ProPublica

Journalism in the Public Interest

Cancel

GE’s Own Safety Team Urged Company to Restrict MRI Drug

GE Healthcare ignored the advice of its own safety experts to “proactively” restrict the use of its imaging drug, Omniscan, after reports in Europe linked the drug to a potentially crippling disease, according to a newly unsealed order in a lawsuit against the company.

A GE Healthcare event in New York City. (Daniel Acker/Bloomberg)

GE Healthcare ignored the advice of its own safety experts to “proactively” restrict the use of its imaging drug, Omniscan, after reports in Europe linked the drug to a potentially crippling disease, according to a newly unsealed order in a lawsuit against the company.

The recommendation came at a May 2006 meeting convened by the company's vice president for drug safety. But instead of immediately alerting doctors to stop using the drug in high-risk patients, GE spent the next year arguing that approach wasn't necessary, even as some government and radiological experts favored such a ban.

GE publicly took the position that its drug was no more dangerous than those of its competitors and argued against putting an exclusive warning on the drug’s label. It wasn’t until September 2007, after pressure from federal drug regulators, that GE and manufacturers of similar drugs revised their labels to warn that patients with serious kidney impairment have a greater risk of nephrogenic systemic fibrosis, or NSF.

The order by a Chicago judge became public this week in one of the approximately 500 lawsuits filed by NSF patients against GE. It provides the first public glimpse at internal documents and other evidence the company has successfully fought to keep sealed.

Plaintiffs’ lawyers say the May 10, 2006, meeting is pivotal because many patients were exposed to Omniscan afterward. According to the judge’s order, an “action” item at the end of the meeting minutes stated, “We should proactively propose to restrict the use of Omniscan in patients with severely impaired renal function.”

The order also references previously sealed evidence indicating that doctors and the U.S. Food and Drug Administration had not seen internal company research that raised questions about Omniscan’s propensity to break down chemically, releasing the potentially toxic metal gadolinium into the body.

Cook County Circuit Judge Deborah Mary Dooling said the evidence she reviewed was sufficient to add a claim of punitive damages, which could have greatly increased GE’s financial exposure. But with a trial set for April 20, the parties reached a confidential settlement.

(Document Viewer: Read the judge's order)

Jeff DeMarrais, a spokesman for GE Healthcare, said Thursday that by the time of the May 2006 meeting, the company had already issued a “global safety alert” to regulators. A few weeks later, GE also sent a letter to doctors about the NSF cases stating that a “causal relationship” between NSF and imaging drugs “has not been established.”

“The company promptly and proactively engaged with regulatory agencies worldwide consistent with Dr. Flaten’s recommendation,” he said, referring to Dr. Hugo Flaten, the vice president of Global Pharmacovigilance, who led the internal meeting.

DeMarrais also said GE has been dismissed from 140 Omniscan cases without any payment and that it has settled “a number of cases,” mostly in state courts. A substantial majority of the cases against GE are in federal courts.

He declined to say how many NSF patients had been exposed to Omniscan after the May 2006 meeting. “It is difficult to put an exact number on this,” DeMarrais said in an e-mail.

Imaging agents such as Omniscan have become an essential part of medical diagnostics, with millions of doses safely administered worldwide each year. Patients are injected with the drugs, which make it easier to read MRI scans. Omniscan belongs to a class that use gadolinium as the key ingredient. Healthy kidneys normally filter out the metal.

Concern about Omniscan first surfaced that April when researchers in Denmark and Austria published studies saying 25 patients, all of them with severely impaired kidneys, had contracted NSF after undergoing magnetic resonance scans with Omniscan. The disease can cause a painful hardening of skin around joints and can affect internal organs.

A few of GE’s competitors also have been sued by NSF patients. But some doctors and drug safety experts, including key medical reviewers inside the FDA, have singled out Omniscan as riskier, in part because of studies suggesting it is less chemically stable. Research by GE Healthcare and other manufacturers is ongoing into the causes of NSF and the possible role of gadolinium, a key ingredient in all the drugs.

GE denies that Omniscan is less stable. The company has maintained, and the FDA so far has agreed, that because “differential risk remains unproven” the various contrast agents on the market should be treated uniformly as a class.

Dr. Sidney M. Wolfe, who heads the health research arm of the consumer group Public Citizen, said it was troubling that Omniscan research may have been withheld.

“I'm concerned, after reading the court order, and as a member of the FDA's Drug Safety Advisory Committee, that there may be evidence pertinent to the safety of this drug that was not promptly or not completely sent to the FDA,” Wolfe said.

Last December, Wolfe and most other members of an FDA advisory panel recommended that the agency revise its policy and effectively ban use of Omniscan and one other contrast agent, Optimark, for patients with severe kidney disease.

The daylong hearing, and the accompanying submissions by GE and the FDA, did not reference some of the internal studies cited by the Judge Dooling, including a 1989 draft report by a GE Healthcare predecessor company, Nycomed, whose scientists were conducting toxicity studies before Omniscan was approved by the FDA in 1993.

According to judge’s order, the draft Nycomed report “concluded that Omniscan caused the highest retention of gadolinium in the liver as compared to other gadolinium contrast agents tested.” The finding “could be taken as an indication for the relative instability of [Omniscan],” Dooling wrote, quoting an exhibit in the case.

GE did not dispute that the draft report was never submitted to the FDA, Dooling’s order states, but said a final version of the Nycomed study dated May 28, 1991, “was submitted to the FDA with complete findings and conclusions.” The judge noted, however, that the final report “does not reference” findings in the 1989 draft.

GE became the owner of Omniscan in 2004, when it acquired Amersham, a British maker of scanning drugs. Nycomed had become part of Amersham several years earlier.

DeMarrais said Thursday that GE “and its predecessors have fully and responsibly complied with their regulatory obligations in submitting data to the FDA.”

The FDA is still reviewing the advisory panel’s recommendation to restrict Omniscan and Optimark, according to Karen Riley, a spokeswoman for the agency.

In spring of 2007, while the FDA considered a labeling change for gadolinium-based imaging agents, two medical reviewers at the agency independently recommended that Omniscan be banned for patients with severe kidney disease.

An outside expert had the same opinion. Dr. Emanuel Kanal, head of a blue-ribbon panel on magnetic resonance safety for the American College of Radiology, said this week that he was “surprised” to learn that GE insiders also had advised restricting the drug.

Judge Dooling’s order allowing a claim for punitive damages was signed April 2 and made public this week.

DeMarrais said the “order was a preliminary ruling under a unique Illinois procedural rule” and “did not mean that a claim for punitive damages would have been allowed to be submitted to a jury for consideration.”

But Tor Hoerman, the attorney for 57-year-old plaintiff Robbie Booker, differed. The judge “had an opportunity to analyze a lot of evidence, and chose to enter a rather detailed, fact-based order rather than a brief procedure-based order,” he said.

Dooling reasoned that there was a "reasonable likelihood" the plaintiffs might prove to a jury that GE engaged in willful misconduct, the standard for punitive damages. If the evidence were proved at trial, she wrote, the "jury could properly conclude" that GE Healthcare "engaged in a course of action that showed an utter indifference to or conscious disregard for the safety" of patients with impaired kidneys.

Hoerman noted that Dooling’s opinion does not apply to the hundreds of federal NSF cases being coordinated by a judge in Cleveland. That judge, in a ruling last fall, allowed GE Healthcare to keep sealed, for the time being, internal corporate documents.

Last year, lawyers for the NSF plaintiffs estimated the litigation could cost the conglomerate $1 billion. GE vigorously disputes this figure, saying it is grossly inflated and that its exposure is so small there is no need to disclose the lawsuits to investors.

Booker’s lawsuit says she had severe kidney disease for more than a decade and had NSF diagnosed in October 2007. The disease left her completely disabled, profoundly disfigured and suffering from severe leg, arm and shoulder pain.

Booker says she received 11 injections of Omniscan, all but one of which took place before 2007.

Jeff DeMarrais

April 16, 2010, 1:42 p.m.

Jeff DeMarrais from GE Healthcare - I apologize for the long post but sometimes context is important.

ProPublica’s April 15, 2010, article, “GE’s Own Safety Team Urged Company to Restrict MRI Drug” is an incomplete and gross mischaracterization of GE’s actions. It casts GE in a false and negative light, as it omits key facts, despite our consistent engagement over several months to explain a complex issue difficult to capture in a short article.

As the facts below demonstrate, GE Healthcare moved quickly, proactively and responsibly in collaboration with health authorities to advise healthcare practitioners about the potential risk of using gadolinium-based contrast agents in renally impaired patients.

All actions and proposals were taken based on the available information at that time, and only with the approval and consent of global regulatory authorities.

GE was the first GBCA manufacturer to proactively advise healthcare practitioners of the potential risk of using GBCAs in renally impaired patients. As a result, there have been no new NSF cases reported to GE and plausibly associated with the administration of Omniscan with a documented date of onset after September 2007.

On May 10, 2006, in the meeting referred to in the article, Dr. Flaten, Global Head of Pharmacovigilence for GE Healthcare’s Medical Diagnostics division recommended consideration of a temporary restriction on the use of Omniscan in at risk patients until more information was known. GE Healthcare recognized that such a proposal could only be undertaken with the approval and consent of global regulatory authorities.

On May 11, 2006, Dr. Flaten met with the Danish Medicines Agency (“DMA,” the Danish equivalent of the US FDA) to discuss the NSF cases reported in Demark in April.  The DMA elected to warn the medical community of the potential but yet unknown risks through a website alert. Because of the limited medical and scientific information available at that time, the DMA did not restrict the use of Omniscan in renally impaired patients in Denmark.  The DMA also stated its intention to request that other contrast agent manufacturers review their pharmacovigilance data for any reports of NSF associated with their respective products. 

In close coordination with and the approval of both the FDA and the DMA, GE Healthcare had prepared both a Global Safety Alert to regulatory agencies worldwide (issued May 9, 2006), and a “Dear Healthcare Professional” letter to healthcare providers globally advising them of the reports of NSF following Omniscan administration in both Denmark and Austria. 

This letter was closely reviewed by FDA and sent to customers in the United States on June 6, 2006.  The ProPublica article inaccurately suggests the letter was intended to convince radiologists that Omniscan was not proven to cause NSF. This is not true. The letter alerted   doctors of a potential but newly-identified risk and, combined with the FDA’s public health alert put the medical community on notice that a potential risk existed.  As a result, radiologists across the United States were able to make informed decisions regarding the safe and effective use of Omniscan.

Importantly, GE Healthcare was the only GBCA manufacturer at the time to work with the regulatory agencies and communicate early and before more information was known. ProPublica’s article also references a draft report of a 1989 study by Nycomed. This study was designed to explore into patent issues as part of an acquisition’s due diligence – not to investigate the safety, toxicity, or efficacy of the product. The study was not suitable for that purpose. The findings of later studies, which were designed to investigate safety and efficacy, were submitted to and accepted by the FDA as part of the New Drug Application (NDA) in October 1990.

Not only did GE Healthcare fully and responsibly comply with their regulatory obligations in submitting data to FDA, the proactive nature of the Company’s actions demonstrate its commitment to work in full cooperation with the FDA and other regulatory bodies around the world.

Every case of NSF is distressing and we have the utmost compassion for those afflicted with it. Throughout this process, we have put patient safety first.

mmmm hmmmm

Jeff DeMarrais

May 5, 2010, 2:47 p.m.

Jeff DeMarrais from GE Healthcare - ProPublica has in its posession a copy of GE’s May 9, 2006 Public Health Alert which directly contradicts this article and its headline.

The letter clearly shows GE’s proactive communications to the medical community regarding the 25 known cases of NSF **before** the May 10, 2006 meeting upon which ProPublica based this column. It is signed by Dr. Hugo Flaten, the doctor who ProPublica claimed was “ignored.”

While we shared this timeline with ProPublica before it published its story, ProPublica can now add the May 9, 2006 Public Health Alert to its online archive so readers can decide for themselves.

We look forward to seeing the timeline corrected and future reporting that accurately portrays GE’s actions.

From the editors:

DeMarrais writes that, prior to publication, he shared GE’s “timeline” of actions in response to NSF reports. Accordingly, we reported the fact that GE Healthcare issued an alert prior to the May 10 meeting of the company’s drug safety officials. In advising us of the alert, DeMarrais did not provide a date. He did not provide a copy of the alert until May 5, more than two weeks after we requested it. Nor did DeMarrais provide pages that detail the NSF cases that prompted the alert. The May 9 alert does call attention to reports of NSF in Omniscan patients, but it stops short of advising health care providers to restrict use of Omniscan as GE’s Dr. Hugo Flaten and his drug safety group recommended the following day. The existence of the alert does not change the fact that GE did not implement the recommendation. A copy of the May 9 alert is here. Citing confidentiality, DeMarrais declined to provide an attachment, also sent to regulators, describing the NSF cases. He also declined to provide a version of the regulatory attachment without information that might identify the patients.

Some assertions in DeMarrais’ earlier comment imply that we omitted or distorted facts, and they deserve a further reply.

DeMarrais writes:  “In close coordination with and the approval of both the FDA and the DMA, GE Healthcare had prepared both a Global Safety Alert to regulatory agencies worldwide (issued May 9, 2006), and a “Dear Healthcare Professional” letter to healthcare providers globally advising them of the reports of NSF following Omniscan administration in both Denmark and Austria.”

Response: As noted above, the story did not neglect or misrepresent this information; in fact, it reports that GE sent both the letter and global alert.

DeMarrais writes: “ProPublica’s April 15, 2010, article, ‘GE’s Own Safety Team Urged Company to Restrict MRI Drug’ is an incomplete and gross mischaracterization of GE’s actions ... [GE]  moved quickly, proactively and responsibly in collaboration with health authorities to advise healthcare practitioners about the potential risk ...” of NSF.

Response: DeMarrais is entitled to his opinion. The story, however, is based on facts described in a ruling by Cook County Circuit Judge Deborah Mary Dooling in an Omniscan lawsuit. Our portrayal of GE’s handling of the NSF reports in April and May of 2006 hews to the judge’s even-handed summary of proposed evidence and testimony in the case.

Judge Dooling wrote that Flaten and GE’s safety group met on May 10, 2006, “to discuss the possible measures to be taken to protect patients.” Quoting the minutes of the meeting, she said “the group felt that precautionary measures should be taken, e.g., a temporary stop of use in patients with severely impaired renal function.” She wrote that “action” items that closed the minutes stated, “We should proactively propose to restrict the use of Omniscan in patients with severely impaired renal function.” Dooling continued: “However, instead of directly warning about the risks of Omniscan in renally impaired patients, [GE Healthcare] sent a “dear doctor” letter on June 6, 2006, which … concluded, “to date a causal relationship between [NSF] and [gadolinium-based] contrast agents has not been established.”

DeMarrais writes: “This (Dear Healthcare Professional) letter was closely reviewed by FDA and sent to customers in the United States on June 6, 2006.  The ProPublica article inaccurately suggests the letter was intended to convince radiologists that Omniscan was not proven to cause NSF. This is not true.”

Response:  The story does not dispute that GE was communicating with regulators about Omniscan and NSF. It simply quotes from GE’s letter, which asserts no “causal relationship” between NSF and drugs like GE’s. The story does add context, as we have reported separately. When the FDA considered a possible black box warning label the following year, GE argued that its drug wasn’t more dangerous than competing drugs in the same class. FDA medical experts found that Omniscan did present a higher risk of NSF, but the agency decided the label warning should apply broadly to Omniscan and other gadolinium-based contrast agents then on the market.

This article is part of an ongoing investigation:
Omniscan

Omniscan: Specter of MRI Disease Haunts General Electric

A rare disease linked to MRI scans has left GE fending off claims of liability.

The Story So Far

General Electric is in a liability fight over a rare and sometimes fatal disease that has been linked to a dye used for MRI scans, with a preponderance of cases involving a GE product called Omniscan.

The disease, nephrogenic systemic fibrosis, or NSF, isn’t fully understood, but nearly all cases have involved patients with kidney problems who were injected with MRI contrast agents.

More »

Get Updates

Stay on top of what we’re working on by subscribing to our email digest.

optional

Our Hottest Stories

  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •