Transcript Document

Ruolo della terapia antiangiogenica
nel carcinoma mammario
Il punto di vista del
clinico
Fabio Puglisi, MD PhD
Therapeutic Options in
Metastatic Breast Cancer
• No single “Gold Standard” for
therapy in metastatic breast cancer
• Therapy should be individualized
based on goals, tumor, and patient
characteristics
Goals of Therapy
Overt metastatic disease is generally incurable
Judicious use of agents individualized
to the patient’s clinical situation
•
•
•
•
Maximize survival
Maintain disease control
Minimize symptoms from disease
Minimize toxicity from treatment
QUALITY
OF LIFE
Goals of Therapy
Maximize survival
Bevacizumab:First-line trials
E-2100
Efficacy
PFS
months
HR
OS
months
HR
AVADO
Ribbon-1
Capecitabine
Ribbon-1
A/T
Control
Arm
Beva
Arm
Placebo
Arm
Beva
Arm
7.5/15 mg/kg
Placebo
Arm
Beva
Arm
Placebo
Arm
Beva
Arm
5.9
11.8
8.2
9/10.1
5.7
8.6
8.0
9.2
0.60
P<.0001
25.2
26.7
0.88
P=.16
0.86 P=.12 (7.5 mg)
0.77 P=.006 (15 mg)
31.9
30.8/30.3
1.05/1.03
P=.72/.85
0.69
P=.0002
21.2
0.85
P=.27
29
0.64
P<.0001
23.8
25.2
1.03
P=.83
Endpoints in phase III Metastatic
Breast Cancer trials
9/73 (12%) of trials
demonstrated OS
gains
OS gains less
frequently noted in
first-line trials (8%)
than in second-line+
trials (22%)
Verma S, et al. The Oncologist 2011
Why OS gain is rarely
noted?
• Potentially active subsequent lines
(including crossover) are not controlled
in most RcTs
• Many RcTs lack statistical power to
detect plausible increases in OS
• Larger sample size is requested
• Longer follow-up period is requested
Survival post-progression
OS = PFS + SPP
If the progression event
is death, then SPP = 0
Broglio KR & Berry DA, JNCI 2009
Probability of statistically significant differences in
overall survival (OS) as a function of median survival
postprogression (SPP)
Broglio, K. R. et al. J. Natl. Cancer Inst. 2009
Chance of seeing a survival
benefit according to SPP
> 90% if SPP = 2 months
< 50% if SPP = 8 months
< 20% if SPP = 24 months
First-line trials and SPP
E-2100
Efficacy
AVADO
Ribbon-1
Capecitabine
Control
Arm
Beva
Arm
Placebo
Arm
Beva
Arm
7.5/15 mg/kg
Placebo
Arm
Beva
Arm
25.2
26.7
31.9
30.8/30.3
21.2
29
OS
months
0.88
P=.16
HR
SPP
months
19.3
1.05/1.03
P=.72/.85
14.9
23.7
21.8/20.2
0.85
P=.27
15.5
Ribbon-1
A/T
23.8
25.2
1.03
P=.83
20.4
15.8
Meta-analysis: summary of results
SPP, mos
17.5
19.7
Beva
Arm
Placebo
Arm
-
16
Meta-Analysis
crossover and post-study therapies
Therapies used upon progression in AVADO and RIBBON-1a
CT +
Beva
(n=1071)
CT
(n=654)
Chemotherapy
65
71
Bevacizumab
40
51
Hormonal therapy
23
25
Number of subsequent agents
–1
–2
–3
–≥4
15
26
12
23
10
27
15
27
Therapy, %
aData
not available from E2100
O’Shaugnessy J et al, ASCO 2010. Abstract 1005
Estimating scenarios for survival
• 36 first-line chemotherapy trials for metastatic
breast cancer published from 1999 to 2009
–
–
–
–
–
–
–
Mean for Median PFS: 7.6 months (6.0-9.0)
Mean for Median SPP: 14 months (10.8-15.6)
Mean for Median OS: 21.7 (18.2-24.0)
Mean for Median ratio OS/PFS: 3 (2.4-3.5)
Mean 1-year survival: 73% (69-78%)
Mean 2-year survival: 45% (38-50%)
Mean 5-year survival*: 12% (7-17%)
*information available only in 14 trials
Published Ahead of Print on December 28, 2010 as 10.1200/JCO.2010.30.2174
Survival curve percentiles and their
corresponding scenarios
Published Ahead of Print on December 28, 2010 as 10.1200/JCO.2010.30.2174
Simple rules of thumb: bevacizumab
Estimates by multiplying median by four
simple multiples:
• 0.25 (worst-case)
 0.25 x 26.7 = 6.67
• 0.5 (lower-typical)  2 (upper-typical)
 0.5 x 26.7 = 13.3  2 X 26.7 = 53.4
• 3 (best-case)
 3 x 26.7 = 80.1 (= 6.7 years)
The Main Question
Who are these patients,
and what characteristics
predict for the tail of the
curve?
Goals of Therapy
Maintain disease control
Minimize symptoms
from disease
Endpoints in Clinical Trials
What Matters Most?
• Progression-free survival and response rate
are important achievements in their own right
– Shrinking a cancer may minimize a patient's acute
symptoms.
– Prolonging progression-free survival may be
associated with enhanced quality of life, even
without an improvement in overall survival.
E2100: response ratea
Investigator assessment
IRF assessment
100
100
80
80
p<0.0001
60
48%
40
Patients, %
Patients, %
p<0.0001
60
40
23%
aPatients
22%
20
20
0
0
Paclitaxel
50%
Bevacizumab
+ paclitaxel
with measurable disease at baseline
Klencke et al. ASCO 2008
Paclitaxel
Bevacizumab
+ paclitaxel
AVADO: Overall Response Ratea
(Bevacizumab 15 mg/kg q3w)
100
p=0.0003
Patients, %
80
64%
60
46%
40
20
0
Placebo +
docetaxel
(n=207)
aPatients
with measurable disease at baseline
Miles et al. SABCS 2009
Bevacizumab 15
mg/kg q3w +
docetaxel
(n=206)
Goals of Therapy
Minimize toxicity from
treatment
Bleeding/hemorrage
• Serious hemorragic events (grade ≥ 3) were
uncommon
≤ 1.7% of patients in the bevacizumab arms
(only in the taxane-BV arm of RIBBON-1: 5.4%)
• Trials allowed use of anticoagulants and aspirin
• Exploratory analysis of AVADO data
– No CNS bleeding events in pts who developed brain metastases
while on study
Hamilton EP & Blackwell KI. Oncology 2011; 80: 314-25
Wound-healing complications
• Incidence of grade 3 or 4 wound-healing
complications
≤ 1.5% of patients in the bevacizumab arms
≤ 1% of patients in the control arms
• Interval between bevacizumab
administration and elective surgery
– Based on 20-day half-life
– Do not administer bevacizumab at least 4 weeks
before and 4 weeks after surgery
Hamilton EP & Blackwell KI. Oncology 2011; 80: 314-25
Thromboembolic events
• Arterial thrombotic events
– Twice as frequently in patients treated with
bevacizumab
• 3.8% vs. 1.7% (meta-analysis of trials in mCRC,
MBC, NSCLC)
• No increased risk for venous
thromboembolic events
Hamilton EP & Blackwell KI. Oncology 2011; 80: 314-25
When Meta-analyses add little
to our body of evidence:
Bevacizumab and Heart Failure Risk
• TRIALS
–
–
–
–
–
Miller JCO 2005
E2100
AVADO
RIBBON-1
RIBBON-2
• Bevacizumab in pts with
MBC increase the risk of
G3-4 CHF five-fold with an
overall incidence of 1.6%
(vs 0.4% in the
control/placebo group)
Choueri, J Clin Oncol 2011
•
Retrospectively collected heart
failure data
•
Lack of information about individual
patients
•
No information about underlying risk
factors
–
–
–
–
•
Cumulative anthracycline dose
Prior radiation
Atherosclerotic disease
Hypertension/Diabetes/Obesity
Lack of accurate definition of heart
failure
– Heart failure is not equilavent to
cardiomyopathy or to left ventricular
dysfunction
Verma & Swain, J Clin Oncol 2011
Cardiovascular events
• RIBBON-1 is the only phase III trial
including a prospective cardiac
evaluation
– No significant increase of grade ≥ 2 left
ventricular systolic dysfunction when
bevacizumab was combined with
anthracyclines
• 6.2% vs. 6%, respectively, at the primary
data cut
Robert NJ, et al. J Clin Oncol 2011
To understand the
risk/benefit ratio
Clinical benefit and molecular
heterogeneity of breast cancer
unselected population
ORR/PFS
Survival
• Predictors of response/PFS may not predict OS in unselected cases
• A single predictive biomarker cannot fit all tumor types
Clinical benefit and molecular
heterogeneity of breast cancer
unselected population
Population
A
Population
C
ORR/PFS
Survival
Population
B
• Predictors of response/PFS may not predict OS in unselected cases
• A single predictive biomarker cannot fit all tumor types