contrast enhanced - vanderbilt university medical … enhanced sonography arthur c. fleischer, m.d....

7/8/2015

1

Clinical Applicationsof

Contrast Enhanced Sonography

Arthur C. Fleischer, M.D.

Vanderbilt U Med Ctr

Radiology and Ob/Gyn

Disclosures• These applications are real,

not just “academic”

• Microbubble contrast is currently not approved by FDA for non‐cardiac general applications, but is approved for cardiac applications

• There is an imminent chance of approval within 1‐2 years

USA is only country in world with this constraint (should be lifted within 1‐2 yrs)

Microbubbles can be used

“off label”

Research Support

• NIH/NCI grant R21 CA 125227‐01

• AIUM discovery grants

• Multiple NIH/NCI grants imminent

• Philips Healthcare‐software support

• Bracco‐contrast and software

• Lantheus‐Definity contrast

Clinical Applications

• Hepatic Masses‐DDx of metastasis, focal nodular hyperplasia, hepatocellular carcinoma, hemangioma

• Renal Masses‐evaluation of “indeterminate” renal mass, particularly in patients with poor renal function‐DDx of renal cell carcinoma, oncocytoma

• Ovarian masses‐DDx of carcinoma, fibroma, dermoid, endometrioma

• Assessment of tumor response

• Potential targeted therapy ‐“theranostics”

Other Applications• Assessment of completeness of TACE, ablation

• Prostate cancer‐localization prior to Bx

• Reflux‐urethral, ureteric

• Splenic, renal, liver laceration 2/2 trauma

• Adnexal torsion‐both depiction of vascular pedicle and “viability” of parenchyma

• Tubal patency

• Vascular complications‐endovascular shunt leaks

• Breast cancer detection (Liu JUM 34:117,’15)

Tumor angiogenesis

(according to Judah Folkman, MD, PhD)

7/8/2015

2

Tumor angiogenesis

Kuszyk, B. S. et al. Am. J. Roentgenol. 2001;177:747-753

Microbubble vs Gd‐MRI

Kuszyk, B., AJR 177:747, 2001

MB stay intravascular, whereas Gd-MRI leaks into interstitium

Scientific American, 2005

Tumor Neovascularity

Increased vascularity, clustered vessels, irregular branching pattern, irregular caliber

McDonald, Choyke Nat Med 9:713, 2003

7/8/2015

3

Microbubbles used forContrast Enhanced Sonography (CE‐US)

____5 um

Definity microbubbles

_____5 u

Microbubbles for US ContrastCapillary Perfusion

Perfusion ~ blood flow (ml/s)volume (mg)

Blood flow ~ vascular area (α) x mean blood velocity (β)

Mean blood = slice x replenishmentvelocity thickness rate(β) (α)

Contrast AgentsContrast Agents• Shell

– albumin, lipid/phospholipid or polymer

• Gas– air, low diffusivity gas

• Diameter– 2-15 microns

• Resonant frequencies - 2-10 MHz• Nonlinearity allows bubbles to be differentiated from surrounding

tissue

• Persistence – circulate for minutes

5 microns 50 microns

c/o C. Caskey

POWER

Interaction of Ultrasoundand Microbubbles

Linearresonance

Nonlinearresonance

Transientscattering

POWER POWER

Fundamentalenhancement

Bubble disruptionHarmonic enhancement

Burns. In: Rumack et al, eds. Diagnostic Ultrasound. Vol 1. 2nd ed. St. Louis: Mosby; 1998:57.

7/8/2015

4

Microbubble Destruction• At diagnostic output levels,

bubble can expand a few times original radius

• Most stabilizing coatings give way, leaving a free bubble to dissolve

• After insonification at normal imaging levels, most agents are destroyed

• This can be avoided with new sensitive imaging modes, or can be used to advantage

Animation adapted from Dr. K Ferrara, UC Davis. © Becher H and Burns PN, Handbook of Contrast

Echocardiography, Springer 2000, www.Sunnybrook.utoronto.ca/EchoHandbook/

5

Principles of Harmonic Imaging

Burns. In: Rumack et al, eds. Diagnostic Ultrasound. Vol 1. 2nd ed. St. Louis: Mosby; 1998:57.

3.0 MHz

3.0 MHz• Tissue and blood reflect at the fundamental frequency

• Microbubbles reflect at both the fundamental and the harmonic frequencies

3.0 MHz3.0 MHz + 6 MHz

Continuous vs Interval Time Delay Imaging

Continuous ImagingWith High MI

Interval Time Delay ImagingWith High MI

Continuous ImagingWith Low MI

TransmittedSequence (red have

inverted phase)

B-mode

PulseInversion

Contrast pulse

sequencing (CPS)

Method

Power Modulation

Algebra to

combine

N/A

Echo1 + Echo2

Echo1 + Echo2 + Echo3

Echo1- 2*Echo2

Summed Tissue Echo

Summed Bubble Echo

a

b

c

Caskey, C . “Leveraging the Power of Ultrasound for Therapeutic Design and Optimization” Journal of Controlled Release, 2011

Caskey, C J Ac S A, 2009 CE‐US steps

• CDS to identify region of interest

• Prepare contrast

• IV injection‐contrast + 10 cc saline

– Bolus

– Infusion for Destruction/reperfusion

• Record in cine‐loop for approx. 3 min

• Analyze enhancement parameters‐time to peak, peak enhancement, washout, AUC,microvessel perfusion

7/8/2015

5

Contrast Agent Preparation

A 3‐µL/kg dose of Definity (~0.3 mL) injected intravenously as a bolus, followed by bolus of 10 mL of 0.9% sodium chloride solution

Contrast Injection

Destruction/ReperfusionI=(1‐et) Hepatic Enhancement‐phases

• Arterial‐begins 10‐20s, ends 30‐45s

• Portal venous phase‐begins 10‐45s, ends 120s

• Late‐gr than 120 s, ends 5‐6 mins (300‐360s)

• Sustained=likely BENIGN

• Slow washout=likely MALIGNANT

• Patterns‐diffuse, nodular, peripheral

Liver Masses – Contrasted US

Hemangiomia

Focal Nodular Hyperplasia

Hepatocellular carcinoma

Metastases

Vascularity

Marginal pools

Profuse stellate pattern feeding artery

Profuse, often dysmorphic

Variable – Usually low

Arterial Phase

Peripheral nodular enhancement

Hypervascularwith scar

Hypervascular

Hypovascular

Portal Venous Phase

Central progression

Sustained enhancement

Rapid washout

Hypovascular


Baseline

Arterial Phase

Portal Phase

Late Phase

7/8/2015

6

Baseline


Arterial phase Portal phase


Late phase


Hemangioma

Portal Phase

Baseline

Arterial Phase

Late Phase

Hemangioma

Early phase

Hemangioma

Late phase

7/8/2015

7

Baseline

Arterial phase

Portal - late phase

MetastasisHCC

HCC

Early phase

HCC

Late phase

HCC

Early phase

HCC

Late phaseBaseline

Metastasis

7/8/2015

8

Metastasis

Early phase

Metastasis

Late phase

Easy to remember…

• WATCH OUT for WASH OUT!

• Rapid wash out is a sign of malignancy

• Whereas……

• Most benign lesions retain contrast in the late arterial, venous and portal venous phase

CE‐US of Liver Masses• PVP enhancement + in 92% benign, ‐ in 93% malignant; sustained PVP with arterial phase nodularity and centripetal progression in 92% of hemangioma; diffuse arterial phase enhancement gr liver in 95% of FNH (Wilson, S AJR, 186:1401, 2006)

• 92‐95% concordance with CT, MRI (Wilson, S Radiology 257: 24, 2010)

• Advantages‐Non‐toxic renal, non‐ionizing

CEUS of renal masses• Approximately 50% of people over 50 have a renal mass

• Vast majority are cysts

• Most of the solid renal masses are malignant

• Classified by Bosniak criteria‐

– B1=cyst, 0% chance of malignancy

– B2=complex cyst, 0‐5% chance of malignancy

– B2F=increased # of septations, thick wall, 5% chance of malignancy (F=followup)

– B3=thick septations, solid parts, 31‐100% chance of malignancy

– B4=solid, vascular mass, 100% chance of malignancy

7/8/2015

9

CE‐US for DDx of indeterminate renal masses

(Barr, R Radiology, 2014)

• Sensitivity=100%; specificity=95%

• PPV=95%; NPV= 100%

• 5 false +’s= 3 oncocytomas, 2 B3 cysts

• Based on 721 pts, 1018 indeterminate lesions

CE‐US in patients with compromised renal function

• Major advantage of CEUS over CT (potential nephrotoxicity of iodinated contrast)

• And MR (potential risk for nephrosclerosissyndrome)

• CEUS‐Low cost compared to CT, MR

• Readily repeatable

• The use of CEUS could decrease the needs for both CT and MRI (diminish radiation exposure)

CE‐US for ovarian Ca

• IOTA study (US O/G, 2009)‐72 adnexal masses

– AUC (0.84) better than grey scale (0.75), but less than pattern recognition (0.93)‐problem DDx of borderline vs Ca‐used parametric analysis, not time/intensity curves‐they found CEUS could not DDx b/w borderline tumor and malignancy

Fleischer, A et al (JUM,2008;AJR 2009)‐57 adnexal masses

sens.=93%; spec.=96%

max. enhancement=90% accuracy

washout/AUC=100% accuracy The time to peak (TTP) = time from injection to the peak intensity

CEUS Image Analysis

The area under the enhancement curve (AUC) calculated from the arrival of the contrast agent to the end of the wash‐out period

CEUS Image Analysis

Wash‐out (WO) = time between the peak intensity and the return to the baseline.

CEUS Image Analysis

7/8/2015

10

The area under the enhancement curve (AUC) calculated from the arrival of the contrast agent to the end of the wash‐out period

CEUS Image Analysis Paraovarian cyst

Paraovarian cystParaovarian cyst-3D TV-CDS

Cystadenoma Cystadenoma

7/8/2015

11

Peak Enhancement Time to peak

Parametric Imaging

Cystadenoma Adenocarcinoma (stage II)-r ov

Adenocarcinoma (stage II)-r ovStage II AdenoCa

Parametric map of a serous adenocarcinoma during peak enhancement (a), time to peak (b), wiAUC (c), woAUC (d), wiwoAUC (e), and woR (f). Areas of tumor neovascularity in red contrast sharply with the cystic portion of the tumor shown in blue.

Serous adenocarcinoma, Stage II

Peak Enhancement Time to peak

Parametric ImagingAdenocarcinoma (stage II)-l ov

7/8/2015

12

Adenocarcinoma (stage II)- l ov Bilateral Ov Ca (stage II)

Borderline mucinouscystadenocarcinoma




T1 T2 fat-suppressed

7/8/2015

13




Parametric Imaging

Time to peakPeak Enhancement

Fibroma

Emax = 0.41 dB T½ = 7.5 sec

AUC = 15.3 sec-1

Peak Enhancement Area under curve

Parametric Imaging

FibromaBenign (n = 38) n =

endometrioma 12serous cystadenofibroma 7hemorrhagic corpus luteum 5mucinous cystadenoma 5teratoma 3serous cystadenoma 3paratubal/paraovarian cyst 2fibroma 1

Borderline/Malignant (n = 12)serous adenocarcinoma 6endometroid carcinoma 2breast carcinoma metastatsis 2borderline mucinous cystadenocarcinoma 2

Histology of Ovarian Masses

7/8/2015

14

Enhancement parameters (n=57)

Benign Malignant0

50

100

150

200

1/2

Wa

sh-O

ut

Tim

e,

sec

Benign Malignant

500

1000

1500

2000

2500

Are

a U

nd

er

the

Cu

rve

Benign Malignant15

20

25

30

35

Tim

e to

Pea

k, s

ec

P=0.7

P<0.01

Benign Malignant

10

15

20

25

Pe

ak

En

ha

nce

me

nt,

dB

P<0.01

P<0.01

Diagnostic accuracy

70%

80%

90%

100%

2D VFI 3D VI 3D VFI Emax AUC T 1/2

Sensitivity

Specificity

Labelled MB in Animal Modelsof Ovarian Cancer

• Barua, A (Int J Gyn Ca; 24, 2014)‐labelled

– Alpha v beta 3 detected angiogenetic ov ca in hens

– Lutz, A (Cl Ca Res, 2014) CD276 targeted MB in mice

Procedural Applications

• s/p TACE‐assess completeness of ablation

• s/p RF ablation of renal masses‐assess for recurrence

• Detect abnormal sentinal nodes in breast ca

CE‐US:Tumor Response

• Lassau, N Radiology 2011‐Accurate assessment of HCC to antiangenic Rx‐predictive of tumor response, progression‐free interval, and overall survival

• Williams, R Radiology 2011‐can depict changes in vascular volume earlier than RECIST

• Pei, X unpublished, 2014‐Changes in cervical Ca with Rx

Volume = 19.1 cm3

Good responderpre-treatment

7/8/2015

15

Good responderDay 15 post treatment

Volume = 19.6 cm3

Good responder @ d0, d15

0

5

10

15

20

25

30

0 20 40 60 80 100 120 140 160 180

En

han

ce

me

nt

inte

nsi

ty,

dE

Time, sec

Pre-treatment

Day 15

0

5

10

15

20

25

30

0 1 2 3 4 5 6 7

En

han

cem

ent

inte

nsi

ty,

dE

Time, sec

Pre-treatmentDay 15

Bolus sequence Destruction-reperfusion sequence

Good responder

Pfizer Confidential – For Internal Use Only 889/2009

Patient #1 Good Responder

0

5

10

15

20

25

30

0 20 40 60 80 100 120 140 160 180

En

han

cem

en

t in

ten

sity

, d

E

Time, sec

Pre-treatment

Day 15

0

5

10

15

20

25

30

0 1 2 3 4 5 6 7

En

ha

nc

em

en

t in

ten

sit

y, d

E

Time, sec

Pre-treatmentDay 15


Baseline Day 15

Volume, cm3 19.1 19.6

Contrast Enhancement, dB 25 17

Microvascular density, dB 22.7 13.7

Blood flow velocity, 1/sec 1.1 0.7

Good responderpre-treatment

Volume = 9.3 cm3

Good responderDay 15 post treatment

Volume = 9.5 cm3

7/8/2015

16

Good responder @ d0, d15

0

5

10

15

20

0 20 40 60 80 100 120 140 160 180

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre-treatmentDay 15

0

5

10

15

20

25

30

0 2 4 6 8 10 12

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre treatmentDay 15


Good responder


Patient # 2 Good Responder

Baseline Day 15

Volume, cm3 9.3 9.5

Contrast Enhancement, dB 15 10




Poor responder @ d0, d15

0

5

10

15

20

0 20 40 60 80 100 120 140 160 180

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre-treatmentDay 15

0

5

10

15

20

25

30

0 2 4 6 8 10 12

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre treatmentDay 15


Good responder

0

5

10

15

20

0 20 40 60 80 100 120 140 160 180

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre-treatmentDay 15

0

5

10

15

20

25

30

0 2 4 6 8 10 12

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre treatmentDay 15


Good responder

7/8/2015

17

0

5

10

15

20

0 20 40 60 80 100 120 140 160 180

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre-treatmentDay 15

0

5

10

15

20

25

30

0 2 4 6 8 10 12E

nhan

cem

ent

inte

nsity

, dB

Time, sec

Pre treatmentDay 15


Good responder

0

5

10

15

20

0 20 40 60 80 100 120 140 160 180

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre-treatmentDay 15

0

5

10

15

20

25

30

0 2 4 6 8 10 12

Enh

ance

men

t in

tens

ity,

dB

Time, sec

Pre treatmentDay 15


Good responder


Poor responder @ d0, d15


Patient # 3 Poor Responder

0

1

2

3

4

5

0 2 4 6 8 10 12Enh

ance

men

t in

ten

sity

, d

B

Time, sec

Pre treatmentDay 15

Baseline Day 15

Volume, cm3 4.2 4.0

Contrast Enhancement, dB 8.9 11.3




Changes with Rx (c/o A. Lyshchik)


Barriers Affecting Rx

Kuszyk, B., AJR 177:747, 2001

7/8/2015

18

Averkiou, M. et al. UMB. 2010

Normalization of CEUS Poor responder (bolus)

Poor responder (perfusion) Fair Responder (bolus)

Fair Responder (bolus) Good Responder (perfusion)

7/8/2015

19

Excellent Responder (perfusion)

Lots of unanswered ?’s

• What defines good, fair, poor responders?

Lassau‐40% reduction in AUC

• How does this correlate with clinical response, survival?

• Multiple lesions?

• Areas of necrosis?

• Rx to decrease interstitial pressures, improve

hypoxic areas

Future Opportunities and Studies

• Perfusion quantification

• Angiogenesis/anti‐angiogenesis monitoring

• Targeted drug delivery

• Gene Therapy

Targeting LigandEncapsulatedDrug

Gas

Courtesy: Evan Unger – ImaRx

Markers for endothelial cell surface

• Inflammation (p‐selectin)

• Thrombosis (aIIb3 integrin)

• Angiogenesis (VEGF2)

Future Opportunities and Studies

• Perfusion quantification

• Angiogenesis/anti‐angiogenesis monitoring

• Targeted drug delivery

• Gene Therapy

Targeting LigandEncapsulatedDrug

Gas

Courtesy: Evan Unger – ImaRx

7/8/2015

20

Targeted US with labelled MB(c/o C. Caskey, PhD)

Inject microbubbles with targeting ligand1 Minute post-injection: some free, some bound7 minutes post-injection: only bound remain

Leong-Poi, H. et al. Circulation 2003;107:455-460

Neovessels with retained targeted microbubbles

Anti‐angiogenesis therapy

VEGF

Endothelial cell

Capillary

Angiogenesis

VEGF release

VEGF binding

anti-VEGF Ab

VEGF receptor

Genentech Inc

VEGF receptors targeting

7/8/2015

21

Imaging strategy

VisualSonics Inc

Control UCA imaging

UCA signalB-mode ultrasound

VEGFR-2 targeted UCA imaging

UCA signalB-mode ultrasound

Molecular sonograms

High VEGFR2 expression Low VEGFR2 expression

Targeted ultrasound contrast agents Targeted UCA as a drug delivery vehicle

M. Tartis et al. Ultrasound in Medicine & Biology 2006

7/8/2015

22

Jets

C. F. Caskey, S. P. Qin, P. A. Dayton, and K. W. Ferrara, "Microbubble tunneling in gel phantoms," Journal of the Acoustical Society of America 125, EL183-EL189 (2009).

Treating brain metastasis with BBB MB Disruption + DOX

0.0E+00

2.0E+05

4.0E+05

1 2 3 4

Total B

LI Counts

Weeks after Injection

ROI 1=1.463e+05ROI 2=11467

Week 3 Week 4 c/o Caskey, C

Microbubble Rx for Alzheimers(C. Jones Ctr, U Queensland‐Science Translational Med)

Rx cleared plaque-restored memory in 75% of mice w/i several wks

Conclusions

• CE‐US excellent for DDx of hepatic, renal, and ovarian masses

• Potential means to assess tumor response

• Potential means to enhance specifically targeted therapies=“theranostics”

• Potential use to enhance sonothrombolysis,antibodies to β‐amyloid in Alzheimer's disease

Thanks to• Charles Caskey, PhD VIIS

• Chelsea Samson, VUMC; Katrina Kohornen, HUP; Ryan Moore, U Oregon

• Andrej Lyschik, MD, PhD Jefferson U Med Ctr

• Lars Thorielus, MD, Linkoping U, Sweden

• Lanteus (Definity)

• Bracco (Optison/Lumason)

• Pfizer Oncology

• John Bobbitt, VUMC

contrast enhanced - vanderbilt university medical … enhanced sonography arthur c. fleischer, m.d....

Documents