Lax oversight by Pitt enabled Schatten fiasco
The University of Pittsburgh's lax policies and disregard for federal guidelines allowed biologist Gerald Schatten to participate in one of the biggest scientific frauds in history, according to a Pittsburgh Tribune-Review examination.
At issue is a Pitt-sanctioned collaboration between Schatten and Korean scientists headed by Hwang Woo-Suk in the controversial field of human embryonic stem cell research.
In a month-long review of university and federal records and through interviews of dozens of experts in medical research and bioethics, the newspaper found:
These policies and decisions enabled Schatten to collaborate with Hwang on now-discredited stem cell research that became embroiled in an international scandal.
"A stem cell trial like this without full review should just never, ever happen," said Glenn McGee, editor-in-chief of the American Journal of Bioethics and founder of the nation's largest research program on the ethical implications of stem cell science. "The University of Pittsburgh should take the lead in the nation to ensure that this option is never provided, since they are the ones that did it wrong."
Schatten heads the Pittsburgh Development Center at Magee-Womens Research Institute in Oakland and serves as a professor and vice chair of obstetrics, gynecology and reproductive sciences at Pitt's School of Medicine. He could not be reached for comment for this story.
In a paper published in June 2005 in the journal Science, Schatten and Hwang claimed to have custom-designed stem cells for individual patients from cloned human embryos for the first time. The discovery was hailed as an important step toward developing cures for diseases such as Alzheimer's and diabetes.
Six months later, a Korean investigation revealed that Hwang and possibly others deliberately fabricated data for the paper, which Science later retracted.
Korean criminal prosecutors are investigating Hwang, who was fired from Seoul National University, for potential misuse of millions in taxpayer dollars. Schatten faces no criminal charges and a Pitt investigative panel absolved him of intentional fraud. Any disciplinary or corrective action taken against him by the university will remain secret.
Medical school dean Dr. Arthur Levine referred questions to Pitt spokeswoman Jane Duffield. The referring of questions to Duffield has been a common refrain. All attempts over the last month to question Pitt officials or scientists for this story were met with either no response or questions being referred to Duffield, who is neither a scientist nor a physician.
"I do not believe that I have any information that you do not already have," Levine wrote in an e-mail.
Duffield did not respond to phone calls. She only responded by e-mail.
Duffield wrote that the university's Institutional Review Board (IRB) decided it did not have to approve Schatten's work because Schatten had told the board his research did not involve identifiable people.
In the world of science, research involving unidentifiable people is not considered to be human-subject research.
Protecting subjects
Shielding the identities of human research subjects frees scientists to complete their work without extra layers of oversight from IRBs.
These internal watchdogs are mostly composed of physicians, scientists and community advocates. They are charged with approving and monitoring biomedical research involving people.
Their ultimate responsibility is to protect the rights and privacy of study participants and to make sure people aren't harmed when they donate organs and tissues for research.
The rationale for these ethical principles has a long history, growing out of violations such as the Nazi human experimentation and the U.S. Public Health Service's Tuskegee, Ala., experiment from 1932 to 1972 on untreated syphilis in illiterate black sharecroppers.
"Research in the past has wound up abusing, harming or hurting people," said Arthur Caplan, director of the Center for Bioethics at the University of Pennsylvania. "You want to stick to these regulations as carefully as possible, because they really make sure the welfare and interests of human subjects are protected."
Scientists should consult with the IRB to decide whether their work involves human subjects, said Dr. John Falletta, senior IRB chairman at Duke University Health System in Durham, N.C.
"The investigator is in a potentially conflicted situation in making that determination for himself," Falletta said. "It's safer for everybody if the determination is made by a person or group of people who are trained in interpreting the regulations."
Recommendations of the federal Office for Human Research Protections also advise that investigators "not be given the authority to make an independent determination that research ... does not involve human subjects."
That is not the case at Pitt, according to Duffield.
"Most investigators conducting research make that determination themselves," Duffield wrote.
Scientists may request the help of the IRB when they are not sure whether the board has jurisdiction, but there is no mandate to do so, according to Duffield.
"This may suggest to one unfamiliar with the routines of clinical research that an investigator can simply go forward with their human-subject research without IRB approval," Duffield wrote. "However, this does not happen in practice because of the culture of the community and the built-in checks and balances."
The research
Schatten began working with Hwang on the stem cell project in early January 2005, according to Pitt documents.
In many ways, Schatten was the brains behind the research. He reviewed figures and tables, analyzed and interpreted data generated in Korea, drafted versions of the manuscript and communicated with journal editors. As senior author of the Science paper, he was responsible for its integrity.
Scientists are supposed to seek formal written clearance from an IRB before beginning their research when there is any question about whether a project might involve human subjects, Caplan said.
"Basically, you shouldn't be sending out anything until you have that letter in hand," Caplan said.
Duffield said Schatten first approached Pitt's IRB about his collaboration with the Korean scientists in February 2005, at least a month after the project had begun.
He informed the board that his collaboration with Hwang did not involve identifiable people, she said. This declaration prompted the board to determine it had no jurisdiction over the work, she said.
"Based on the information provided to the IRB, this project includes no involvement of human subjects, according to federal regulations," states a letter sent to Schatten on March 16, 2005, signed by Christopher M. Ryan, a Pitt neuropsychologist who is vice chair for the board's exempt and expedited reviews.
The day before, Schatten and Hwang submitted a first draft of their paper to Science.
Ethical obligations
Ryan, who did not return phone calls or e-mails, wrote to Schatten that the review board made its decision because "the investigator conducting research ... will not obtain identifiable private information."
The board used as a reference federal guidelines issued in March 2002 that say the study of human cell lines, where the donors cannot be identified by the investigator, is not human-subject research, Ryan's letter said.
But an online supplement that Schatten and his Korean co-authors published with their Science 2005 paper indicates otherwise.
The supplement says that when patients donated their cells for the research, their identities were "encoded by the responsible clinician and donor identity was unknown to the investigators and others."
Under Korean law, as explained in that supplement, the families of these donors must receive priority for any future treatments that result from the research to which they contributed.
For this to happen, however, at least one of the researchers must be able to determine the identities of the donors from the records encoded by the clinicians who collected the eggs and skin cells, said Mildred Cho of the Stanford Center for Biomedical Ethics in Palo Alto, Calif.
This breach of anonymity would have been good reason for Pitt's IRB to subject Schatten's work to full review, rather than give him clearance to move forward without its further scrutiny, Cho said.
"I think a lot of IRBs would not have made the same decision as this IRB did," Cho said.
Duffield said Pitt's IRB did not review the online supplement.
For the IRB to allow a project to proceed without its full review, there should be no doubt about whether the investigator could identify patients, Duke University's Falletta said.
"If there are questions, the IRB has the responsibility to ask more questions," he said. "If the (human) subjects are not anonymous, then it's human subject research (that is) subject to IRB review."
Subsequent investigations at Pitt and Seoul National University revealed the identities of the egg donors were not kept secret from the Korean scientists.
Hwang himself knew at least one of the women -- a staff member in his lab who later worked in Schatten's Pitt lab -- because Hwang drove her to the hospital for the egg extraction procedure, according to Pitt's own investigation.
Allegations about this ethical violation prompted Schatten to sever his collaboration with Hwang in November.
To protect the identities of the tissue donors, Schatten was required by federal regulation to have a written agreement showing he had no way to identify these people.
Specifically, as of Jan. 26, 1999, the U.S. Office of Human Research Protections has mandated that research involving coded data be subject to full review "unless a written agreement unequivocally prohibits release of identifying codes to the consultant."
"If there is a linking code, there has to be some kind of written confirmation or affidavit that the researcher will not be given access to the information," Cho said.
Pitt did not respond to requests to produce such a document.
"We have no evidence that (Schatten) ever had the codes, or any information that would permit him to identify subjects," Duffield wrote in an e-mail.
Some experts say Pitt was ethically obliged to fully review Schatten's participation in human embryonic stem cell research, regardless of whether the patients could have been identified.
Study of these cells is contentious, because obtaining them requires the destruction of 5-day-old human embryos from in vitro fertility treatments or cloned specifically for research. In the latter technique, used in Schatten's research, egg extraction procedures put women at health risk for research that might never benefit them personally.
Egg extraction involves using powerful hormones to manipulate a woman's body into producing many eggs. Some of these women experience abdominal discomfort, ovarian enlargement, nausea and vomiting. The women are at risk of fertility loss, kidney or multiple organ failure and death.
Members of Pitt's IRB "knew they were reviewing a study that would be controversial, and knew they were reviewing research where the (egg) procurement process would put people at risk," Cho said. "That, in and of itself, could've been an argument to put the study through full IRB review."
An open territory
Duffield said federal rules requiring IRB approval also would not have applied to Schatten's research because it was paid for by Magee-Womens Health Foundation, not federal money.
Indeed, federal documents filed by Pitt show the university opted not to apply federal rules governing protection of human subjects to privately funded research.
"It's an open territory," said Office of Human Research Protections spokeswoman Patricia El-Hinnawy. "Most researchers continue to do the right thing anyway, going through the IRB process, but there is no mandate for them to do so."
Duffield said Pitt's decision gives the university the flexibility to use a single, central IRB for industry-sponsored research in which more than one medical center is involved.
But El-Hinnawy said federal rules would not prevent a university from using a centralized IRB.
Allegheny General Hospital in the North Side requires any
research conducted there to be approved by its IRB, regardless of who pays for the work, spokesman Dan Laurent said.
Some experts say the decision to shield privately funded research from federal oversight is not necessarily uncommon. But such decisions could indicate the university doesn't want to be targeted by government audits.
Serious violations can lead the federal government to cut off a university's human research grants, El-Hinnawy said.
For example, federal regulators temporarily shut down Johns Hopkins University's human medical program when systemic problems were revealed after the 2001 death of a healthy woman in an asthma study.
"I think there are always issues of liability that make people hesitate," El-Hinnawy said.
Duke's Falletta said some institutions opt not to give the federal government oversight of privately funded research so scientists can do work that is federally restricted. This would include some types of human embryonic stem cell research.
Those scientists most likely go through the IRB process, said Dr. Richard Nelson, former chair of the Institutional Review Board at Children's Hospital of Philadelphia, who hosts an online IRB discussion and news forum.
"I personally don't know of any institution that does not apply these sort of regulations to all research that they conduct," Nelson said. "The standard is applied uniformly."
Pitt applies the same general ethical standards to all research, Duffield wrote.
But it is possible that privately funded research would not be accountable to anyone, Falletta said.
"Society has given investigators a mandate to advance knowledge, and that's not a free ride," Falletta said. "If investigators start behaving in a manner that doesn't meet societal standards, then those investigators are going to lose the opportunity to do that research."
More safeguards
Bioethicist McGee said Pitt must create mechanisms for oversight to prevent situations like the Schatten incident.
"The only thing that really could've prevented this would've been a policy that says, if you are working on stem cell research, you get full, full review no matter what," McGee said.
Full review might have slowed the research, possibly prompting the IRB to notice irregularities that pointed to the fraud perpetrated in Korea, McGee said.
For example, Hwang told Schatten in mid-January 2005 that some stem cell lines had become contaminated. But Schatten failed to conclude that their replacements could not have been grown and analyzed in time for the March 15 paper submission date as Hwang claimed, a Pitt investigation showed.
Earlier this month, Pitt issued draft regulations to help ensure that stem cell research is conducted legally and ethically.
Universities nationwide are taking similar steps, based on the recommendations of a report issued by the National Academy of Sciences in April 2005.
"There was a lot of concern that work was going to go forward without standardized guidelines," said University of Virginia bioethicist Jonathan Moreno, who served as co-chair of the panel that wrote the National Academy report.
"But the regulatory system has been designed for people who want to do the right thing but aren't sure how to proceed," Moreno said.
Faulty research
A chronology of events surrounding University of Pittsburgh scientist Gerald Schatten's role in fraudulent Korean human embryonic stem cell research.
March 15, 2005: Schatten and Korean researcher Dr. Hwang Woo-Suk submit a paper to the journal Science, claiming creation of patient-matched stem cells by cloning of human embryos.
March 16, 2005: Pitt's Institutional Review Board overseeing human subjects research sends Schatten a letter saying his project does not involve human subjects, which means it will not review his work further.
June 17, 2005: Paper is published in Science.
Nov. 11, 2005: The authors of the paper submit corrections to Science, saying they created fewer cloned cell lines than they originally reported.
Nov. 14, 2005: Schatten publicly severs ties with Hwang over ethical concerns about human egg procurement.
Dec. 4, 2005: Hwang contacts Science to report some cell images submitted as part of the paper unintentionally showed the same thing.
Dec. 6, 2005: Pitt's research integrity officer, Jerome L. Rosenberg, learns of a Korean news report questioning the validity of the research.
Dec. 14, 2005: Pitt notifies Schatten that it has opened an investigation two days after he asks Science to remove his name from the paper.
Dec. 15, 2005: Seoul National University (SNU) opens an investigation.
Dec. 23 and 30, 2005; Jan. 10, 2006: SNU announces in three stages that the supposed cloned cell lines did not exist.
Jan. 12, 2006: Science editors retract paper.
Feb. 8, 2006: Pitt panel appointed by Dr. Arthur Levine finds Schatten did not commit research misconduct, but said he shirked his responsibilities as co-author of the paper. It recommends the university take corrective or disciplinary action.
Source: University of Pittsburgh investigative report on allegations of possible scientific misconduct; Science magazine.