In the spring of 2000, Dr. S. Charles Schulz attended a national medical conference to present favorable research on a new psychiatric drug called Seroquel.

Schulz, chief of psychiatry at the University of Minnesota, reported that the drug was "significantly superior" to the old gold-standard treatment for schizophrenia. In a press release by the manufacturer, AstraZeneca, he touted the "dramatic benefits" of Seroquel's class of drugs.

But newly released documents show that AstraZeneca knew the research didn't support the claim -- and knew two months before Schulz went public with it.

The disclosures have raised questions about Schulz's ties to the company as a paid consultant at a time when Congress and the university itself are intensifying their scrutiny of potential conflicts of interest in medical research.

"The data don't look good," an AstraZeneca official, John Tumas, warned in an e-mail on March 23, 2000. That month, an internal company analysis of the raw data concluded: "It is clear that a claim of superiority for Seroquel over Haloperidol (Haldol) could not be generated using these data."

In an interview this week, Schulz said the pharmaceutical company never shared its doubts about Seroquel, which went on to become a blockbuster, with annual sales of $4.5 billion today. "I don't recall anybody calling up and saying, oh my goodness, we have this problem," he said.

At the same time, Schulz acknowledged that his own study did not really show that Seroquel was more effective than the older drug. "That's a bit of a misunderstanding," he said. "I think the overall message is that it works about the same."

In a statement, AstraZeneca spokesman Tony Jewell said Schulz accurately presented the data at the American Psychiatric Association (APA) meeting in 2000, and "clearly explained the methodology he used."

Dr. Frank Cerra, the university's senior vice president for health sciences, noted that medical research can be interpreted in more than one way, and cautioned against jumping to conclusions based on Internet documents. He said he was not familiar with details of the Schulz study, but that there was nothing unusual about his relationship with AstraZeneca. "I think the role of university professors, particularly in health sciences, is to engage with the pharmaceutical industry and the device industry," he said. As long as Schulz disclosed his ties, he said, "then the university is OK with this."

Today, Seroquel is widely used worldwide for a range of psychiatric conditions, from schizophrenia to borderline personality disorders. In fact, next month AstraZeneca, a $32 billion British concern, will ask the Food and Drug Administration for approval to market it for major depression, which could greatly expand the drug's patient base and sales.

However, experts say doubts remain about the value of the drug. "It does work, but it's not necessarily better than the pills that are already out there," said Dr. Daniel Carlat, a psychiatrist at Tufts University in Boston. "And it does have some nasty side effects."

Ed Blizzard, a Houston lawyer who is handling consumer lawsuits against the company, said he believes AstraZeneca inflated the drug's value. "They marketed it as having unsurpassed efficacy," he said.

The internal company documents became public last month in connection with a federal court case in Florida over Seroquel's side effects. Since the documents started circulating on the Internet, at least one outside critic has called for an investigation by the university.

"Is it any wonder that psychiatric research, and by implication psychiatric practice, is losing credibility due to the economic influence of big pharmaceutical companies," wrote Brent Robbins, a psychologist at Point Park University in Pittsburgh, on a blog site of the Society for Humanistic Psychology.

Robbins said he had contacted the Medical School to look into the matter. "How can we trust science when the people who are conducting the science are only publishing or presenting on findings that favor the economic well-being of the company for which they are hired?" Robbins asked.

Schulz denied that his ties to the company affected his research. "I don't think I would have been comfortable standing in front of a poster that exaggerated the results," he said.

Over a 10-year period, Schulz received at least $88,000 from AstraZeneca for consulting and research on its drugs, according to records at the Minnesota Board of Pharmacy, and authored five studies of Seroquel. AstraZeneca would not reveal the amount it has paid Schulz in consulting and research fees, saying the information is proprietary.

By early 2000, Seroquel had been on the market for three years when Schulz prepared his study for the APA meeting-- an analysis of 1,800 patients in four separate trials. Schulz reported that patients on Seroquel were more likely to show marked improvement on a questionnaire about delusions, hallucinations and other symptoms of schizophrenia.

But company officials were taken aback when their own analysts took a closer look at the overall results and came to a different conclusion, the internal documents show.

"What seems to be the case is that we were highlighting the only good stuff," wrote Tumas, the AstraZeneca official, in his March 23 e-mail to colleagues. "Once you get a chance to digest this," he wrote, "let's ... discuss where to go from here. We need to do this quickly, because Schulz needs to get a draft ready for APA."

When the study was presented, AstraZeneca issued a news release headlined "An analysis suggests Seroquel (R) has greater efficacy than Haloperidol." It quoted Schulz saying: "I hope that our findings help physicians better understand the dramatic benefits of newer medications like Seroquel." Jewell, the company spokesman, said the release spoke "narrowly to the analysis that was to be presented at a medical meeting."

Schulz said this week he did not mean to assert that Seroquel was more effective, but that it had fewer serious side effects, such as muscle tremors, than Haldol, which meant patients were more likely to stay on the medication.

"Looking back, I could have said it differently. In the current climate, I certainly think about these things many times over before opening my mouth."

Asked whether the company should have let him go ahead with the presentation, Schulz said, "That's a good point."

In 2003, Schulz published a paper in a scientific journal saying that multiple studies showed that Seroquel was comparable in its effectiveness to Haldol, not better.

Maura Lerner • 612-673-7384 • mlerner@startribune.com Janet Moore • 612-673-7752 • janet.moore@startribune.com