At AdverseEvents, we believe that monitoring and reviewing post-approval side effect data is vital to the analysis of a drug’s true safety profile. As we discuss in a recent White Paper entitled Post FDA-approval drug safety data: why they are vital and how they can be made accessible, actionable, and predictable, pre-approval clinical trials are, by necessity, conducted in relatively homogenous group. Trial subjects may react in similar ways to a test drug, and therefore, the clinical trail process does not uncover many of the side effects that occur once the drug is introduced to real-word, heterogeneous, patient populations. Therefore, a drug’s true side effect profile is almost never properly understood until many months to years after FDA approval.
A front-page article in yesterday’s Wall Street Journal brought this matter to the forefront once again with allegations of under-reporting of deaths and other serious outcomes and side effects during a major clinical trial of the anticoagulant drug Brilinta.
Oral anticoagulants such as warfarin and antiplatelet agents such as aspirin and clopidogrel (Plavix) have been on the market for decades. They are prescribed for patients with various venous and arterial thromboembolic disorders. Newer classes of anticoagulant drugs include direct thrombin inhibitors (dabigatran, rivaroxaban, and apixaban), while new antiplatelet medications include ticagrelor (Brilinta) and prasugrel. These drugs represent some of the most widely used medicines in the world.
Some of the more eye-catching allegations mentioned in the WSJ article included:
- “the number of deaths in the study's control group - those taking a competing drug - was unusually high compared with earlier trials,”
- “12 patients reported their own deaths by telephone,”
- “in the U.S. portion of the overall study, among 1,413 patients, Brilinta actually did worse than Plavix and was linked to a 27% greater incidence of vascular deaths, heart attacks and stroke,”
- “Dr. Thomas Marciniak, the FDA review-team leader who asked Dr. Serebruany to get involved stated "literally this submission is the worst submission I have encountered for collecting - or at least submitting - information on serious adverse events,”
- a partial recount by Duke University's Clinical Research Institute added an extra 45 heart attacks to the Plavix group, and
- “Dr. Serebruany said there were just two countries, Poland and Hungary, that showed a statistically significant advantage for Brilinta. He said these two countries accounted for 21% of the enrolled patients but 46% of the excess strokes, heart attacks and vascular deaths experienced by those on Plavix.”
In response, AstraZeneca defended the “Plato” trial in question and the efficacy of Brilinta. They further noted that the drug had been “"extensively reviewed by top-tier journals and approved in over 100 countries.” Additionally, the company stated that if the results for Poland and Hungry were omitted, the trial still would have demonstrated that Brilinita was superior to Plavix.
In order to better understand the post-approval safety data available on these drugs, we conducted a detailed review of the FDA’s large database of post-marketing side effects, the Adverse Event Reporting Database (FAERS), in order to compare Brilinta (ticagrelor) with Plavix (clopidogrel).
Using FAERS data (from 7/20/2011 to 12/31/2012), aggregated and standardized by the AdverseEvent RxFilter process, we identified 2,280 serious adverse events that listed Brilinta as the primary suspect. Of these reports, we identified 68 cases of stroke and 180 cases of heart attack. The most commonly reported side effects were shortness of breath, myocardial infarction (heart attack), and death. We identified 861 hospitalizations and 317 patient deaths where Brilinta was indicated as the primary suspect.
For Plavix, using FAERS data from 11/01/1997 to 12/31/2012, we identified 11,359 serious adverse event reports that listed it as the primary suspect. Of these reports, we identified 547 cases of stroke and 549 cases of heart attack. The most commonly reported side effects were death, gastrointestinal hemorrhage, and anemia. We identified 4,922 hospitalizations and 2,221 patient deaths where Plavix or its generic equivalent was indicated as the primary suspect.
Disproportionality measures are used to determine the “unexpectedness” of a given adverse event/drug “pair.” The methods calculate an “expected” rate of the adverse event by averaging all counts of that adverse event across all drugs in the FAERS database. The Reporting Odds Ratio (ROR) is the disproportionality measure we use and it represents the ratio of incidence of the adverse event in the ‘‘exposed reports’’ to ‘‘unexposed reports.” We derive ROR by the use of standard formulas.
The tables below list the number of “primary suspect” case reports for each drug and the corresponding “Preferred Term” adverse event categories (as defined by MedDRA (http://www.meddra.org/)). RORs were calculated for the “primary suspect” cases and are listed in the right hand columns.
Date range for calculations, below: Plavix (11/17/1997 - 2/3/2014) and Brilinta (7/20/2011 - 2/3/2014).
Clinical trials are the established means for determining a drug’s safety and efficacy during the approval process, but they are by no means perfect. When a new drug comes to market, it is used by a more heterogeneous population and real-world side effects begin to appear.
When the clinical trial itself is called into question, the need for post-approval monitoring is even more pronounced.
Our review of post-approval data suggests disproportionally elevated reporting of certain adverse events such as various hemorrhages and hematomas linked to both Brilinta and Plavix.
Between the drugs, many of the increased disproportionally scores were similar, except for pulmonary haemorrhage and haemorrhage where Brilinta had higher RORs than Plavix.
When we tallied case counts for “outcome,” we found that Plavix had 2,221 “primary suspect” cases of “death” (representing 20% of its’ total primary cases), while Brilinta had 317 cases (representing 14% of its’ primary suspect cases).
To see more information on Brilinta, Plavix, and the 4,100 other drugs contained in our platform, register for a complimentary 7 day trial of AdverseEvents Explorer.
While these data support the black box warnings already added to both the drug’s labels, they cannot definitively ascertain causality or incidence. Post-marketing data may be subject to many biases, such as underreporting, stimulated reporting, and confounding by comorbidities. Despite such limitations, we hope these data offer an update for clinicians, while we wait for further information regarding the Brilinta lawsuit filed by Dr. Serebruany.
Related Read: Drug Safety Advocate Call to Action
Neither AdverseEvents, nor its officers or employees currently hold, and will not hold, a direct or indirect investment position (long or short) in any of the securities mentioned in this report.
Neither AdverseEvents, nor its officers or employees have been directly compensated by any party for the preparation of this report. AdverseEvents offers its services for sale to enterprise customers, including managed care organizations, financial institutions, pharmaceutical companies, and others.
The inclusion of a particular company, drug, class or indication in this report is determined wholly by our quantitative signaling and scoring systems along with our qualitative analysis work. The inclusion or exclusion of any drug, company, or indication has not and will not be influenced by any third party, including any clients of AdverseEvents.
For more reports, go to our RESOURCE CENTER
Keith Hoffman, PhD
Vice President, Scientific Affairs