U of M acknowledges mistakes in stem cell researchby Art Hughes, Minnesota Public Radio
University of Minnesota officials say discrepancies found in promising stem cell research appear to be mistakes, and not deliberate fabrications. A journal determined U of M stem cell researcher Catherine Verfaillie doubled up some data.
St. Paul, Minn. — The research offered some hope that adult stem cells might be as effective as the more controversial embryonic stem cells in treating certain medical conditions.
Experts say the subject needs additional research to determine whether the flaws significantly weaken the findings.
Verfaillie's 2002 paper in the journal Nature revealed cells taken from the bone marrow of rodents had the ability to develop into many other tissues in the body. This was previously thought to be restricted only to embryonic cells, which, unlike the adult cells, carried with them the controversy associated with destroying human embryos during research.
But now questions have surfaced about the research, and Verfaillie has acknowledged parts of the study are flawed. The magazine New Scientist claims some graphs were duplicated and some other information was misinterpreted.
Tim Mulcahy, the University of Minnesota's vice president for research, says the university has concluded its review of the concerns in the research.
"we know that there is some data included in that paper that was flawed technically," said Mulcahy. "We and Catherine have communicated that error or the flawed data to the journals involved, so the scientific community is aware of it."
Mulcahy says the university sought guidance from a panel of three experts. One of the experts found the flaws weakened the study. Another said the conclusions are uneffected by the flaws. The third reviewer didn't respond.
Mulcahy says Verfaillie stands by the main conclusions in the research, and that the final say will come through further study.
"Because of the fact there's always this diversity in opinion and assessment, it was our opinion that the most important thing to do -- and Dr. Verfaillie concurred -- was to make the facts known and let the community weigh the relevance, and how that would impact their own conclusions regarding the work," Mulcahy said.
The journal Experimental Hematology published one of Verfaillie's papers now being questioned. Editor in Chief Esmail Zanjani says the journal will print the appropriate corrections.
Zanjani says he believes the flaws do not compromise the conclusion that the adult cells can produce the variety of cell types Verfaillie's group originally described, and similar studies since concur.
"More recently in fact, not only to they do it in culture, but in vivo as well," said Zanjani. "In a recent paper they show they produce blood cells as well. So I'm fairly convinced these cells can do what's been proposed by Dr. Verfaillie's group for them to do."
Dr. Dale Hammerschmitt edited the Journal of Laboratory of Clinical Medicine for 15 years. He's a physician in the Hemotology, Oncology and Transplantation division in the University of Minnesota's department of medicine.
While Hammerschmitt couldn't comment on Verfaillie's research, he says, generally, flawed research falls into two categories -- mistakes or deliberate fabrication. He says, for instance, labs must always guard against duplicating information from common sources.
"It's not that hard for something accidentally to get published more than once, without the second publisher realizing he had to acknowledge the first publication. It really can happen by accident in a large laboratory," said Hammerschmitt. "And how egregious the event is would determine whether you think it was really a failure of oversight, or just an 'oops.'"
Hammerschmitt sasy the best response to such a mistake is to own up to it, correct the published record and let additional research confirm or disprove the findings.
- All Things Considered, 02/23/2007, 5:20 p.m.