C. Lieu, H. Tran, Z. Jiang, M. Mao, M. Overman, C. Eng,

C. Lieu, H. Tran, Z. Jiang, M. Mao, M. Overman, C. Eng, J. Morris, L. Ellis, J. Heymach, and S. Kopetz

Departments of Gastrointestinal Medical Oncology, Surgical Oncology, and Head and Neck/Thoracic Medical Oncology

The University of Texas MD Anderson Cancer Center

The Association of Alternate VEGF Ligands With Resistance to Anti-VEGF Therapy in Metastatic

Colorectal Cancer

Cell membrane

VEGF-A

VEGF-R1(Flt-1)

MigrationInvasionSurvival

VEGF-R3(Flt-4)

Lymphangio-genesis

VEGF-R2(KDR/Flk-1)Proliferation

SurvivalPermeability

PlGF VEGF-C, VEGF-D

Func

tions

VEGF Biology

Y

Bevacizumab

Adapted from Bergers et al. Nat Rev Cancer. 2008;8:592-603.

VEGF-A VEGF-A

X VEGF-C

VEGF-D PlGF

Baseline Treatment with bevacizumab

AngiogenesisRestored

VEGF-A

X

Hypothesis: Alternate VEGF ligands are associated with the resistance to anti-VEGF therapy in patients with metastatic colorectal cancer

Study Design

Prospective Clinical Trial· Phase II trial of FOLFIRI +

bevacizumab in patients with metastatic colorectal cancer

· Forty-three patients enrolled· Intensive cytokine measurements

Retrospective Validation Cohort· The Texas Genetic Consortium

database (n = 710)· Heterogeneous treatment

histories· Single cytokine measurement

PD

PD

PD

Study Design

Prospective Clinical Trial· Phase II trial of FOLFIRI +

bevacizumab in patients with metastatic colorectal cancer

· Forty-three patients enrolled· Intensive cytokine measurements

Retrospective Validation Cohort· The Texas Genetic Consortium

database (n = 710)· Heterogeneous treatment

histories· Single cytokine measurement

PD

PD

PD

Prospective Cohort: PlGF Increased Prior to Progression

Kopetz et al. J Clin Oncol 28:453-459

0

20

30

40

ULN

p<0.001*

p<0.001*

p<0.01*

PlG

F (p

g/m

L)

Baseli

ne

Post Bev

Post FOLFIR

I

First R

estag

ing

Second R

estag

ing

Prior t

o Progressio

n

Progres

sion

0

500

1000

1500

2000

2500VE

GF-

C (p

g/m

L)

ULN

**

* p<0.05 by Mann Whitney U test

Prospective Cohort: VEGF-C Increased Prior to Progression

Baseli

ne

Post Bev

Post FOLFIR

I

First R

estag

ing

Second R

estag

ing

Prior t

o Progres

sion

Progres

sion

0

100

200

300

400VE

GF-

D (p

g/m

L)

ULN

Prospective Cohort: VEGF-D Minimally Increased at Progression

*

* p = 0.04

Study Design

Prospective Clinical Trial· Phase II trial of FOLFIRI +

bevacizumab in patients with metastatic colorectal cancer

· Forty-three patients enrolled· Intensive cytokine measurements

Retrospective Validation Cohort· The Texas Genetic Consortium

database (n = 710)· Heterogeneous treatment

histories· Single cytokine measurement

PD

PD

PD

· Separated patients into three groups:- Patients presenting prior to frontline therapy- Patients treated with chemotherapy without bevacizumab - Patients treated with chemotherapy and bevacizumab

· To minimize heterogeneity, samples were matched for:- Metastatic disease sites- Chemotherapy cycles- Time from last chemo to plasma collection

· 533 patients were included in the analysis

Retrospective Cohort

PlGF Elevated After Bevacizumab

No Treatment Chemo Only Chemo+Bev0

20

40

60

80

Group

Con

cent

ratio

n, p

g/m

L

ULN

p < 0.0001

p < 0.0001

No Chemo Chemo Only Chemo+Bev0

500

1000

1500

Group

Con

cent

ratio

n, p

g/m

L

ULN

VEGF-C Elevation Unable to be Confirmed

p < 0.0001

p = 0.64

No Chemo Chemo Only Chemo+Bev0

100

200

300

400

500

Group

Con

cent

ratio

n, p

g/m

L

ULN

p < 0.0001

Minimal VEGF-D Elevation Confirmed

Summary

PlGF

VEGF-C

VEGF-D

Prospective Retrospective Conclusions

Baseli

ne

After B

evac

izumab

After F

OLFIRI+B

Prior t

o Progres

sion

Progres

sion

0

20

30

40

ULN

p<0.001*

p<0.001*

p<0.01*

PlG

F (p

g/m

L)

No Treatment Chemo Only Chemo+Bev0

20

40

60

80

Group

Con

cent

ratio

n, p

g/m

L

ULN

Baseli

ne

Post Bev

Post FOLFIR

I

First R

estag

ing

Second Res

taging

Prior to

Progres

sion

PD 0

500

1000

1500

2000

2500

VEG

F-C

(pg/

mL)

ULN

No Chemo Chemo Only Chemo+Bev0

500

1000

1500

Group

Con

cent

ratio

n, p

g/m

L

ULN

Baseli

ne

Post Bev

Post FOLFIRI

First R

estag

ing

Second R

estag

ing

Prior to

Progressio

n PD

0

100

200

300

400

VEG

F-D

(pg/

mL)

ULN

No Chemo Chemo Only Chemo+Bev0

100

200

300

400

500

Group

Con

cent

ratio

n, p

g/m

L

ULN

1) PlGF is elevated after FOLFIRI+B

2) A similar elevation was seen after chemotherapy + bev but not after chemotherapy alone

1) VEGF-C is elevated after FOLFIRI+B

2) No difference was seen in the second cohort between the two “post-therapy” groups

3) Limitations include heterogeneity and high inter-patient variability

1) Modest elevations in VEGF-D were seen after FOLFIRI+B

2) Elevations were seen in the “post-therapy” groups but not impacted by bevacizumab therapy

How long do PlGF and VEGF-D stay elevated after bevacizumab?

Time from last bevacizumab dose (mo)

VEG

F-D

(pg/

mL)

0 1 2 3 4 5 60

100

200

300

400

500

600

700 VEGF-D T1/2 = 1.5 months(95% CI: 1.2 - 2.0)

Time from last bevacizumab dose (mo)

PlG

F (p

g/m

L)

0 1 2 3 4 5 60

20

40

60

80

100

120 PlGF T1/2 = 1.6 months(95% CI: 1.4 - 1.9)

No temporal change in PlGF and VEGF-D in patients treated with chemotherapy only

Limitations of Cytokine Analysis· How well do circulating levels of VEGF ligands reflect the tumor

microenvironment?- Or host response?

· Difficult to place magnitude of changes into context- What degree of elevation would be necessary to evoke a biologic

response

· Association vs. Causation- Are alternate VEGF ligands driving resistance to bevacizumab

· Return to preclinical models· Clinical trial

VEGF-APlGF VEGF-C, VEGF-D

Large molecule VEGF inhibitors

Y

Bevacizumab

YVGX-100

YRamucirumab(IMC-1121B) II

CT-322

Y

IMC-18F1

Aflibercept (VEGF Trap)Y

TB403

Phase III VEGF-Trap (Aflibercept) after Bevacizumab

2nd line CRC (after treatment with

oxaliplatin-based therapy)

N=1200 patients

Primary endpoint: OS

R

FOLFIRI + Placebo

FOLFIRI + Aflibercept 4mg/kg

“VELOUR” Study Primary endpoint OS met

Conclusions· VEGF family ligands other than VEGF itself are associated with

bevacizumab-containing chemotherapy resistance in mCRC

· Plasma levels of PlGF are increased prior to radiographic progression of disease

· Changes in VEGF-C were not able to be validated- Limited by technical concerns in the validation cohort

· VEGF-D is minimally increased at the time of progression- Unclear biologic significance

· Further study of agents targeting multiple VEGF-ligands are ongoing

Acknowledgments

· GI Medical Oncology- Scott Kopetz- Karen Mao- Camilla Ziang- James Abbruzzese

· Thoracic/H&N Medical Oncology- Hai Tran- John Heymach- Stef Fiorentino

· GU Medical Oncology- Gary Gallick

· Funding- ASCO Cancer Foundation

Young Investigators Award- Circadian Technologies- T32 Training Grant

Plasma PlGF in mCRC· In a prospective cohort, plasma PlGF

levels are elevated prior to progression and at the time of progression on a bevacizumab regimen

· In a validation cohort, compared with untreated patients and patients who have received chemotherapy only, patients who have received chemotherapy and bevacizumab had elevated levels of PlGF

· These changes appear specific to patients receiving bevacizumab

0

20

30

40

ULN

p<0.001*

p<0.001*

p<0.01*

PlG

F (p

g/m

L)

Lieu et al. ASCO 2011 Abstract #3533

Lieu et al. ASCO 2011 Abstract #3533