Holding the Government Accountable

FDA Set Low Bar for Approval of Blood Thinners

See other articles in POGO’s “Drug Money” Series: Part 1 and Part 2.

The Project On Government Oversight’s past reporting on prescription blood thinners illustrates how low the FDA has set the bar for drug approval and what the FDA has glossed over or overlooked along the way.

For example, in 2011, the FDA approved the blood thinner Xarelto to prevent strokes in heart patients despite the fact that the manufacturer-sponsored clinical trial on which the FDA based its decision had a potentially crippling flaw: it relied on faulty blood testing devices that had already been the subject of two FDA warning letters.

One of the companies behind the drug, Johnson & Johnson, had disclosed in a submission to the FDA that it employed the devices. In addition, the published protocol for the clinical trial disclosed that other blood test data were gathered during the trial—data the FDA could have used to check the performance of the suspect devices before accepting the clinical trial results.

The FDA got around to conducting such a comparison in a report it issued on October 11, 2016. That was almost five years after it approved Xarelto, paving the way for billions of dollars in sales and millions of prescriptions; almost two years after a notice posted on the FDA website said the devices were prone to potentially fatal errors; almost a year after POGO reported that the clinical trial used the faulty devices; and three months after the manufacturer of the testing devices announced that it was withdrawing them from the market.

The FDA concluded that patients in the Xarelto trial’s control group “may have been unintentionally over-anticoagulated to some degree.” Left to speculate about how much that overdosing skewed the trial’s results, the FDA opined that the effect was “too small” to alter its original assessment that, compared to a much older drug, Xarelto’s benefits outweighed its risks.

As POGO reported, to win FDA approval, drug makers are supposed to show that the new drug is “non-inferior” to one already on the market. However, the FDA uses the term “non-inferior” loosely. In the case of Xarelto, the FDA set the bar so low that the product could be approved if it was proven only 50 percent as effective as an inexpensive generic drug in use since the 1950s.

Then there was Pradaxa, a new-generation blood thinner approved in 2010 and later named in many reports of deadly bleeding. The FDA approved Pradaxa on October 19, 2010, precisely the target date for FDA action—also known as the “PDUFA date”—dictated by the user fee regime’s accelerated timetables.

As part of its drug reviews, the FDA inspects a sampling of medical sites where clinical trial data were gathered, but FDA records show that the agency had not completed its assessment of inspection results when it approved Pradaxa. In addition, as POGO reported, an FDA memo dated October 13, 2010, shows that the agency’s Division of Scientific Investigations had not yet received or reviewed key inspection reports almost two months after the FDA staff completed a major paper reviewing Pradaxa, almost a month after an FDA advisory committee voted unanimously to endorse the drug, and mere days before the FDA announced its October 19, 2010, decision approving the drug.

As POGO reported in “Drug Problems: Dangerous Decisionmaking at the FDA,” the FDA approved Pradaxa in the absence of an antidote that could thicken the blood and stop patients from bleeding uncontrollably. Five years later, the FDA approved a reversal agent, saying there were situations in which reversing the drug’s effects was “medically necessary.” Still, the FDA has allowed Xarelto, a competing drug, to stay on the market without a reversal agent.