Transcript Death or Q-wave MI - Clinical Trial Results

An Overview of Delayed Stent Thrombosis and DrugEluting Stent Mortality Meta-Analysis and
Drug-Eluting Stent Late Thrombosis Meta-Analysis
Presented at
European Society of Cardiology Scientific Congress,
September 2006
Presented by Dr. Alain J. Nordmann and Dr. Edoardo Camenzind
Initial Reports of the Potential Hazard: Bare Metal and Drug
Eluting Stent in Same Pt:
DES thrombosis, BMS stays open
“We report four cases of angiographically-confirmed stent
thrombosis that occurred late after elective implantation of
polymer-based paxlitaxel-eluting (343 and 442 days) or
sirolimuseluting (335 and 375 days) stents, and resulted
in myocardial infarction.
All cases arose soon after antiplatelet therapy was
interrupted.
If confirmed in systematic long-term follow-up studies, our
findings have potentially serious clinical implications.”
McFadden EP et al, Lancet 2004; 364: 1519–21
www. Clinical trial results.org
Wall Street Journal September 7th, 2006
“Previously, Boston Scientific had reported that among some 3,400 patients
participating in its clinical trials, there had been eight thrombosis cases in the
Taxus group versus one case in the bare-metal group during follow-up from six
months to three years.
But this summer the company analyzed data that included longer follow-up times - four years for some of the trials -- and the results for an increased risk of
thrombosis, roughly 0.5%, became statistically significant.
The increase means that in comparison to bare-metal stents, an added one in
every 200 patients gets a stent blood clot between six months and four years after
implantation. But stent thrombosis can be deadly about 30% of the time, estimates
suggest, and with more than a million people getting the devices every year, some
experts believe the issue has become a public-health matter.”
Wall Street Journal, By SYLVIA PAGÁN WESTPHAL and RON WINSLOW, September 7, 2006
www. Clinical trial results.org
Wall Street Journal September 7th, 2006
“According to figures provided by J&J, based on an analysis of the
company's four key stent trials, there were five cases of thrombosis
between one and four years in Cypher patients, compared to no
cases of thrombosis in bare-metal patients, an increased risk of
0.6% for the Cypher.”
“In the analysis, which included about 1,800 patients, this difference
didn't reach statistical significance.”
“The company's position for now, said Dr. Donohoe, is that there
seems to be no "safety signal" about late thrombosis in Cypher
stents.”
Wall Street Journal, By SYLVIA PAGÁN WESTPHAL and RON WINSLOW, September 7, 2006
www. Clinical trial results.org
Drug-Eluting Stent Mortality Meta-Analysis: Background
• Compare mortality rates of patients treated with drugeluting stents (DES) versus bare-metal stents (BMS)
among randomized DES trials
• Compare mortality rates of different specific drug-eluting
stents: sirolimus-eluting stents (SES) and paclitaxel-eluting
stents (PES)
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Late Thrombosis Meta-Analysis:
Background
• Compare death or Q-wave myocardial infarction (MI)
rates among patients treated with DES versus BMS in
randomized trials of DES
• Compare death or Q-wave MI rates for treatment with
SES and PES versus BMS
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Mortality Meta-Analysis: Study Design
Meta-analysis of 17 randomized trials of treatment with DES versus BMS in patients
undergoing percutaneous coronary intervention (PCI) for de novo coronary lesions
Treatment with BMS
Treatment with DES: SES or PES
Follow-up > 1 year post PCI
(Mean follow-up = 3-4 years)

Primary Endpoint: Mortality (total, cardiac, and noncardiac)
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Late Thrombosis Meta-Analysis: Study
Design
Meta-analysis of first-generation DES versus BMS in patients undergoing percutaneous
coronary intervention (PCI) for de novo coronary lesions (n = 5,108). Studies used in metaanalysis: RAVEL, SIRIUS, ESIRIUS and SIRIUS for SES and TAXUS I, II, IV, V and VI for PES
Treatment with BMS (n = 870
for SES trials, n = 1,675 for
PES trials)
Treatment with DES: SES (n = 878)
or PES (n = 1,685)
Mean follow-up = 3 years

Primary Endpoint: Death or Q-Wave MI (surrogate for late thrombosis)
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Mortality Meta-Analysis: Total Mortality
• Total mortality at 1 year did not differ for BMS vs DES
(OR 0.94, 95% CI 0.66-1.34) or for individual type of
stents: SES (OR 0.86, 95% CI 0.49-1.51) and PES (OR
0.98, 95% CI 0.64-1.48).
• Total mortality at 3 years trended toward higher, but
non-significant, rates for DES vs BMS (OR 1.25, 95% CI
0.91-1.73) as well as for SES vs BMS (OR 1.48, 95% CI
0.91-2.42).
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Mortality Meta-Analysis: Cardiac and
Noncardiac Mortality
• Cardiac mortality at 3 years did not differ with DES
treatment vs BMS treatment (OR 1.00, 95% CI 0.621.60).
• Non-cardiac mortality at 3 years was directionally, but
not significantly, higher for treatment with DES vs BMS
(OR 1.45, 95% Cl 0.93-2.25).
• Treatment with SES was associated with significantly
higher rates of non-cardiac mortality at 2 years (OR 2.74,
95% CI 1.22-6.13, p < 0.05) and 3 years (OR 2.04, 95%
CI 1.00-4.15, p < 0.05) compared with BMS.
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Late Thrombosis Meta-Analysis:
Mortality/MI
Death or Q-wave MI (%) in patients treated with BMS
(n=870) versus SES (n=878)
8%
p=0.06
6.0%
Death/MI
6%
p=0.03
6.3%
p=0.09
4%
p=0.30
p=0.21
2%
2.3%
1.7% 1.4%
0.9%
3.7%
4.0%
3.9%
3 years
last
followup
2.0%
• Death or Q-wave
MI was
significantly
higher in patients
treated with SES
versus BMS at
last available
follow-up (6.3%
vs 3.9%, p =
0.03).
0%
6-9
months
1 year
BMS
www. Clinical trial results.org
2 years
SES
Presented at ESC 2006
Drug-Eluting Stent Late Thrombosis Meta-Analysis:
Mortality/MI
Death or Q-wave MI (%) in patients treated with BMS
(n=1,675) versus PES (n=1,685)
5%
p=0.60
Death/MI
4%
p=0.78
3%
p=0.88
2%
p=0.80
1.5%1.6%
1.6%1.7%
6-9
months
1 year
2.8%
2.6%
3.5%
3.1%
p=0.68
2.6%
2.3%
1%
0%
BMS
www. Clinical trial results.org
2 years
3 years
• Death or Qwave MI did not
differ
significantly in
patients treated
with PES
versus BMS at
last available
follow-up (2.6%
vs 2.3%, p =
0.68).
last
followup
PES
Presented at ESC 2006
Drug-Eluting Stent Mortality and Late Thrombosis MetaAnalysis: Limitations
• The scope of these meta-analyses was limited in that
none of the randomized trials were powered to
investigate harder endpoints of death or Q-wave MI.
• Longer-term follow-up and greater insight into the
specific causes of death will be useful.
www. Clinical trial results.org
Presented at ESC 2006
Drug-Eluting Stent Mortality and Late Thrombosis MetaAnalysis: Summary
• From Late Thrombosis Meta-Analysis:
– Treatment with SES was associated with significant increase in late death
or Q-wave MI at 3 years versus treatment with BMS.
– Treatment with PES has no significant difference on incidence of death or
Q-wave MI compared with BMS treatment.
• From Mortality Meta-Analysis:
– Treatment with DES was not associated with a significant difference in
total mortality compared to treatment with BMS at 3 years
– Higher rates of noncardiac death at both 2 and 3 years with SES
compared with BMS
– Present meta-analysis suggests a potential, but nonsignificant hazard in
late mortality with DES compared with BMS
www. Clinical trial results.org
Presented at ESC 2006
6 Month angiographic follow-up is when most target
lesion revasc occurs; this may overestimate clinical
need for revasc
www. Clinical trial results.org
Potential Magnitude of the Problem from a Public Health
Perspective: Liberal Estimate
• 1,000,000 stents placed per year
• Assume meta analysis figure of 2.4% increase in the risk
of death or MI with use of drug eluting stent vs bare metal
stent (6.3% vs 3.9%)
• 2.4% of 1,000,000 would be 24,000 death or MIs per year
associated with drug eluting stent use over bare metal
stent use.
www. Clinical trial results.org
Potential Magnitude of the Problem from a Public Health
Perspective: Conservative Estimate
• 1,000,000 stents placed per year
• Assume a conservative figure of 0.5% incremental risk of stent
thrombosis with use of drug eluting stent vs bare metal stent (Boston
Scientific estimate)
• Assume 30% of stent thromboses are fatal
• Then 30% of 0.5% = 0.15% more patients would die if a drug
eluting stent was placed
• 0.15% of 1,000,000 would be 1,500 deaths per year associated
with drug eluting stent use over bare metal stent use
www. Clinical trial results.org
Potential Magnitude of the Problem from a Public Health
Perspective: Vioxx Risk vs DES Risk
Absolute Increase in Risk of MI
0.7
0.6%
0.6
• Incremental risk of MI with
Vioxx in the VIGOR Trial: 0.3%
(increased from 0.1% to 0.4%)
0.5%
0.5
• Incremental risk of MI with
Taxus stent according to
Boston Scientific: 0.5%*
0.4
0.3%
0.3
• Incremental risk of MI with
Cypher stent according to J
and J: 0.6%*
0.2
0.1
0
Vioxx
Taxus
www. Clinical trial results.org
Cypher
* Wall Street Journal, By SYLVIA PAGÁN
WESTPHAL and RON WINSLOW,
September 7, 2006
Challenges in Evaluating Late Stent Thrombosis
• Event rates are low, hard to be certain of risk
• Larger randomized trials needed with longer duration of
follow-up
• Devices have traditionally not had to demonstrate a
durable benefit in hard endpoints such as death or MI
• A non fatal surrogate endpoint (restenosis) has been
accepted over death or MI
• Registries are not going to be adequate given the
differences in the types of patients and lesions that DES
and BMS are placed in (lesion length, vessel diameter,
presence of thrombus, crush technique, bifurcation
location are all confounders)
www. Clinical trial results.org
Challenges in Evaluating Late Stent Thrombosis
• It will be critical to know if thrombosis was in target
vessel, target lesion, or uninstrumented artery (may not be
clear if an angio is not performed)
• The role of prothrombotic events must be identified (i.e.
did discontinuation of antiplatelet therapy before surgery
lead to stent thrombosis?)
• Optimal duration of thienopyridines is unclear (unclear if
longer duration may permit greater intimal hyperplasia to
form)
• Optimal dose of ASA is unclear
• Role of clopidogrel and ASA resistance is unclear
• Role of ticlid (not a pro drug, little resistance, rapid acting)
is unclear
www. Clinical trial results.org