GRAVITAS: No Benefit of Doubling Dose in Clopidogrel Nonresponders

November 16, 2010

November 16, 2010 (Chicago, Illinois) — There was no benefit on cardiovascular outcomes or stent thrombosis with a double dose of clopidogrel in patients receiving drug-eluting stents with high residual platelet activity on the regular clopidogrel dose in the GRAVITAS trial.

Presenting the results to the media early in the day at the American Heart Association 2010 Scientific Sessions, lead investigator Dr Matthew Price (Scripps Clinic, La Jolla, CA) concluded: "In post-PCI patients with low-risk clinical presentations, a treatment strategy of high-dose clopidogrel for high residual reactivity on platelet-function testing is not warranted."

"Alternative Treatment Strategies Should Be Tested"

Dr Matthew Price

He added: "The high dose of clopidogrel doesn't appear to improve outcomes, so alternative treatment strategies should be tested." He said he expects these disappointing results to change the way cardiologists treat these patients. "Many physicians have been using high-dose clopidogrel as a default strategy in patients who are nonresponsive to the drug. We show that this strategy is probably ineffective."

Designated discussion of the trial, Dr Jessica Mega (Brigham and Women's Hospital, Boston, MA) said: "It is important to underscore that this trial does not definitely rule out the benefit of tailoring therapy based on platelet-function testing, but it suggests that in future we should be looking at higher-risk populations, larger cohorts, and different drugs."

Dr Jessica Mega

In his presentation, Price explained that ex vivo platelet reactivity varies widely among individuals on clopidogrel therapy. While high residual platelet reactivity on the regular clopidogrel dose has been strongly associated with an increased risk of cardiovascular events after PCI, a treatment strategy has not been defined for these patients.

In the GRAVITAS trial, 5429 patients on the regular clopidogrel dose underwent platelet-function tests with the VerifyNow assay (Accumetrics, San Diego, CA) 12 to 24 hours after PCI. Of these, 2214 (41%) had high residual platelet reactivity (platelet reactivity units [PRU] >230) and were randomized to continue on the 75-mg regular clopidogrel dose or to receive another 600-mg loading dose and a higher maintenance dose of 150 mg daily.

Baseline characteristics showed that most patients were at relatively low risk, with 80% having stable coronary artery disease.

At six months of follow-up, the composite end point of cardiovascular death/MI/stent thrombosis was identical in both groups, at 2.3%.

GRAVITAS: Primary End Point

End point

High-dose clopidogrel (%)

Standard-dose clopidogrel (%)

HR (95% CI)


Cardiovascular death/MI/stent thrombosis



1.01 (0.58-1.76)


Stent thrombosis occurred in 0.5% of the high-dose group and 0.7% of the standard-dose group, a nonsignificant difference.

There was also no difference in bleeding.

GRAVITAS: Bleeding Results


High-dose clopidogrel (%)

Standard-dose clopidogrel (%)


GUSTO severe/moderate bleeding




Any GUSTO bleeding




Dr Eric Topol (Scripps Clinic), chair of the executive committee of the trial, commented to heartwire : "We showed, for the first time, that double-dose clopidogrel for six months [in patients undergoing PCI] was safe, but that a single platelet-function test was inadequate to properly guide its use."

Only a Modest Effect on Platelet Function

Price reported that the percentage of patients who had persistently high platelet reactivity (PRU >230) was significantly reduced at 30 days and at six months.

GRAVITAS: Percentage of Patients With Persistently High Platelet Reactivity at 30 Days

Platelet reactivity units

High-dose clopidogrel (%)

Standard-dose clopidogrel (%)






Results at six months were virtually the same as those at three months.

Price told heartwire that the absolute platelet reactivity was reduced from an average of about 280 PRU at baseline to 200 PRU in the high-dose clopidogrel group vs 240 PRU in the standard-dose group. "This is only a modest reduction, and it is obviously not enough to show any effect on event rates, so I would say the pharmacodynamic effect is consistent with the clinical effect," he commented. He added: "We can tell the increased dose was not doing much because there was also no effect on bleeding."

Perhaps we should be treating to a lower target of platelet reactivity.

Price suggested that if this trial were conducted again, it would have been better to use much higher doses of clopidogrel, but he pointed out that this was not known at the time the trial was started. "There was a paucity of data on the effect of doubling the dose of clopidogrel on platelet reactivity, and it was prior to the approval of prasugrel, so at the time the trial started, doubling the dose of clopidogrel seemed like a good idea for these patients."

Noting that some patients did show a good reduction in platelet activity with the higher dose, Price suggested that future trials should incorporate serial platelet tests with serial individual adjustments of the clopidogrel dose according to the platelet-function results. "Perhaps we should be treating to a lower target of platelet reactivity, something like 200 PRU rather than 230 PRU," he said.

What Do We Do Now?

Answering the question of "What do we do now?" Price said: "If you were measuring platelet function before GRAVITAS, then there is nothing in these results to tell you to stop doing this. But what it does tell you is that we shouldn't be treating the clopidogrel nonresponders with a double dose of that drug. We need to be using a stronger drug."

A Boost for Prasugrel and Ticagrelor

Dr Elliot Antman

Dr Elliot Antman (Brigham and Women's Hospital, Boston), who was one of the lead investigators in the TRITON trial with prasugrel, said the GRAVITAS results might persuade more people to use prasugrel or ticagrelor across the board, rather than having to worry about platelet-function testing and what to do in the clopidogrel nonresponders.

On the idea of using a platelet-function test to select out the nonresponders to clopidogrel and then just treat these patients with one of the more potent agents, Antman said that this was a possible strategy, but he pointed out that because platelet function is so variable, multiple tests would probably be needed to make sure the correct individuals were targeted, and this would be just as expensive, if not more so, than using either prasugrel or ticagrelor in everyone. "This is not really practical. It is probably simpler just to use one of the other agents at baseline," he commented.

Gene Testing May Be Better?

Based on this trial, I am looking more favorably toward the genetic approach rather than platelet function testing.

Antman believes that if any measure of clopidogrel variability was going to be done, this trial might tip people in the direction of the gene test. "Based on this trial, I am looking more favorably toward the genetic approach rather than platelet-function testing. Unlike platelet function, your gene status stays the same forever, so you only need one test, and those results can be put in your medical records and will never change. If you find a patient who is a carrier of the loss-of-function alleles, you have them marked down at being at higher risk with clopidogrel for life."

Price noted that they are looking at genomics in a GRAVITAS substudy called GIFT, but these results are not yet available. "My hope is that platelet-function testing and genomics will be complementary," he added.

Underpowered Trial

Mega suggested GRAVITAS was underpowered to show an effect, as the 50% relative risk reduction aimed for was perhaps too ambitious and the reduction in platelet activity with the double dose was too low. In addition, the researchers were expecting an event rate of around 5%, but the actual rate was half that. "So while they set out for a power of 80% to show a difference, they probably only actually had a power of 12%."

But Price replied that because the event rates and bleeding were not at all different in the two groups, he did not think a larger trial with more power would find any meaningful difference in event rates between the two clopidogrel doses in this population.

Were the Right Patients Selected?

Dr Dirk Sibbing

Commenting on the GRAVITAS results for heartwire , Dr Dirk Sibbing (Deutsches Herzzentrum Munich, Germany) suggested two reasons for the negative results. "First, the proportion of patients with high on-treatment platelet reactivity (>40%) seems rather high and does not match with clinical experience in PCI-treated patients. In the large majority of patients, clopidogrel works very well and is successful in avoiding thrombotic complications. Second, in a well-defined cohort of real nonresponders, just doubling the dose of clopidogrel is not the best solution. This group of patients needs alternative and more potent agents."

Sibbing added: "Importantly, besides the VerifyNow assay, other devices for platelet-function testing are available, and we must also compare these devices for identifying the patients who benefit most from an intensified antiplatelet treatment."

Other Trials Under Way

Price told heartwire that several other trials are under way looking at other strategies in patients with high residual platelet activity on clopidogrel. These included TRIGGER-PCI , which is using prasugrel, and TARGET-PCI using serial platelet-function testing.

The Results on Those Who Kept Taking 75-Mg Clopidogrel

The GRAVITAS trial also involved a third observational arm, in which 5429 patients underwent platelet-function testing on the regular clopidogrel dose after PCI and were divided into those with high residual platelet reactivity and those without high residual activity, again using the cutoff of 230 PRU. All patients were then just continued on the regular 75-mg dose of clopidogrel and followed for six months.

Results showed that the composite of cardiovascular death/MI/stent thrombosis was higher in the high-residual-activity group (HR 1.68), but not significantly so. Price commented: "Although not significant, the large hazard ratio is consistent with previous observational studies that document increased events after PCI in patients with high residual platelet activity."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.