Transcript Imaging evaluation of clinical benefit in sarcomas: Dynamic MRI

Imaging evaluation of clinical
benefit in sarcomas: Dynamic MRI
Dr Anwar Padhani
[email protected]
Mount Vernon Cancer Centre
London
Montreal November 2004
Mount Vernon Cancer Centre &
Gray Cancer Institute
Jane Taylor, James Stirling
Gordon Rustin, Sue Galbraith, Kate Lankester,
Andreas Makris, Mei-Lin Ah-See
Ross Maxwell, Gill Tozer
Royal Marsden Hospital &
Institute of Cancer Research
Janet Husband and Martin Leach, David Collins,
James d’Arcy, Simon Walker-Samuel, Carmel
Hayes, Geoff Parker, John Suckling, Ian Judson
I acknowledge other contributors who have provided additional materials of their work in support of this lecture
Dr H Choi, MD Andersen Cancer Cemtre, Houston
Dr WE Reddick, St Jude Children Research Hospital, Memphis
Talk outline


Dynamic MRI – biological basis & quantification
Illustrate utility of dynamic MRI to assess benefit
of therapy in patients with bone sarcomas
– Predict response to neoadjuvant chemotherapy
– Assess activity of residual disease


Biomarker for assessing effects of treatment with
antiangiogenesis/vascular targeting drugs
Biomedical challenges in clinical implementation
specific to patients with sarcomas
Perfusion MR imaging of extracranial tumor angiogenesis. A DzikJurasz, AR Padhani. Top Magn Reson Imaging. 2004;15(1):41-57.
Talk outline


Dynamic MRI – biological basis & quantification
Illustrate utility of dynamic MRI to assess benefit
of therapy in patients with bone sarcomas
– Predict response to neoadjuvant chemotherapy
– Assess activity of residual disease


Biomarker for assessing effects of treatment with
antiangiogenesis/vascular targeting drugs
Biomedical challenges in clinical implementation
specific to patients with sarcomas
Perfusion MR imaging of extracranial tumor angiogenesis. A DzikJurasz, AR Padhani. Top Magn Reson Imaging. 2004;15(1):41-57.
Dynamic contrast enhanced MRI
(DCE-MRI)

Technique where
enhancement of a tissue or
organ is continuously
monitored using MRI after
bolus IV contrast medium
– Low molecular weight contrast
media (<1 kDa)
– Diffuse into extravascularextracellular space (does not
cross cell membranes)
– Experiment lasts a few minutes
7 minutes
Haemangiopericytoma
Data courtesy of David Collins and Ian
Judson, Institute of cancer Research,
London
Basis of dynamic contrast enhanced MRI
T2*W DCE-MRI of Mixed Mullerian Tumour
Typical acquisition 1-2 mins
T1W DCE-MRI of Mixed Mullerian Tumour
Typical acquisition 5-8 mins
T2*W versus T1W DCE-MRI
Evaluation of signal enhancement
during DCE-MRI
 Qualitative - shape of signal intensity (SI) data
curve
 Semi-quantitative - indices that describe one or
more parts of SI or [Gd] curves
 Upslope gradient, max amplitude, washout rate or area
under curve at a fixed time point
 True quantitative - indices from contrast medium
concentration changes using pharmacokinetic
modelling
Patterns of enhancement on T1W DCEMRI and histological correlates
Type I
Type II
Type III
(semi-necrotic with
reactive changes)
(viable tumour)
(rapidly proliferating
tumour edge)
kep
(min-1)
= 0.5
kep
(min-1)
= 3.4
kep
(min-1)
(Taylor and Reddick, Adv Drug Del Rev, 2000)
= 8.9
Pharmacokinetic modelling of
T1W DCE-MRI data



Transfer constant (Ktrans)
Extracellular leakage space
(ve) assumed for modelling
Figure cc. Compartments
epmethods*
Rate constant (kep)
K
k 
ve
Bolus
injection of
Contrast
medium
trans
Whole body
extracellular
space
Blood plasma
Ktrans
kep
Tumour
extracellular
space (ve)
Renal
Excretion
Modified from Tofts 1995
Quantitative analysis with
pharmacokinetic modelling

Advantages
– Whole curve shape is analysed
– Biologically relevant physiological parameters
– Independent of scanner strength, manufacturer and
imaging routines
– Enables valid comparisons of serial measurements and
data exchange between different imaging centres

Disadvantages
– Data acquisition and analysis is more complex
– Lack of commercial software for analysis
– Models may not fit the data observed
Clinical indications for DCE-MRI in
patients with musculoskeletal lesions
To improve characterisation of lesions*
 Monitoring response to treatment

– Conventional treatments
(chemotherapy/physical treatments)
– Novel biological treatments including
antiangiogenic/vascular targeting drugs

Assess activity of residual disease after
definitive treatment
*Ma LD, et al. Radiology 1997; 202(3):739-44
*van der Woude HJ et al. Radiology 1998; 208(3):821-8
*Verstraete KL, Radiology. 1994; 192(3):835-43
Importance of predicting early
tumour response to chemotherapy

If pathological response can be reliably
predicted after a few cycles of neoadjuvant
chemotherapy
– Treatment regimen could be adjusted (early surgery,
cryotherapy, isolated limb perfusion etc)


Pathological response rates may be improved
Changing treatment could increase expense
and exposes patients to greater toxicity
Good response to treatment (99% necrosis)
Baseline
120.00
SI (Baseline Corrected)
100.00
80.00
60.00
40.00
20.00
0.00
0
50000
100000
150000
200000
250000
300000
-20.00
Time (ms)
2 months on treatment
80.00
SUV 13.0
SI (Baseline Corrected)
70.00
60.00
50.00
40.00
30.00
20.00
FDG-PET scans
10.00
0.00
-10.00
0
50000
100000
150000
200000
250000
Time (ms)
Pre-operative
35.00
SUV 2.4
2A
SI (Baseline Corrected)
30.00
25.00
20.00
15.00
10.00
5.00
0.00
-5.00
0
50000
100000
150000
200000
250000
Time (ms)
Courtesy of Dr H Choi, MD Andersen Cancer Center, Houston
Correlation of DCEMRI and necrotic
fraction after
chemotherapy
Dyke JP, et al. Radiology 2003; 228:271-278
Disease-free Survival (%)
Tumors < 56 cm2
100
Prognostic value
of DCE-MRI in
osteosarcomas
kep < 1.167 min-1
80
kep  1.167
min-1
60
40
20
Change in kep as a function of pre-treatment
value. Higher permeability at presentation
results in greater decreases with therapy
0
0
1
2
3
5
4
6
100
80
kep < 1.167 min-1
60
kep > 1.167 min-1
40
20
P = 0.05
0
kep During Therapy (min -1)
Disease-free Survival (%)
Tumors > 56 cm2
2
0
-2
-4
-6
0
1
2
3
Year
4
5
6
Disease free survival for 31 patients stratified by
tumour size and DCE-MRI after 9 weeks of Rx;
0
1
2
3
4
5
6
kep at Presentation (min-1)
Reddick WE, et al. Cancer 2001; 91:2230-2237
7
Disease-free Survival (%)
Tumors < 56 cm2
100
Prognostic value
of DCE-MRI in
osteosarcomas
kep < 1.167 min-1
80
kep  1.167
min-1
60
40
20
Change in kep as a function of pre-treatment
value. Higher permeability at presentation
results in greater decreases with therapy
0
0
1
2
3
5
4
6
100
80
kep < 1.167 min-1
60
kep > 1.167 min-1
40
20
P = 0.05
0
kep During Therapy (min -1)
Disease-free Survival (%)
Tumors > 56 cm2
2
0
-2
-4
-6
0
1
2
3
Year
4
5
6
Disease free survival for 31 patients stratified by
tumour size and DCE-MRI after 9 weeks of Rx;
0
1
2
3
4
5
6
kep at Presentation (min-1)
Reddick WE, et al. Cancer 2001; 91:2230-2237
7
Poor access to contrast before treatment
Baseline
40.00
SI (Baseline Corrected)
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
-5.00
0
50000
100000
150000
200000
250000
300000
Tim e (m s)
SUV 5.9
FDG-PET scans
Poor response to treatment (75% necrosis)
20.00
18.00
SI (Baseline Corrected)
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-2.00 0
50000
100000
150000
200000
250000
300000
Time (ms)
Pre-operative
Courtesy of Dr H Choi, MD Andersen Cancer Center, Houston
SUV 8.3
Drugs targeting tumour neovasculature
 Permeability
 rBV
or  rBF
Probably
depends on drug
duration and
dose
Vascular targeting
drugs
Anti-VEGF
drugs
 Permeability
 rBV
rBF
Time course of Combretastatin
effects on microvasculature
IAP 10 mg/kg
2 hours post
CA4P
10 mg/kg
trans
K
100 mg/kg
125
Relative Change (%)
PreRx
trans
K
IAP 100 mg/kg
100
75
50
25
0
0
5
10
15
20
Time post treatment (hours)
Window chamber view
P22 Carcinosarcoma
B Vojnovic and G Tozer, Gray Cancer Institute
IAP - radiolabelled
iodoantipyrine
25
Morphological & kinetic changes
After 1st dose of CA4P (52mg/m2)
24 hrs
Pre
4 hrs
Post
tra n s
R e la tiv e C h a n g e K
(% )
D o s e m g /m
140
2 0 -4 0 -- -------5 2 ------- -------6 8 ------- -----8 8 ----- -----1 1 4 ----
*
Biologically active dose 52 mg/m2
120
20
0
2
DLT 114 mg/m2
MTD 88 mg/m2
*
*
*
*
-2 0
-4 0
-6 0
4 H o u rs
-8 0
-1 0 0
2 4 H o u rs
4 4 98 5 %
9 2C9 I3 f1o3r2 a3n3 i9n d2 3i v2i 5d 2u 8a 3l 0 9 1 2 1 4 1 6 1 7 1 9 2 0 2 1
P aGalbraith
t ie n t SM,
N uetmal.bJeClin
r Oncol – 2003;21:2831-42.
Galbraith SM, et al. J Clin Oncol – 2003;21:2831-42
Phase I goals and DCE-MRI
achievements in the CA4P study
Achievement
Goal
Modulation of vascular kinetics
+
Dose response relationship
+ (threshold)
Identify therapeutic window
+
Drug exposure kinetic response
relationship
+
Galbraith SM, et al. J Clin Oncol – 2003;21:2831-42
Dose response in Ki for PTK787/ZK in
colorectal cancer on Day 2
160
No maximum
tolerated
dose was
reached
140
Ki (% Baseline)
SEM bars, all
colorectal
liver
metastases
25 patients with
metastatic colon cancer
evaluated at baseline,
on day 2 and 28
120
100
80
60
40
20
0
50
300
500
750
1000
1200
Dose (mg)
Morgan, B., et al., J Clin Oncol, 2003. 21(21): p. 3955-3964.
Phase I goals and DCE-MRI
achievements in the PTK787/ZK study
Goal
Modulation of vascular kinetics
Achievement
+
Dose response relationship
+ (threshold)
Identify therapeutic window
+ (no MTD)
Drug exposure kinetic response
relationship
?
Morgan, B., et al., J Clin Oncol, 2003. 21(21): p. 3955-3964.
Conclusions




Dynamic MRI provides unique information on the
vascular characteristics of tumours
DCE-MRI can predict extent of histological
response to chemotherapy in patients with
osteosarcomas/Ewing tumours
Intriguingly, DCE-MRI may inform on drug access
(? predict responsiveness) and patient prognosis
Acts as a biomarker that provides
pharmacodynamic (PD) information in early trials
of antivascular drug and should be used for
evaluating combination therapies in sarcomas
Dynamic MR imaging of tumor perfusion: approaches and biomedical challenges.
DJ Collins, AR Padhani. IEEE Engineering in Medicine and Biology Magazine 2004