PROBE BY DINGELL RAISES ISSUE: WHO POLICES SCIENTISTS?

Author: By Peter G. Gosselin, Globe Staff

Date: 04/30/1989 Page: 1
Section: NATIONAL/FOREIGN

In recent months, a team of Secret Service agents, wielding methods usually reserved for tracking writers of presidential death threats, went to work on the notebook of a Greater Boston woman.

Their finding: Entries on some of its pages were written after the dates on the pages seemed to indicate they were.

But the document was no death threat nor the woman a would-be assassin. Instead it was the laboratory log of Thereza Imanishi-Kari, a pathologist at Tufts University School of Medicine and research associate of David Baltimore, a Nobel laureate at the Massachusetts Institute of Technology and one of the nation's most eminent scientists.

Secret Service entry into the fray is the latest in what is fast becoming the most celebrated probe of alleged research misconduct in decades. It sets the stage for an explosive confrontation later this week between the scientists and Rep. John Dingell of Michigan, the powerful Democratic chairman of the House Energy and Commerce subcommittee on oversight and investigations.

Lawyers for the scientists described the newly disclosed Secret Service involvement yesterday and expressed open horror about it.

"This has all the classic trappings of a witch hunt," said Imanishi- Kari's lawyer, Bruce A Singal.

At issue is a 1986 paper published in the scientific journal Cell that focuses on how the immune system is altered when foreign genes are added to a test animal. Questions about portions of the paper were first raised by Margot O'Toole, a former postdoctoral fellow of Imanishi-Kari's. The dispute over the paper, now in its third year, has called into question the dramatic conclusion of the paper, which was co-authored by Imanishi-Kari, Baltimore and four others, including Moema Reis, now of the Instituto Biologico in Sao Paulo and David Weaver of Boston's Dana-Farber Cancer Institute.

The authors have already published one correction about portions of the paper in the journal Cell, and a second one is in the works. The case has raised the troubling issue of whether junior scientists can challenge their seniors without destroying their own reputations in the process. O'Toole has said her career came to a screeching halt after she questioned whether her boss' data supported the published paper's conclusions.

And it has opened up the contentious question of who can -- and should -- investigate allegations of scientific wrongdoing. If they agree on nothing else, virtually all American scientists seem to agree on who should not investigate, and that is John Dingell.

"How science should be reviewed is really what the Dingell hearing is all about," said Baltimore's lawyer, Norman D. Smith. "There are those who believe that bureaucrats should breathe down the necks of scientists in order to detect fraud and those who believe that science is a self-correcting process and ought to be left to the scientists."

What Smith and others in the case described of the Secret Service involvement is anything but leaving it to the scientists. The lawyers said that Dingell staff members briefed them and their clients about the agency's findings last week in preparation for subcommittee hearings this Thursday and the following Tuesday. It was unclear how the Secret Service became involved.

Secret Service agents used what is called "impression analysis" to examine pages from one of Imanishi-Kari's laboratory notebooks, Smith said. By bringing out previously invisible indentations left on lower pages from writings on the pages above them, he said, they contended they were able to date the lower pages. When they compared dates thus determined to those actually written on some pages of the notebook, they found what they contend are discrepancies, he said.

Subcommittee staff members refused to discuss the Secret Service analysis or any other aspect of the case yesterday.

One possibility for the discrepancies seems to be that the notebook entries were predated to make it appear that more evidence supported the conclusions of the scientific paper by Imanishi-Kari and others than existed at the time of publication.

Another is that whoever made the notebook entries in question innocently recopied old information onto newer pages, including the date when the information was originally developed without intending to claim that was when the information was actually entered in the notebook.

Smith and other spokesmen for the scientists strongly advocate this explanation, saying it involved none of the sinister intent suggested by the Secret Service findings.

"We believe from what we've been told that the Secret Service investigation is meaningless," said Alfred G. Kildow, a spokesman for Baltimore.

Meaningless or not, the Secret Service findings are likely to play a central role in Thursday's hearing before Dingell, and the hearing's outcome is likely to affect profoundly how misconduct allegations are handled in the future.

Dingell has repeatedly said he does not think scientists, universities and the National Institutes of Health, which fund most of the nation's biomedical research, can police themselves when it comes to allegations of research fraud, misconduct and error.

Some of the problems with self-policing seemed evident in the controversy surrounding the Baltimore group's paper even before last week. And disclosure at week's end of a 1986 letter from Baltimore to a colleague about the case did little to allay concern.

Since 1986, scientific panels at both Tufts and MIT have investigated the central allegation in the case -- that laboratory data did not support all of the conclusions of the research goup's paper -- and the panels decided no serious problem existed. For example, Herman Eisen, who headed the MIT probe, and his colleagues concluded that while one of the allegations about the paper was "disturbing," the error involved was not flagrant and did not merit a published correction.

Before last week, an NIH panel also had taken up the matter. Only two months ago, it absolved the scientists of the most serious accusations against them -- fraud, misconduct, manipulation of data and serious conceptual errors -- but found that they had committed "significant errors of misstatement and omission."

Concern about the paper's accuracy and the quality of the scientific review of it seemed to erupt again with full force last week as the NIH announced it was reopening its investigation of the publication.

James B. Wyngaarden, NIH's outgoing director, told associates he was reviving the probe because of additional questions raised by O'Toole. But sources speculated that the NIH official acted only after learning of the Dingell subcommittee's new evidence, including both the Secret Service findings and a confidential 1986 letter from Baltimore to Eisen.

The letter, written less than six months after the April 1986 publication of the disputed paper, suggests that Baltimore himself had serious doubts about portions of the paper but nevertheless argued that no need for a correction existed.

In telephone interview Friday night, Baltimore denied he ever had serious doubts and asserted that whatever concerns he may have discussed in the letter were based on a misunderstanding. Kildow and Smith said yesterday that the misunderstanding occurred because Imanishi-Kari, whose native tongue is Portuguese, is difficult to understand.

In the letter, Baltimore acknowledged that a test for an antibody, or diease-fighting protein discussed in the paper "doesn't do as described" and said that Imanishi-Kari told Eisen that she "knew it at the time."

He called her statement to Eisen "a remarkable admission of guilt."

Despite that, Baltimore argued that a correction should not be required
because it would damage the professional reputation of David Weaver, one of the paper's authors.

"All authors do have to take responsibility for a manuscript so all of us are in a sense culpable," the letter said, "but I would hate to see David's integrity questioned for something he accepted in good faith."

Last modified: Wed Sep 20 20:30:12 EDT 2000