Transcript ASCO_GI_2010_files/Ellis AntiAngio Educ ASCO GI 2010

Future Perspectives in Anti-Angiogenic
Therapy for Advanced Stage and
Adjuvant CRC
Lee M. Ellis, MD
Departments of Cancer Biology and Surgical Oncology
UT MD Anderson Cancer Center
Houston, Texas, USA
Overview
• Where are we now
• Where do we need to go
• A few comments on VEGF-targeted therapy
in the adjuvant setting
Anti-angiogenic Therapy:
A Cure for Cancer or Hype????
1998
The Scorecard: Phase III Trials
Chemo +/- VEGF Targeted Rx in CRC
Trial
Line of
Therapy
1o Endpoint
Met
RR AntiVEGF Rx
PFS
mos
5FU/LCV +/- SU5416
First Line
No
?
?
IFL +/- Bev
First Line
Yes
10%
4.4
FOLFOX +/- PTK/ZK
First Line
No
-4%
0.2
XELOX/FOLFOX +/- Bev
First Line
Yes
0
1.4
FOLFOX +/- Bev
Second Line
Yes
14%
2.6
FOLFOX +/- PTK/ZK
Second Line
No
?
1.5
FOLFIRI +/- Sunitinib
First Line
No
?
?
FOLFOX +/- Bev (C-08)
Adjuvant
No
NA
DFS No 
FOLFOX/XELOX +/- Bev
(AVANT)
Adjuvant
?
NA
DFS ?
January
2010
We Have All Seen These Patients, But We Do Not
Have Any Predictive Biomarkers to Inform Us Who Will
Benefit from the Addition of Bev and Who Will Not
Chemo + Bevacizumab
Chun et al. JAMA 2009
PREDICTIVE biomarkers are essential to optimize current therapies,
and future therapies, but so far they have remained elusive in the
field of angiogenesis.
We do not
seemintoFront
be making
muchIII
Overview
of PFS
Line Phase
Trials: Chemoprogress!
+ ONE Targeted Agent
Have we hit the ceiling?
More Must Be Better!
More is Not Better!
Addition of EGFR MoAB to an Oxali/5FU/Bev
Regimen Decreased PFS
CAIRO-2
PACCE
+ EGFR MoAB
Hecht et al. JCO 2009
Tol et al. NEJM 2009
What Have We Learned So Far About VEGF
Targeted Therapy in Patients with mCRC?
• All VEGF targeted agents are not created equal
– As of today, no TKI has been shown to improve upon a
chemotherapy backbone
• There are real toxicities
• Single agent therapy is not an option
– E3200
• We have not identified predictive biomarkers
• More is not better
– 2 trials with FU/Ox/Bev + EGFR MoAB show negative
interaction
We Need to Do Better!
But..how do we do this?
Overview
• Where are we now
• Where do we need to go
• A few comments on VEGF-targeted therapy
in the adjuvant setting
Defining our goal:
To SIGNIFICANTLY improve overall survival
VEGF Targeted Agents in the
Clinic or In Clinical Trials
VEGF-TRAP
TG-403
PlGF
VEGF-A
VEGF-B
BEVACIZUMAB
VEGF-C
VEGF-D
1121B
18F1
VEGFR-1
(Flt-1)
NRP-1/
NRP-2
VEGFR-2
(Flk-1/KDR)
VEGFR-3
(Flt-4)
NRP-2
Tyrosine Kinase Inhibitors
Sunitinib
Sorafenib
Pazopanib
Axitinib
Motesanib
Cedirinib
Brivinib
Many, many others
Vasculogenesis
Angiogenesis
Lymphangiogenesis
Ellis, Hicklin Nat Rev Ca. 2008
Current First Line Trials:
Chemo +/- VEGF Targeted Therapy
Although
these strategies are necessary for companies to
Phase III
achieve
their goals of+/-approval/registration,
these trials
• FOLFOX/CapOX
Cediranib (AZD2171) (HORIZON
II) are
not •likely
to lead
to vs
major
advances
in therapy
for mCRC
FOLFOX
+ Bev
FOLFOX
+ Cediranib
(HORIZON
III)
• Chemo + Bev vs Chemo + Cetuximab (80405)
• Although success is often times measured by “P<0.05”, obtaining a
PPhase
valueIIdoes not always translate into making progress in the field
• FOLFOX +/- Aflibercept (VEGF Trap)
- If you have Phase
a minimally
active
but enroll enough pts on
• Innumerable
II single
armagent,
studies
the study, you will obtain your desired P value, but you may
NOT advance the field
JCO Dec 2009
We Must Be More Creative!!
“Me too” drugs and trials are unlikely to
significantly advance the field
It is time to move new approaches
forward!!
There Are Only a Few Ways to Improve
Outcomes With VEGF Targeted Therapies
• Better drugs?
– Unlikely to advance the field with different VEGF-targeted agents
• No clear hints in Phase II/III trials in CRC or other solid
malignancies
• Although kinase inhibition profiles may be slightly different, I doubt
that they are different enough to distance themselves from the pack
• There may be an advantage with a “me too” drug if toxicity was
significantly less than other drugs in the class, but so far, this is not
the case
• Better patient selection?
– No predictive biomarkers, despite extensive search
• Better drug combinations? Different drugs/drug targets?
Thoughts on Improving VEGF
Targeted Based Therapies
Disclaimers
– For the sake of discussion, I will focus on front
line therapy, but my comments may apply to
subsequent lines of therapy.
– All drugs do not have to be used at once to
achieve an improvement in OS.
Since this is an anti-angiogenic lecture, I will focus on endothelial cell
targets, but please recognize that there are other targeted approaches that
focus on tumor cell biology that should be pursued.
Targeting Resistance Pathways Before It
Occurs
Caveats:
Ellis, Hicklin CCR 2008
– Resistance in mCRC is not for a single drug, but rather for a regimen
containing Bev.
– Preclinical modeling is not likely to reflect the complexities of the cytokine
response to multi-agent therapy
– We probably have not paid enough attention to induction of hypoxia in tumors
Until We Fully Understand the Mechanisms of
Action of VEGF Targeted Therapies, We Will Not
Be Able to Extract Maximal Benefit From Therapy
•
•
•
•
•
•
•
Anti-angiogenic
“Normalization” of the vasculature
Direct effect on tumor cells
Vascular “constriction”
Offset effects of stress
Reverse immuno-suppression due to VEGF
Disruption of the cancer stem cell niche
Acute Effects of Single Agent Bevacizumab Therapy in mCRC:
The Rapid Decrease (4 hrs) in Enhancing Fraction and Plasma
Volume Suggests Vasoconstriction and Hypoxia
(Gordon Jayson, FRCP, PhD)
This rapid induction of hypoxia leads to
induction/stabilization of hypoxia inducible factor (HIF)
©2009 by American Association for Cancer Research
O'Connor et al. Clin Cancer Res 2009
VEGF Targeted Therapy  Hypoxia and
Gene Induction
• HIF Regulated
Angiogenic Genes
–
–
–
–
–
–
–
–
Angiopoeitin-1
Angiopoeitin-2
VEGFR-1
VEGFR-2
MMP-2,9
PDGF-B
Tie-2
VEGF-A
Melillo and colleagues, Cell Cycle 2009
Targeting HIF
The “Response” to VEGF Targeted Therapies
• Intentional HIF
inhibitors
• Topotecan
• Anthracycline
• HSP90 inhibitors
• Proteosome inhibitors
Just because some agents
historically have not shown
activity in CRC, does not mean
they will not have value when
combined with the right drugs.
Onnis, Rapisarda, Melillo: J Cell Mol Med 2009
Daily Administration of Topotecan In Combination With
Bevacizumab, Inhibits Tumor Growth And HIF Induction
Regression!
Topotecan blocks the
hypoxia response element
- luciferase promoter
construct
Rapisarda A et al. Melillo. Mol Cancer Ther 2009
Other Vascular Targets
• DLL4/Notch
• Angiopoietins/Tie-2
Inhibition of Notch Leads to NonFunctional Vessels
Hicklin
Nat
Biotech
2007
Yan, Plowman CCR 2007
Ridgway et al.
Nature 2006
Notch inhibitors in clinical
trials
– MK0752
– PF-03084014
– REGN421 (Dll4 Ab)
Dual Inhibition of VEGF and Ang/Tie-2
Pathways
AMG 386
and
MoAB Ang-2
VEGF TKI
Combo
Bevacizumab
or
Motesanib
MCT 2010
or
Sorafenib
or
Ang-1/2/Tie-2 Inhibitors
− AMG-386 peptibody
− Arry-614 TKI
− MGCD 265 TKI
− CovX 60- Peptide-Ab
− CEP-11981 TKI
− BAY 73-4506
Sunitinib
Hong, MDACC
We Must Develop Rationale Combination Therapy
Targeting the Vasculature:
Chemo + VEGF-targeted Agent + Rationale Drug
• Signaling inhibitors to
–
–
–
–
–
mTOR
PI3K
Src
MAPK/Mek
HIF
• Cell surface/GFR inhibitors to
–
–
–
–
–
–
–
–
C-Met
FGF-R
Ephs
Tie-2 (Ang-1/2)
Notch (DLL4)
EGFL7
SDF-1
“Vertical” VEGF/R inhibition
• Beware toxicity
– Other VEGFRs
• VEGFR-3, NRP-1, NRP-2
We Cannot Ignore the Complex Interactions of
Chemotherapy and VEGF Biology
• Acute administration of oxaliplatin in vitro induced VEGF,
VEGF-C, PlGF, VEGFR-1, NRP-1 (Fan et al MCT, 2008)
• We need better preclinical models
– Including combination chemotherapy
– At the minimum, orthotopic models that replicate the
complexities of the tumor microenvironment
> We need to develop transgenic models of metastasis
• Clinical trials should evaluate biomarkers for
sensitivity/resistance to a regimen, and not just single
agent therapy
– Kopetz et al. JCO 2009
Moving Therapies Forward
• There are some interesting combination studies
in later lines of therapy, and in Phase I studies,
but we need to be brave enough to move these
approaches forward
– Examples
• Sirolimus + Bev, FU, Irinotecan: Ghiringhelli, WJG 2009
• Dasatinib + FOLFOX, Cetuximab: Kopetz, 2010 GI Cancers
Symposium
Overview
• Where are we now
• Where do we need to go
• A few comments on VEGF-targeted therapy
in the adjuvant setting
These Comments (and Those That I Made At ASCO)
Are Predicated On The Idea That
AVANT Is Negative Or The Same As CO-8
Adjuvant Therapy in CRC and Cure
• The goal of adjuvant therapy in CRC is CURE
(OS)
– DFS is not really meaningful without an improvement in
overall survival in asymptomatic patients
– DFS is used a surrogate for OS for chemotherapy
based regimens
- BUT C-08 taught us that early DFS cannot be used
as a surrogate for OS for regimens where Bev is
administered for a finite period of time
C08 - 1 yr of Bevacizumab
100
FOLFOX
+ Bev
80
60
BEV
40
The current question:
Would a longer duration of bevacizumab lead to more cures?
20
0
0.0
0.5
1.0
1.5
2.0
DFS
2.5
3.0
Why Would Longer Bev Exposure Lead to Tumor Cell
Eradication After Already Receiving 12 Months of Bev?
FOLFOX
+ Bev
Bev
After 12 months of therapy, I think we can declare
these tumor cells (and endothelial cells) resistant!
Unique Toxicities Observed in C-08
Grade 3/4 Toxicities Seen With Bev After Chemo
(Allegra et al. JCO 2009)
FOLFOX
FOLFOX + Bev
Definition of Grade 3 Toxicity
Any pain
2.1
4.8*
Severe pain; pain or analgesics severely
interfering with activities of daily living (ADL)
Depression
1.3
2.9*
Severe mood alteration interfering with ADL
Dizziness
0.7
1.8*
Interfering with ADL
• These toxicities are important when benefit of prolonged
therapy may be marginal
AVANT Adjuvant Colon Cancer Study
24 Weeks
48 Weeks
FOLFOX4 q2wk
n=3451
Stage III or
high-risk stage
II colon cancer
1:1:1
FOLFOX4 q2wk
Stratified by
stage and
region
Bev 5 mg/kg, q2wk
Bev 7.5 mg/kg q3wk
XELOX q3wk
Bevacizumab 7.5 mg/kg q3wk
•
•
•
3451 patients were enrolled between November 2004 and June 2007
Primary analysis: compare DFS between control and each treatment arm in stage III
patients
Projected final analysis time: Q3, 2010
Toxicities Observed In AVANT Trial
Focus on FOLFOX vs FOLFOX/Bev Arms
Grade 3/4/5 Toxicities
FOLFOX
+
Bleeding
0.6
FOLFOX/
Bev
1.2
Fistula/abscess
0.3
1.4
1.1
GI Perforation
0.1
0.7
0.6
VTEs
5.5
8.3
2.8
ATEs
1.0
1.6
0.6
SAE
19.7
25.9
6.2
Discontinuation due to AE
28.4
40.2
11.8
0.6
Hoff et al. ESMO 2009
Although Preclinical Modeling Can Support Nearly
Any Hypothesis
(including a benefit of prolonged administration of VEGF targeted therapy),
We Cannot Ignore Clinical Data
• We now know that the benefit of adjuvant Bev lasts
only as long a Bev is administered
– Or maybe even a little longer
• We either need to administer Bev:
– Forever (not feasible, nor would patients comply)
– Or not at all
• I would not expect any other VEGF targeted agent
to be effective either, as toxicity and compliance are
likely to be more of an issue
Conclusions/Recommendations
• We must be more aggressive for patients with mCRC
Suggestion:
– Phase II trials: Chemo + VEGF Targeted Rx + ? (rationale selection)
Future Suggestion:
– Consider “dropping” a chemotherapeutic agent if efficacy is good
• 5FU + VEGF Inhibitor + New Agent?
• Defer irinotecan and oxaliplatin for second line therapy limiting the
duration of long term toxicity of front line therapy
• If AVANT is the same as C-08, then we should not consider
long term VEGF blockade as a therapeutic option in the
adjuvant setting
– Likewise…lets be more creative in adjuvant therapy
Thank You for Your Attention!