AstraZeneca Papers Raise Seroquel Issues

By SHIRLEY S. WANG and AVERY JOHNSON

ORLANDO, Fla. -- AstraZeneca PLC instructed its U.S. sales representatives to tell doctors that its powerful psychiatric drug, Seroquel, didn't cause diabetes even though a company physician had at one point stated years earlier that such a link was probable in some individuals, documents unsealed in a federal court case here show.

The documents -- ranging from unpublished study results to previously undisclosed depositions -- are among more than 100 the U.K. drug maker agreed to unseal Thursday in lawsuits brought by plaintiffs who allege they were harmed by the multibillion-dollar antipsychotic drug. Many of the cases have been consolidated in the U.S. District Court for the Middle District of Florida.

In an Aug. 15, 2005, voicemail message addressed to company salespeople, an AstraZeneca employee named Christine Ney followed up on a "weight and diabetes sell sheet" they had recently been sent. The sales representatives should assuage doctors' fears about their patients' weight gains, she said in the voicemail, by telling them that data showed no causal link between diabetes and the drug.

"Our objective is to neutralize customer objections to Seroquel's weight and diabetes profile," Ms. Ney said, according to a transcript of the voicemail message. She then instructed representatives to "refocus the call" away from diabetes to the drug's tolerability, the transcript shows.

The voicemail to the sales representatives raises questions about whether there was a contradiction between Ms. Ney's instructions and an AstraZeneca drug-safety expert's own assessment of Seroquel's link to diabetes years earlier. In a 2000 position paper about the safety of Seroquel sent to Dutch regulatory authorities, an AstraZeneca doctor named Wayne Geller wrote that there was a relationship between the drug and diabetes.

"There is reasonable evidence to suggest that Seroquel therapy can cause impaired glucose regulation including diabetes melliutus in certain individuals," Dr. Geller wrote.

AstraZeneca acted "responsibly and appropriately as it developed and marketed Seroquel," company spokesman Tony Jewell says. "From the time it was first approved, the Seroquel labeling alerted physicians that diabetes mellitus, hyperglycemia and weight gain had been observed in clinical trials. We've continued to update the label as the findings have developed."

Mr. Jewell says the document written by Dr. Geller doesn't accurately reflect the company's position in 2000. "In fact, it was not Dr. Geller's ultimate view either. It was an initial draft for discussion purposes. After rigorous discussion of the scientific evidence, Dr. Geller and his colleagues concluded the evidence did not establish that Seroquel causes diabetes."

Dr. Geller retracted his statement in a May 2008 deposition with plaintiffs' attorneys. In response to a plaintiffs' attorney's question, Dr. Geller responded that the statement was "an artifact of an earlier discussion document." AstraZeneca declined to make Dr. Geller and Ms. Ney, who are both still employed by the company, available for comment. Neither could be reached for comment late Thursday night.

AstraZeneca had argued that the documents should remain under seal because they contained proprietary information that could hurt the company if it was revealed to competitors and could harm the public if interpreted out of context. But the company agreed to make public most of the documents following negotiations with plaintiffs' attorneys that concluded in the early hours of Thursday morning.

"AstraZeneca believes communications from the FDA [Food and Drug Administration] to doctors is the appropriate way to notify patients and the public about a medicine's benefits and risks, not a court proceeding," Mr. Jewell says.

Makers of antipsychotic drugs have come under intense scrutiny over whether they knew early on that the powerful psychiatric medicines -- which are used to treat schizophrenia and bipolar disorder -- caused serious side effects such as diabetes. Most of the drugs were approved for sale in the 1990s, but their side effects didn't become widely known until earlier this decade. The drug makers have also been accused of marketing the drugs outside of their approved indications, which is against U.S. law.

Internal company documents proved damaging to Eli Lilly & Co., which agreed to pay $1.4 billion to settle off-label promotion claims last month with the U.S. attorney for the Eastern District of Pennsylvania. Company e-mails and memos leaked in 2006 showed that Lilly played down the health risks of its antipsychotic drug, Zyprexa.

In 2007, Bristol-Myers Squibb Co. agreed to pay $515 million to settle another federal investigation over its marketing practices for its antipsychotic drug, Abilify.

AstraZeneca faces more than 9,000 lawsuits from patients who allege they have been harmed by Seroquel. Most of the cases have been consolidated into one group being heard in the Florida federal court. The first two of these cases, considered bellwethers that would have helped predict the outcomes of the remaining lawsuits, were supposed to have been tried earlier this month, but the judge dismissed them on the grounds that the evidence failed to prove that Seroquel had caused their diabetes. More than 2,300 other cases have also been dismissed. A separate tranche of state cases is also working its way through the courts, with the first set to go to trial in Delaware, where AstraZeneca has its U.S. headquarters. The company says it will defend itself vigorously against the suits.

Pennsylvania, Montana, Arkansas and South Carolina have all sued AstraZeneca, alleging that they were bilked into paying for the medicine for off-label usage and seeking reimbursement for alleged injuries sustained by individuals as a result of the drug. AstraZeneca acknowledges that the U.S. attorney for the Eastern District of Pennsylvania, Laurie Magid, is investigating Seroquel but wouldn't say when the investigation started or what it involves. A spokeswoman for Ms. Magid declined to comment.

When Seroquel was approved in 1997, it contained information in the "adverse reactions" section of the product label concerning diabetes, hyperglycemia and weight gain that had been observed in clinical trials, says Mr. Jewell. In 2003, the FDA mandated a stricter precaution, upgrading the diabetes risk to the label's "warnings" section -- even though the agency said at the time that it wasn't certain that a causal relationship existed.

Last month Seroquel's label was updated to include data on children and adolescents, including blood-glucose levels, cholesterol, weight gain and increased appetite. It also voluntarily agreed to move information about increased blood sugar to the "warnings" section of the label, according to a company lawyer's testimony in court Thursday.

Among the unsealed documents, another set goes back to the late 1990s and raises further questions about whether AstraZeneca kept a lid on unflattering Seroquel studies.

A document dated Feb. 12, 1997, describes internal company deliberations over how to report "Study 15," a study comparing Seroquel to Haldol, an older-generation psychiatric drug. In the document, an AstraZeneca employee named Richard Lawrence writes to his team that one of his colleagues had done a great "smoke and mirrors job," and another had suggested an approach that "should minimise (and dare I venture to suggest) could put a positive spin (in terms of safety) on this cursed study."

It isn't clear what Mr. Lawrence's comments were referring to. He couldn't be reached late Thursday, and AstraZeneca declined to discuss the emails. Ed Blizzard, an attorney with the Houston law firm Blizzard, McCarthy & Nabers, who is one of the lawyers for the plaintiffs, says the document refers to the fact that "the weight gain data was bad, and that's why we believe the study was buried."

In a series of emails in 1999, AstraZeneca executives discuss withholding the results of another Seroquel study, known as COSTAR. That study compared Seroquel to Risperdal, another antipsychotic drug made by Johnson & Johnson.

On Dec. 6, 1999, an AstraZeneca employee named John Tumas, the publications manager for Seroquel, wrote to his colleagues, "There is growing pressure from outside the industry to provide access to all data resulting from clinical trials conducted by industry. Thus far, we have buried" certain studies. Referring to COSTAR, he added: "We must find a way to diminish the negative findings. But in my opinion we cannot hide them."

Mr. Tumas couldn't be reached late Thursday. AstraZeneca's Mr. Jewell declined to comment about the 1990s e-mails and documents other than to note that the FDA "vetted and substantiated the safety data for Seroquel," including the data from both Study 15 and COSTAR. "The FDA has approved the medicine as a safe and effective treatment for schizophrenia and bipolar disorder," he says.

Write to Shirley S. Wang at shirley.wang@wsj.com and Avery Johnson at avery.johnson@WSJ.com