Scientists Who Judged Pill Safety Received Fees
The manufacturer of the diabetes pill Rezulin provided fees or hefty research grants to at least a dozen scientists who also weighed the safety of the controversial drug for the federal government, records and interviews show.
Over the last two years, as many as seven of the researchers obtained up to $300,000 in grants. In one instance, funding that could total $50 million was awarded to the research firm of a UC San Diego scientist who had promoted the drug and helped lead the government study.
The payments exceed what was previously reported about the relationship between the researchers and the Warner-Lambert Co. Such ties are not uncommon in drug research, and in this case the manufacturer generated funding or compensation for at least 12 of the 22 scientists selected to operate a nationwide diabetes study by the National Institutes of Health.
The financial arrangements stand out in the case of Rezulin, which last year became one of the hottest selling drugs in America--only to come under a cloud after the liver failures and deaths of some patients.
The scientists at times played two different--and potentially incompatible--roles: As the government’s “principal investigators,” they judged the worthiness of Rezulin in a diabetes-prevention study conducted by the NIH. As researchers or consultants whose efforts were financed by Warner-Lambert, they were part of the drug maker’s drive to win quick acceptance of Rezulin as a front-line diabetes treatment.
Guidelines Warn Against Conflicts
Federal ethics guidelines warn against “any conflicting financial interest” that could bias the decisions of private or academic scientists who conduct research on behalf of NIH.
“If you’re ostensibly doing an objective study for NIH, and on the side you’re getting money from the drug company, that’s wrong,” said Dr. George J. Galasso, who helped draft the institutes’ current ethics guidelines before retiring. “They may be perfectly honest. But if they say anything good about the drug, are they saying it because the drug is good? Or are they saying it because they don’t want to cut off their supply of funds?”
A spokeswoman for NIH Director Harold E. Varmus said officials are uncertain “whether or not” ethics guidelines have been followed, noting that aspects of the diabetes study are now under federal investigation. The spokeswoman, R. Anne Thomas, said that, generally, “NIH does have concerns about objectivity, conflict of interest, perceived conflict of interest and public trust.”
In a statement, Warner-Lambert said it “has the utmost confidence in the integrity and honor” of the researchers.
“The suggestion that an academic researcher should not conduct clinical studies for the pharmaceutical industry and the NIH at the same time is naive,” the company said. “There is a very limited pool of diabetes experts who can conduct complex clinical trials.”
Warner-Lambert declined to disclose the amounts paid. The Times identified the dozen or more researchers through interviews, company-sponsored literature and Internet postings. The paper previously disclosed the potential conflicts of two leaders of the NIH study.
The blessing of a drug by NIH can have far-reaching influence--from investors seeking the next blockbuster to doctors choosing which drugs to prescribe. Overall, ethics sensitivities have heightened as the pharmaceutical industry, the government and academic institutions collaborate more closely.
In mid-1996, Warner-Lambert hailed the NIH researchers’ selection of Rezulin as one of two diabetes drugs in the $150-million nationwide Diabetes Prevention Program study. The company then won quick approval from the Food and Drug Administration in January 1997, and Rezulin has by now produced sales topping $1.6 billion.
But in June of this year, the FDA took the unusual step of severely narrowing the recommended use of the drug, citing its capacity to cause liver failure. The agency said Rezulin, intended to lower blood-sugar levels of those with adult-onset diabetes, should no longer be used as an initial, stand-alone treatment.
Responding to earlier reports in The Times, the inspector general for the Department of Health and Human Services began investigating the circumstances surrounding the NIH study of Rezulin. On May 17, 1998, a participant in the NIH study--a 55-year-old woman from East St. Louis, Ill.--suffered liver failure and died after taking Rezulin. Her death prompted senior NIH officials in Bethesda, Md., to banish Rezulin in June 1998 from the study. A second diabetes pill continues to be studied.
Despite liver-failure deaths that began appearing in general medical practice in the fall of 1997, the drug was not promptly withdrawn from the NIH study. The NIH researchers discussed the early liver failures with Warner-Lambert and considered withdrawing Rezulin, but did not. The decision to pull the drug was made by other NIH officials.
One NIH researcher, Dr. Mohammed F. Saad of UCLA, said Warner-Lambert agreed to provide him and his staff a $300,000 grant for one year with a renewal option. Saad termed the grant to research Rezulin “a big amount of funding,” compared with the average drug study.
Saad, who along with most of the researchers initially voted against using Rezulin in the NIH study, said Warner-Lambert convinced him the drug was safe by the time he agreed, in fall 1997, to accept the grant.
But, upon learning in May 1998 of the death of the woman who took Rezulin in the NIH study, Saad said, he “immediately” canceled his company contract.
“The drug is not safe,” Saad said. “If my mother were diabetic, I would not give her the medicine.”
The government study was financed in part by Warner-Lambert, which pledged to contribute $20.3 million in exchange for exclusive rights to market “any inventions arising” from the project, NIH documents show. The company withdrew this previously undisclosed commitment last year, soon after NIH officials dropped the drug from the study, according to Dr. Richard C. Eastman, chief of NIH’s diabetes division.
The potential conflicts of both Eastman and another NIH study leader, Dr. Jerrold M. Olefsky of UC San Diego, are the focus of the inspector general’s investigation. Eastman accepted more than $78,000 in compensation from Warner-Lambert from 1995-1997, while overseeing selection of Rezulin for the NIH study. Olefsky, who is listed on three patents as the sole inventor or a co-inventor of Rezulin, advocated the drug’s selection for the NIH study. He also was a founding co-chairman of the National Diabetes Education Initiative, a group established and financed by the company that has encouraged doctors to prescribe Rezulin.
Yet Olefsky has more ties with Warner-Lambert than previously reported: He is the co-founder and president of a privately held firm, Metabolex Inc., which accepted a diabetes research grant from Warner-Lambert that “could total in excess” of $50 million, according to an Internet posting in January by Olefsky’s firm.
The funding generated by Warner-Lambert for the NIH researchers has taken at least three forms--annual grants of up to $300,000 to conduct a separate, company study of Rezulin’s effect on newly diagnosed diabetics, speaking fees of typically $1,000 per address and other stipends.
Some researchers received compensation, including consulting and speaking fees, through Warner-Lambert’s Rezulin National Speakers Bureau or the National Diabetes Education Initiative.
Work Independent, Researchers Say
Several of the NIH researchers said their dealings with Warner-Lambert did not compromise their judgments on Rezulin.
“I certainly think that I’m capable of exercising independent scientific judgment,” said Dr. Christopher Saudek, a Johns Hopkins University specialist who has taken speaking fees from Warner-Lambert. Saudek also accepted a company grant to conduct the separate study of Rezulin’s effect on newly diagnosed diabetics.
Saudek said that he and other NIH study leaders have participated in the Warner-Lambert project “entirely out of scientific interest and good faith.”
Also among the researchers who have accepted grants or fees from Warner-Lambert or its affiliated diabetes group are Doctors Steven E. Kahn of Seattle; Edward S. Horton of Boston; Kenneth Polonsky of Chicago; Samuel Dogogo-Jack of St. Louis; Steven M. Haffner of San Antonio; Abbas E. Kitabchi of Memphis, Tenn.; Xavier Pi-Sunyer and Harry Shamoon of New York, and Ronald B. Goldberg of Miami.
Five of the researchers declined to comment. Kahn, echoing most of his colleagues, said that while he has taken speaking fees, there can be “absolutely no question of total impartiality” on his part.
Times researcher Janet Lundblad contributed to this story.
