Full Text
I bet few, if any, interventional cardiologists fail to prescribe dual antiplatelet therapy (DAPT) using a P2Y12 inhibitor
plus aspirin after performing a PCI. After all, the ravages that
balloons, stents and other devices incur on the fragile endothelial
lining of arteries that result in immediate platelet deposition have
been known since the early days of PCI.1
We’ve
come a long way from using low molecular weight dextran, coumadin,
aspirin and ticlopidine to the current DAPT combinations. But serious
questions remain. Why do some patients appear “resistant” to P2Y12 inhibition with clopidogrel? (Sidestepping the aspirin resistance issue.) And if this P2Y12 resistance is real, how do we identify these patients? Does such
resistance make a difference in clinical outcomes? The list of questions
continues.
Clopidogrel
“resistance” is well described. Clopidogrel is a “pro-drug.” To be
effective it requires activation via the cytochrome P450 CYP2C19 pathway
in the liver. But our genome may get in the way by containing within it
so-called CYP2C19 loss-of-function (LOF) alleles. These alleles reduce
CYP2C19 activity — resulting in reduced plasma levels of the active
clopidogrel metabolite and thus less inhibition of platelet aggregation
during clopidogrel therapy.
But platelet function while on
DAPT is highly variable within genetic groups — because CYP2C19
genotype and platelet reactivity are imperfect correlates of each other.
Nonetheless, one might guess that patients with an LOF allele, when
treated with clopidogrel after PCI, would have a higher risk of
platelet-induced events. And they do! Increased major adverse
cardiovascular events (MACE) have been shown in both randomized trials
and patient registries comparing patients with and without CYP2C19 LOF
alleles.
With
all the uncertainty, or need for additional genetic testing, why not
simply abandon clopidogrel and switch standard of care to alternative
P2Y12 inhibitors? In fact, prasugrel and ticagrelor were
better than clopidogrel in preventing MACE in acute coronary syndrome
(ACS) patients in the TRITON-TIMI 38 and recent PLATO trials.2,3 The answer is that the devil is in the details.
How
much better is actually better? For example, in the TRITON-TIMI 38
trial, at six to 15 months follow-up, the primary efficacy composite
endpoint (cardiovascular death, nonfatal myocardial infarction [MI] or
nonfatal stroke) occurred less frequently with prasugrel vs. clopidogrel
(9.9 vs. 12.1 percent; hazard ratio [HR], 0.81; 95 percent confidence
interval [CI], 0.73-0.90; p<0.001). The prasugrel group also beat the
clopidogrel group in the rates of MI (9.7 vs. 7.4 percent; p<0.001),
urgent target-vessel revascularization (3.7 vs. 2.5 percent;
p<0.001) and stent thrombosis (2.4 vs. 1.1 percent; p<0.001).
All the p values make prasugrel a clear winner in TRITON-TIMI 38. But stent thrombosis (arguably the main reason a P2Y12 inhibitor is used post PCI) was only 1.3 percent less in the prasugrel
group. What price was paid for that difference? Major bleeding was
observed in 2.4 percent of patients receiving prasugrel and in 1.8
percent of patients receiving clopidogrel (HR, 1.32; 95 percent CI,
1.03-1.68; p=0.03). Fatal bleeding was three-times more common in the
prasugrel group (0.4 vs. 0.1 percent; p=0.002).
"Platelet function while on DAPT is highly variable within genetic groups — because CYP2C19
genotype and platelet reactivity are imperfect correlates of each
other."
Add
to that the trial limitations. Genotype-guided therapy was not
randomized and antiplatelet therapy selection was left to physician
discretion in TRITON-TIMI 38. Finally, outcomes were ascertained from
electronic records, without adjudication, and bleeding outcomes were not
systematically collected — all worrisome limitations. The trialists
themselves point out that the number of patients needed to genotype,
with alternative antiplatelet therapy then prescribed for all patients
with LOF alleles, to prevent one cardiovascular event was 93.
But wait — there’s more! Alternative P2Y12 inhibitors are more expensive than generic clopidogrel and their
increased bleeding risk can deter its use. Safety is a problem. Even
though TRITON-TIMI 38 patients were not genetically tested but simply
randomized to clopidogrel or ticagrelor, the differences in efficacy
outcomes across the board (though arguably small) favor prasugrel.
One option already commonly used is to simply prescribe an alternative P2Y12 inhibitor. Alternatively, should we genetically test all patients
needing DAPT and use clopidogrel only in patients without LOF alleles?
The problems with this strategy are that genetic testing is not always
immediately available and it is expensive.
There
is more information to help with this dilemma. In a secondary analysis
of patients from TRITON-TIMI 38 patients with the LOF allele were
estimated to have a substantial reduction in the risk of cardiovascular
death, MI or stroke when treated with prasugrel compared with those
treated with clopidogrel (relative risk=0.57).4 The analysis went a bit
further, splitting out three groups of patients.
The
overall group treated in a randomized fashion (without genetic testing)
with either prasugrel or clopidogrel showed the differences in
cardiovascular outcomes seen overall in TRITON-TIMI 38. Prasugrel was
the winner — but not by much. The second group, testing negative for the
LOF allele and able to metabolize clopidogrel to its active form,
favored clopidogrel therapy for cardiovascular death and major and minor
bleeding outcomes. But for the third group (those testing positive for
the LOF allele) prasugrel therapy resulted in substantially less
cardiovascular death and nonfatal MI — albeit at the cost of more
bleeding.
One
must conclude from this secondary analysis that genetic testing can
indeed identify patients best treated with prasugrel or ticagrelor.
Perhaps testing patients for “high on-treatment” platelet reactivity and
making the appropriate switch from clopidogrel is another answer.
The one-year outcomes from PRAGUE-18 were published in January.5 It was designed to answer the question of whether prasugrel or
ticagrelor therapy produced better outcomes in patients with acute MI
(about 5 percent of enrollees had high-risk NSTEMI). In short, the
outcomes showed no differences between the two groups. However, the
actual outcome numbers were well below those in TRITON-TIMI 38, perhaps
indicating a lower overall risk in the PRAGUE-18 participants or other
confounding factors.
Both
drugs provided good results in PRAGUE-18. Cardiovascular death,
nonfatal MI or stroke at one year were 6.6 vs. 5.7 percent (p=0.50),
respectively, for prasugrel vs. ticagrelor. The authors concluded that
clinical use of either drug was acceptable. But then came a surprise.
The study also looked at outcomes of patients who switched from their
assigned study drug to clopidogrel — mostly for economic reasons. About a
third of patients on prasugrel and about 45 percent of patients
assigned to ticagrelor switched to clopidogrel. Patients who continued
their original study medications had a higher risk of a cardiovascular
event compared with the group who were economically motivated to switch
to clopidogrel. Further analysis showed that the patients who switched
to clopidogrel had fewer risk factors, which could well explain the
differences.
But
an intriguing question remains: Is there a strategy for patients such
as those in PRAGUE-18 that would allow an early switch from an expensive
medication (prasugrel or ticagrelor that also has greater bleeding
risk) to a less expensive, lower-bleeding risk drug (clopidogrel) with
associated good outcomes?
"TROPICAL
ACS points out that an individualized strategy based on genetic testing
and early de-escalation of antiplatelet treatment can be considered as
an alternative approach in patients with ACS managed with PCI."
This strategy is supported by the recent TROPICAL ACS trial.6 This trial divided patients with ACS into standard treatment with
prasugrel for 12 months (control group) or a step-down regimen (one week
of prasugrel followed by one week of clopidogrel and genetic
testing-guided maintenance therapy with clopidogrel or prasugrel from
day 14 after hospital discharge (guided de-escalation group). Despite
the medication switch, there was no increase in the combined risk of
cardiovascular death, MI or stroke in the de-escalation group vs. the
control group. TROPICAL ACS points out that an individualized strategy
based on genetic testing and early de-escalation of antiplatelet
treatment can be considered as an alternative approach in patients with
ACS managed with PCI.
Thus,
there is little doubt that clinical use of CYP2C19 genotyping to guide
post-PCI antiplatelet strategy provides information that can result in
better outcomes. How this is best used clinically remains unanswered.
DAPT strategy is still based primarily on retrospective analyses and not
large randomized trials. Such a randomized trial is currently underway
and hopefully will be completed by 2020. TAILOR-PCI (Tailored
Antiplatelet Therapy Following PCI) is designed to determine if genetic
testing can identify the best antiplatelet therapy for patients
undergoing PCI. Enrollees will be randomized to a conventional therapy
arm (clopidogrel) without prospective genotyping guidance vs. a
prospective CYP2C19 genotype-based antiplatelet therapy approach
(ticagrelor 90 mg bid) in reduced-function allele patients and
clopidogrel (75 mg qd) in non-LOF allele patients.
I
think we can predict the outcomes of TAILOR-PCI from what we already
know. Yet, having a supporting, large randomized trial will be helpful.
Perhaps the investigators will also tally those patients who opt out
early after initiation of DAPT using ticagrelor and switch to
clopidogrel to see if the intriguing strategy of an early post-PCI
switch from ticagrelor to clopidogrel suggested in the PRAGUE-18
one-year follow-up is clinically sound.
References
1. Cavallari LH, Lee CR, Beitelshees AL, et al. JACC: Cardiovasc Interv 2018;11:181-91.
2. Wallentin L, Becker RC, Budaj A, et al. N Engl J Med 2009;361:1045-57.
3. Wiviott SD, Braunwald E, McCabe CH, et al. N Engl J Med 2007;357:2001-15.
4. Sorich MJ, Vitry A, Ward MB, et al. J Thromb Haemost 2010;8:1678-84.
5. Motovska Z, Hlinomaz O, Kala P, et al. J Am Coll Cardiol 2018;71:371-81.
6. Sibbing D, Aradi D, Jacobshagen C, et al. Lancet 2017;390:1747-57.