Superior Patency, Durable Clinical and Angiographic Benefit of DES

1
Superior Patency, Durable Clinical and
Angiographic Benefit of DES

• David O. Williams, MD
• Director, Cardiovascular Laboratory
• and Interventional Cardiology
• Professor of Medicine
• Rhode Island Hospital
• Brown University Medical School
• Providence, Rhode Island

2
Scope of Discussion

• Drug-eluting versus bare-metal stents
• Results of randomized clinical trials
• Angiographic outcomes
• Clinical outcomes
• Results of observational studies
• Research Registry
• E-Cypher Registry
• Dynamic Registry
• Unique clinical subsets
• Multivessel CAD

3
DES vs BMS Randomized Clinical Trials

• Introduction
• Sirolimus has been the prototype for all DES vs
  BMS comparisons
• The evaluation of DES has been an evolutionary
  process, beginning with small cohorts of
  highly-selected patients from a few centers, to
  large enrollments derived from multiple centers,
  including more complex anatomy
• Since the sirolimus-eluting stent was the first
  to be investigated, long-term follow-up data are
  available for sirolimus-treated patients

4
Relevant Terms

• Angiographic restenosis
• Presence of a gt50 diameter stenosis within the
  treated segment
• Clinical restenosis
• Clinically driven target lesion revascularization
  (TLR) any repeat revascularization of the
  original target lesion
• Clinically driven target-vessel revascularization
  (TVR) any revascularization of the vessel
  containing the original target lesion
• Minimal lumen diameter (MLD)
• Luminal diameter (mm) at the narrowest site
  within a stenosis
• Late loss
• Difference (mm) in luminal diameter between
  immediately post-PCI to follow-up angiogram

5
Quantitative Coronary Angiography (QCA)
Angiographic Restenosis
Refer. dia. 3.9 mm
MLD 1.4 mm
Diameter stenosis 64
6
Late Loss
Post-PCI
Follow-Up
Late loss MLDpost-PCI MLDfollow-up
Within stents, late loss reflects the magnitude
of neointimal proliferation. Late loss within a
drug-eluting stent should be less than within a
bare-metal stent. The degree of late loss permits
a comparison of potency of drug-eluting stents
7
Superiority of DESRandomized Clinical Trials

• RAVEL
• The first RCT ever comparing a drug-eluting stent
  to a bare-metal stent
• Sirolimus was the drug
• Small sample size (N238) and shorter lesions
  (lt18 mm)
• Unique finding was the large magnitude of
  effectiveness for sirolimus (no angiographic or
  clinical restenosis or late loss) in Cypher
  group
• Safety endpoints were similar in frequency
• Duration of augmented anti-platelet therapy was
  brief (8 weeks) and no stent thrombosis was
  observed

8
Superiority of DESRandomized Clinical Trials

• SIRIUS
• Pivotal, US multi-centered, controlled,
  randomized clinical trial of Cypher vs
  bare-metal stent (Velocity)
• Large sample size (N1058) with more complex (2.6
  mm diameter, 14.4 mm length, 26 diabetes)
  lesions
• Duration of anti-platelet therapy 3 months
• Confirmed the effectiveness and safety of Cypher
  both early and late
• Restenosis rate and late loss
• TVR
• Peri-stent restenosis noted

9
SIRIUS Restenosis RatesAccording to Location
in-stent
prox
distal
in-segment

Four-fold reduction in in-segment restenosis
Limited effectiveness for proximal margin
restenosis
10
SIRIUS Late Loss Analysis (mm)
in-segment

Reduction in restenosis mediated by less late
loss Benefit compared to bare metal stents within
stent and at margins
11
SIRIUS Clinical Events at One Year
30
Sirolimus
Control
Plt0.001
25
20.0
20
Patients Percent ()
15
10
4.9
3.4
5
3.0
1.3
0.8
0.8
0.4
0
Death
MI
TLR
Thrombosis
Clinical follow-up demonstrated effectiveness in
reducing the incidence of repeat
revascularization without any excess of adverse
events No effect on late MI or death compared
to BMS
Holmes DR. Circulation. 2004109634-640.
12
Superiority of DESRandomized Clinical Trials

• NEW SIRIUS
• A combination of two randomized clinical trials
  that followed SIRIUS
• Canadian and European centers using a modified
  SIRIUS protocol
• Direct stenting permitted
• Less pull back post-dilation
• Longer stent/lesion ratio
• Demonstrated results superior to SIRIUS and
  convincingly validated the effectiveness and
  safety of Cypher

13
Angiographic Results 8 Months
In-Lesion Restenosis
42.9
Sirolimus
ns
36.3
Control
?75
?86
Patients Percent ()
ns
8.9
5.9
SIRIUS
E-SIRIUS
14
New SIRIUS Peri-Stent Analysis
Restenosis (8 months)
42.7
CYPHER Stent
Plt0.001
Control (n353)
85
93
72
?
?
?
Percent ()
Plt0.001
Plt0.001
10.1
7.4
3.1
2.1
1.5
Proximal Margin
In-Stent
Distal Margin
15
New SIRIUS Clinical Data
TLR at 9 months
CYPHER Stent n225
Plt0.001
Control n227
80
?
Percent ()
20.3
4.0
TLR
Further reduction in incidence of repeat
revascularization in comparison to SIRIUS
reflecting treatment effects at stent margins
16
Superiority of DESRandomized Clinical Trials

• TAXUS IV
• Pivotal, US multi-centered, controlled,
  randomized clinical trial of Taxus (paclitaxel)
  vs bare-metal stent (Express)
• Large sample size (1314) with more complex (2.8
  mm diameter, 13.4 mm length, 24 diabetes)
  lesions
• Duration of anti-platelet therapy 6 months
• Confirmed the effectiveness and safety of
  paclitaxel/polymer technology
• Restenosis rate and late loss
• TVR

17
TAXUS 4 Late Loss
Control (n267)
Plt0.0001
Taxus (n292)
1
0.92
Plt0.0001
0.75
0.61
0.5
Late loss (mm)
P0.002
P0.0007
0.39
0.27
0.23
0.25
0.17
0.15
0.05
0
Analysis
Proximal edge
In-Stent
Distal edge
segment
Paired Lesion Analysis
18
TAXUS 4Restenosis Rate
Control (n267)
RR0.23 0.13, 0.38 Plt0.0001
Taxus (n292)
40
30
RR0.30 0.19, 0.46 Plt0.0001
26.6
24.4
Restenosis ()
20
P0.27
7.9
P0.81
10
5.5
3.4
2.7
1.9
0.7
0
Analysis
Proximal edge
In-Stent
Distal edge
segment
19
TAXUS 49-Month TLR and TVR
Control (n652)
Taxus (n662)
20
RR0.27 0.16-0.43 Plt0.0001
RR0.39 0.26-0.59 Plt0.0001
15
12.0
11.3
Event ()
10
4.7
5
P0.48
3.0
1.7
1.1
0
TLR
TVR / non TLR
TVR
20
Superiority of DES Observational Studies

• Research Registry
• Single-center registry enrolling consecutive
  patients receiving Cypher stent (n563)
• Compared DES patients to a historical control
  group (n806)
• Substantial differences in baseline features,
  including the proportion of patients receiving
  stents (DES 100, Control 87)

21
Research RegistryEvents at Six Months
18
Plt0.01
Cypher
16
Control
14.1
14
12
Plt0.01
10
Patients Percent ()
8
6.8
5.7
6
4
2.7
2.7
1.4
1.4
1.2
2
0
Death
MI
Repeat Revase
MACE
Single-center observational study whose findings
confirmed the effectiveness and safety of
randomized trials for an off-label population
22
Superiority of DES Observational Studies

• E-SIRIUS Registry
• Post-marketing surveillance registry in Europe to
  determine
• Safety and reliability of SES in routine clinical
  use
• Reproducibility of RCT results
• Use of SES worldwide in daily practice
• Identification of MACE predictors
• All patients receiving gt1 SES are enrolled
• Both on- and off-label uses are included
• Clinical follow-up at 1, 6, and 12 months
• Target enrollment 15,000 patients

23
E-Cypher Six-Month Outcomes
5
4
9473 patients (83 of those eligible)
2.95
3
Patients Percent ()
2
1.01
0.9
1
0.62
0.37
0.25
0.1
0
Cardiac
Other
Q AMI
NQ AMI
TLR PCI
TLR
MACE
Death
Death
CABG
Demonstrated very low rates of adverse events in
multi-center real-world use of Cypher DES.
Rate of follow-up limits ability to address
infrequent events such as stent thrombosis
24
Superiority of DES Observational Studies

• Dynamic Registry
• Current iteration of the original NHLBI PTCA
  Registry
• Enrolls consecutive patients in whom PCI is
  attempted
• 16 North American clinical sites experienced in
  PCI and clinical investigation
• Multiple waves of enrollment dating back to the
  first PTCA permit comparisons of PCI over time
• Recent Wave 4 enrollment 2004 of 2690
  consecutive PCI patients with gt90 rate of
  consent

25
Dynamic Registry DES vs BMS Baseline
Characteristics
Substantial difference in patient selection
between DES and BMS, especially in regard to
patients presenting with AMI
26
Dynamic Registry DES vs BMS 30-Day Event Rates
4.2
4.5
DES
4
3.7
BMS
3.5
2.9
2.8
3
Patients Percent ()
P0.49
2.5
2
1.6
1.5
1
0.6
0.3
0.5
0.15
0
Death
Death,MI
Death, MI,
Thrombosis
CABG
More favorable outcome with DES. No excess of
stent thrombosis Differences at baseline need to
be considered in interpretation
27
Superiority of DES Observational Studies

• DEScover Registry
• Purpose of Registry
• To characterize the use of DES and the outcomes
  of patients treated with them in the real-world
• To compare the use of DES among hospitals and
  practices
• To compare DES usage and outcomes to those of
  balloon angioplasty alone and bare-metal stents
• To enable a broad sampling of hospitals and
  practices
• To obtain important economic, quality-of-life,
  and clinical information

28
Superiority of DES Observational Studies

• DEScover Registry
• Study Design
• An observational study
• Sample size 15,000 patients
• 200-300 clinical sites
• Experienced data-coordinating center
• Individual site queries
• Clinical Events Adjudication Committee
• Eligibility Criteria
• Consecutive patients undergoing PCI
• Patients included regardless of treatment
  delivered
• Patients eligible if guiding catheter introduced

29
PCI for Multivessel Disease Impact of Stents
Srinivas VS et al. Circulation.
20021061627-1633.
30
Dynamic Registry vs BARI A Matched MVD
Comparison
In-Hospital Events
P0.001
12
10.2
Dynamic
10
BARI
8
Patients Percent ()
6
P0.03
P0.72
4
2.1
1.9
1.1
2
0.9
0.8
0
CABG
MI
Death
Impact of stents was to reduce the incidence of
Peri-procedural MI and need for CABG
Srinivas VS et al. Circulation.
20021061627-1633.
31
Dynamic Registry vs BARI A Matched MVD
Comparison
One-Year Outcomes
30
Plt0.001
Plt0.001
Dynamic
25
22.7
BARI
22.5
20
P0.036
Patients Percent ()
15
12.4
P0.47
11.0
8.6
10
7.9
4.9
4.1
5
0
PCI
CABG
Death/MI
Death
Late effect of stent use is to reduce repeat
revascularization. Stents have no effect on late
MI or death. Likely effect of DES will be further
reduction in repeat revascularization
Srinivas VS et al. Circulation.
20021061627-1633.
32
Dynamic Registry vs ARTS
One-Year Events
25
Dynamic
21.2
19.4
ARTS
20
15
Patients Percent ()
9.5
10
7.9
4.9
5
2.5
0
PCI/CABG
Death/MI
Death
Dynamic Registry predicted the results of the RCT
33
Superiority of DESMultivessel CAD

• ARTS 2
• Non-randomized comparison of consecutive MVD
  patients treated with Cypher stent (n606) to
  CABG arm of ARTS 1 (n605)
• Primary endpoint absence of MACE at one year
• Intervention Cypher DES
• Stable angina/ischemia study population

34
ARTS 2 30-Day Results
9
8.2
8
ARTS 2
ARTS 1-PCI
7
ARTS 1-CABG
6
5
Patients Percent ()
4.1
3.7
4
2.8
3
2.5
2.4
2.3
2
1.5
1
1
0.5
0.5
0.4
0.3
0.2
0
0
Death
MI
Stroke
PCI/CABG
MACE
Successful use of Cypher stents for patients
with MVD with initial results as good if not
better than BMS
35
Impact of DES in Multivessel Disease

• Single-center observational report of 155
  patients
• 30 days death 0.6, Q-MI 1.9, repeat
  revascularization 0.6.
• 6 month death 2.7, Q-MI 3.6, TVR 17.9
• FREEDOM Trial
• RCT of sirolimus stent in patients with
  diabetes,PCI vs CABG

Orlie D, et al. J Am Coll Cardiol.
2004431154-60.