Pre-Rx 1 wk Rx
35
SUV
max
Responders Nonresponders
Day 0 Day 7 Day 0 Day 7
SUV
max
FLT brain tumor imaging to measure response: kinetic analysis
Muzi, J Nucl Med 47: 1612, 2006; Spence, Mol Imag Biol 11: 343, 2009
Kinetic model: Parametric imaging:
36
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
13The screen versions of these slides have full details of copyright and acknowledgements
Bone metastasis response monitoring Bone scan: response? - ???
Pre-Rx Post-Rx
37
Bone metastasis response monitoring (2)
Pre-Rx Post-Rx
FDG PET: response? - Yes!
38
Change in FDG SUV correlates with response and change in tumor markers
39Stafford, Acad Radiol 9:913, 2002
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
14The screen versions of these slides have full details of copyright and acknowledgements
% Decline in FDG SUV Initial SUV
< 5.1
Specht, Br Ca Res Treat 105: 87, 2007
FDG uptake predicts outcome of bone-dominant breast cancer
Time to skeletal-related eventTime to progression
40
> 41%
< 41%
> 5.1
FDG PET/CT to monitor breast cancer bone metastasis response
Tateishi, Radiology 247: 189, 2008
Response = ‘d FDG on PET and ‘d sclerosis on CT
• The baseline FDG PET image shows considerably high uptake in bone metastasis
41
uptake in bone metastasis
• Post therapy shows a dramatic decrease in uptake of FDG, indicating successful therapy
• The combination of a decline in SUV and a change from a lytic to a more sporadic appearance on CT was very strongly predictive of response
FDG PET vs. bone scanUptake depends upon nature of metastases
FDG
Lytic Sclerotic Mixed
42
Fluoride Bone Scan
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
15The screen versions of these slides have full details of copyright and acknowledgements
Imaging and cancer therapy Novel approaches to biomarker imaging
• Choosing the right patients
– Is the therapeutic target present?
• Choosing the right drug
43
– Does the drug reach the target?
• Getting the right result
– Is there a pharmacodynamic response?
• Predicting the outcome
– Will response lead to better patient survival?
Neo-adjuvant therapy of breast cancerPET assessment of response
SurvivalDiagnosis Surgery
4-6 months
Systemic Therapy
44
Baseline
g Surgery (Response)
Systemic Therapy
Mid-TherapyPET scans:
FDG PET predicts survival after neo-adjuvant breast cancer therapy
Post-therapy uptake, disease-free survival (DFS)
Mid-therapy uptake change, overall survival (OS)
Uptake > median
45
Uptake < median
(P < 0.05)(P < 0.01)
Emmering, Annals Oncol 19: 1573, 2008 Dunnwald, J Clin Oncol 26: 4449, 2008
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
16The screen versions of these slides have full details of copyright and acknowledgements
Molecular imaging is a tool for measuring in vivo biology
Genes Gene expression Proteins
Protein function
mRNA
46
Sequencing
Gene expression arrays
Immunohisto-chemistry
Molecular imaging
mRNA
Changes in breast cancer metabolism and blood flow predict pathologic response
Mankoff, J Nucl Med 44: 1806, 2003
• Decreased metabolism• Decreased blood flow• CR, alive
• Decreased metabolism• Increased blood flow • PR, died
47
HeartTumor
Results: multivariate DFSModels adjusted for age, ER, PR, tumor stage,
grade, histology, path response
Characteristic HR CI PLR Test
48*Also adjusted for axillary node status
Dunnwald, J Clin Oncol 26:4449, 2008
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
17The screen versions of these slides have full details of copyright and acknowledgements
Breast cancer flow-metabolism mismatch predicts poor response and survival
Mankoff, J Nucl Med 43:500, 2002; Dunnwald, J Clin Oncol 26:4449, 2008
49
*Low ratio (match) predicts path CR* High ratio (mismatch) predicts early relapse
Mankoff, J Nucl Med 43:500, 2002; Dunnwald, J Clin Oncol 26:4449, 2008
FactorP-value for association with relapse
> 4 + axillary nodes
0.10
FactorP-value for association
with path CR
High tumor grade
0.09
Match predicts good response: Mismatch predicts bad outcome:
50
nodes
< Path CR 0.75
ER-negative 0.36
HER2+ 0.66
High MRFDG/flow
0.007
grade
ER-negative 0.09
High prolif. index 0.74
Tumor size 0.82
Low MRFDG/flow
0.003
Metabolism-flow mismatch predicts poor outcome for a variety of tumors & treatments
Miles, Cancer Imaging 8:81, 2008
Cancer Imaging methods Mismatch associated with:
Breast (Mankoff, 2002) FDG/water PET poor response, early relapse
Lung (Hoekstra, 2002) FDG/water PET advanced tumors
51
Breast (Semple, 2006) FDG PET, MRI poor response
Lung (Miles, 2006) FDG PET, Perf CT large tumors
Liver metastases (Williams, 2007)
FDG PET, Perf CT large tumors
Head/neck (Hirasawa, 2007)
FDG PET, Perf CT large tumors
Pancreatic (Komar, 2009) FDG/water PET poor survival
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
18The screen versions of these slides have full details of copyright and acknowledgements
Altered metabolism is associated with myocardial viability
Tillisch, Brunken, Marshall, Schwaiger, Mandelkern, Phelps, Schelbert, NEJM 314(14):884, 1986
Glucose metabolism Blood flow
52
• Metabolism/flow mismatch
• Predicts recovered function after re-vascularization
Glycolysis and angiogenesis are coupled through a variety of mechanisms
Growth factor activation
Uncontrolled growth
Hypoxia
53Baselga, Oncologist, 2002; Semenza, Nat Rev Cancer, 2003
Gordan, Cancer Cell 12: 109, 2007; Morrish, Oncogene 28:2485, 2009
AngiogenesisProliferation Avoidance of
apoptosis
Increased glycolysis
Hypoxia
Increased glycolysis
Angiogenesisc-myc
HIF-1
Future directions
54
Clinical trial of novel molecular imaging probes for cancer
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
19The screen versions of these slides have full details of copyright and acknowledgements
ACRIN experimental imaging sciences committee (EISC)
55
Acknowledgements: UW PET cancer P01P01CA42045, S10 RR17229, Ken Krohn, PI
• Project 1 - Brain Tumors– Alex Spence
• Peoject 2 - Lymphoma– Janet Eary, Ollie Press
• Project 3 - Head
• Core A - Radiochemistry– Jeanne Link
• Core B - Physics– Tom Lewellen, Paul Kinahan
• Core C - Data analysis
56
and Neck Cancer– Joseph Rajendran
• Project 4 - Breast Cancer– David Mankoff, Hannah Linden
• Project 5 - Sarcoma– Chappie Conrad, Janet Eary
• Project 6- Endocrine Tumors– Jeanne Link, Gary Mann
– Mark Muzi, Finbarr O’Sullivan
• Core D - Molecular Pathology– Jonathan Tait, Kevin Yagle
UW/SCCA nuclear medicine breast cancer imaging research group
Robert DootLisa Dunnwald
NIH Grants CA72064, CA12457
57
Lanell PetersonErin Schubert
Larry MacDonaldMark Muzi
Clinical Applications of Molecular Imaging:Imaging to Guide Cancer Therapy
David A. Mankoff, MD, PhD
20The screen versions of these slides have full details of copyright and acknowledgements
And many other collaborators…Oncology
Julie GralowHannah LindenRobert LivingstonJason RockhillJennifer SpechtAlex SpenceE Y
RadiologyPeter EbyBill EubankJim FinkConnie LehmanJean LeeSavannah Partridge
58
Evan YuBiostatistics
Bill BarlowBrenda Kurland
PathologyKim AllisonAllen Gown (Phenopath)Tom LawtonPeggy Porter (FHCRC)
OtherNora DisisSara EyalDavid HockenberryLupe SalazarJash Unadkat
…And thanks to the audience
59