FDA study spanks drugmakers for poor adverse events reporting

Gilead Sciences' ($GILD) hep C drug Sovaldi (sofosbuvir) has had a stellar year, with huge sales because of cure rates that top 90%. But is it safe? According to adverse events reporting, the FDA says it appears to be, but then that is a bit hard to determine. That, the agency says, is because Gilead, like many other drugmakers, has done such a poor job of reporting drug side effects.

The new report on adverse reporting says drugmakers in general do a bad job of getting basic information, rendering unreliable the FDA's primary tool to track the safety of drugs after approval. The report points out that drugmakers account for 96% of the reports it gets but fewer than half are considered complete. That compares with the FDA-collected reports, which are complete 85% of the time. The study looked at 847,000 reports made during the 12 months that ended March 31, 2014.

As The New York Times points out, the system has some basic flaws, which has left it open to criticism by consumer groups. Doctors and patients provide information voluntarily. Drugmakers are required to report only when they learn about a problem. It is believed the system picks up only a small percent of the total adverse reactions to drugs. The shortcoming is not specific to the U.S. In 2012, Roche ($RHHBY) got into some trouble with U.K. regulators when they discovered 80,000 uninvestigated adverse reaction reports from the U.S., including 15,000 reports of deaths. They had been collected through a Roche-sponsored patient support program but were not passed along for investigation.

Using Sovaldi as an illustration of shortcomings in the system, the study points out the drug was approved without a controlled efficacy trial for its largest patient population. While there was no indication of problems from adverse events, it said only 39% of the reports Gilead ($GILD) turned in were "reasonably complete." It went on to say, "With limited pre-approval testing and weak adverse event reporting, the safety profile of sofosbuvir remains uncertain."

Gilead told The New York Times that it made repeated efforts to get info but that what it gets from doctors and patients is often sketchy and that FDA rules give it little time to follow up.

Gilead was not at all the only drugmaker called out in the report. In fact, the study by the Institute for Safe Medicine Practices scolds drugmakers across the board, saying that it sees no reason why they have more trouble getting basic information like age and sex of a patient, than the FDA.

The report includes tables of the companies that turn in the most reports--Pfizer ($PFE), Novartis ($NVS) and Roche among them--as well as those who do the best job and those that have "weak" reporting. Kudos went to companies that include Ariad Pharmaceuticals ($ARIA), Vertex ($VRTX) and Biogen Idec ($BIIB), while those named for poor reporting include Par Pharmaceutical, Takeda Pharmaceutical's Millennium and Cubist Pharmaceuticals, acquired last year by Merck ($MRK).

The FDA takes some of the heat itself for a system that is not doing the job it is intended to do. The study says the FDA needs to modernize its requirements to meet the digital age. It says the agency needs to do a better job of helping drugmakers. "Providing FDA feedback to companies submitting large numbers of reports with missing data is elementary quality assurance in the digital data era," it says.

It explains the agency is looking into big data mining as a way to get more and more useful reports. It points to its Sentinel system which collects data from insurance records as an import resource, but it says the FAERS reporting system remains the primary avenue for learning of problems and needs to be fixed. "It makes no sense for drug manufacturers to be required to spend millions collecting and submitting adverse drug event reports promptly when so many reports contribute little to the assessment of drug safety," the report concludes.

- here's the study (PDF)
- read the New York Times story (sub. req.)

Read more on