Author: By Peter G. Gosselin, Globe Staff

Date: 03/21/1991 Page: 1

WASHINGTON -- Capping one of the most bitter controversies in modern science, investigators for the National Institutes of Health have concluded that a former MIT immunologist fabricated crucial data in a 1986 scientific paper that she co-wrote with Nobel Prize laureate David Baltimore and others.

Baltimore asked yesterday that the paper be retracted.

NIH investigators accuse Dr. Thereza Imanishi-Kari, currently an immunologist at Tufts University, of "serious scientific misconduct" and of ''repeatedly" presenting "false and misleading information" to them, according to a draft report of their findings obtained by the Globe.

Baltimore was not accused of fabrication, but investigators labeled his five-year defense of the paper and Imanishi-Kari in the face of mounting criticism as "difficult to comprehend" and "deeply troubling." Baltimore is president of Rockefeller University in New York. At the time that the paper, dealing with the genetics of the immune system, was written, he was director of the Whitehead Institute of Biomedical Research at the Massachusetts Institute of Technology.

The draft report says that his behavior is "startling when one considers that Dr. Baltimore, by virtue of his seniority and standing, might have been instrumental in effecting a resolution of the concerns about the paper early on, possibly before Dr. Imanishi-Kari fabricated some of the data later found to be fraudulent."

In a telephone interview, Baltimore rejected the investigators' criticism of him but conceded for the first time that their findings "raise very serious questions about the veracity" of the paper and called for it to be withdrawn until they can be answered. He said it is up to Imanishi-Kari to answer the specific charges of fraud.

The investigators singled out for high praise Dr. Margot O'Toole, a former postdoctoral fellow in Imanishi-Kari's laboratory at MIT who lost her job after criticizing the paper.

"Dr. O'Toole's actions were heroic in many respects," the report says. ''She deserves the approbation and gratitude of the scientific community for her courage and her dedication to the belief that truth in science matters."

O'Toole said in a telephone interview that she felt "relieved and vindicated" by the report and that Baltimore owes her an apology. "A simple apology won't quite do it, but at least it's a beginning," she said. O'Toole is a scientist with Genetics Institute, a Cambridge biotechnology company.

Imanishi-Kari referred all questions to her lawyer, Bruce A. Singal of Boston. Singal refused comment on the substance of the report, saying it had been improperly leaked to reporters, but he denied any wrongdoing by his client. "There is no evidence whatsoever of falsification or fabrication," he said.

Baltimore, Imanishi-Kari and David Weaver, a scientist with Boston's Dana- Farber Cancer Institute and another of the paper's authors, have 30 days to contest the draft report before it is put in final form.

If the finding of misconduct is sustained, the burden of responding will fall on Imanishi-Kari, whose data are at the center of the dispute. At the extreme, she could be barred from receiving federal research grants for up to 10 years and the case could be referred for criminal prosecution, according to Suzanne W. Hadley, deputy director of NIH's Office of Scientific Integrity, which conducted the probe.

The dispute over the 1986 paper has become a notable one in American science, at one time or another drawing many of the nation's most prominent researchers and most of their sharpest critics into its orbit.

What began as a private disagreement between a scientist and her research fellow over results of experiments into the genetic regulation of the immune system of mice mushroomed into a battle royal over scientific fraud, Washington's right to oversee expenditure of public research dollars and the future of intellectual freedom in the nation. Along the way, two universities investigated the controversy, NIH conducted two probes and a congressional subcommittee held three hearings.

The original challenge to the scientific paper, which appeared in April 1986 in the prestigious journal Cell, came from O'Toole, who asserted that data produced by Imanishi-Kari did not support the paper's central conclusion that introduction of foreign genes into an experimental mouse altered the way in which the mouse's immune system developed.

Initially, O'Toole made little progress; review panels at Tufts and MIT concluded that there were no serious problems with the paper. But her complaints eventually reached Walter Stewart and Ned Feder, a pair of NIH researchers who have devoted themselves to exposing scientific fraud, and US Rep. John Dingell of Michigan, chairman of the House oversight and investigations subcommittee. They took up her cause, saying it revealed flaws in the nation's research system and in universities' ability to monitor misconduct.

Dingell's involvement in particular horrified many prominent scientists, who were mobilized to object by Baltimore. In a May 1988 letter to colleagues across the country, Baltimore labeled the congressman's probe "totally unnecessary" and warned that "what we are undergoing is a harbinger of threats to scientific communication and scientific freedom."

But while researchers protested, problems with the papers and allegations against Imanishi-Kari proliferated.

Based in part on work by Stewart and Feder and separate forensic analysis by the Secret Service, NIH investigators now charge that:

- Some data produced by Imanishi-Kari for the original paper and other data used in a subsequent published correction were "fabricated."

- Substantial parts of a notebook of experimental results that Imanishi- Kari used to answer critics were "falsified."

In some cases, the investigators said, computer tapes that Imanishi-Kari said contained supporting data from experiments on mice were produced years before the Tufts scientist had the mice on which she was working.

In reacting to the investigators' report, Baltimore singled out this
finding. "If they can demonstrate unambiguously that those tapes predate the existence of the mice, then it must be concluded that there was an intention to mislead," he said.

Last modified: Wed Sep 20 19:52:01 EDT 2000