July 2nd, 2013
Hundreds of Questions About the Integrity of Stem-Cell Research Group
Serious questions have been raised about the integrity and validity of research conducted by a well-established German stem-cell research group. An article in the International Journal of Cardiology exhaustively details a multitude of discrepancies and contradictions in papers from the group. Further, the revelation of such widespread misconduct may lead to broader, disturbing questions about the reliability of scientific publications and the ability of the clinical research system to police itself.
In their article titled “Autologous Bone Marrow-Derived Stem Cell Therapy in Heart Disease: Discrepancies and Contradictions,” Darrel Francis and colleagues scrutinize 48 papers from the research group Bodo-Eckehard Strauer. According to Francis et al., the 48 papers from Strauer’s group contained reports on only 5 actual clinical studies, or “families” of reports, and duplicate or overlapping reports were common. The Francis article details more than 200 errors in the papers, including contradictory descriptions of the design, protocol, and results of the trials. Francis and colleagues write:
“Readers cannot always tell whether a study is randomised versus not, open-controlled or blinded placebo-controlled, or lacking a control group. There were conflicts in recruitment dates, criteria, sample sizes, million-fold differences in cell counts, sex reclassification, fractional numbers of patients and conflation of competitors’ studies with authors’ own.
Contradictory results were also common. These included arithmetical miscalculations, statistical errors, suppression of significant changes, exaggerated description of own findings, possible silent patient deletions, fractional numbers of coronary arteries, identical results with contradictory sample sizes, contradictory results with identical sample sizes, misrepresented survival graphs and a patient with a negative NYHA class.”
Strauer: Attracting Controversy
In 2009, Strauer retired from his position as the head of cardiology at Düsseldorf University Hospital. He has remained active in research at Rostock University. In December 2012, the German newspaper Süddeutsche Zeitung reported that fraud allegations against Strauer were being investigated by Düsseldorf University Hospital. The news seems not to have been reported outside Germany, and no further details have been revealed to date.
Strauer has been the subject of previous controversies. In 2010, he presented the results of the STAR trial at a Hot Line session during the annual meeting of the European Society of Cardiology. However, all the data had been published a month earlier in the the European Journal of Heart Failure. When the fact of the publication became public during the meeting, the ESC leadership rebuked Strauer and said that it would not accept his abstracts for two years.
In 2007, several prominent stem-cell researchers were sharply critical of Strauer for hyping the results of a single case to a German newspaper and calling it a “global innovation.” Andreas Zeiher, a stem-cell researcher who was also the head of cardiology at the University of Frankfurt, told the Scientist that “Science is not [done by] reporting a single case… What Strauer has done is to give a patient report, not a scientific study.”
When contacted by CardioExchange about the new allegations, Zeiher said he agreed with the major findings of Francis’s International Journal of Cardiology paper. “It is indeed more than scary and disappointing to learn that the extent of ‘imperfectness’ in reporting clinical data by the group of Strauer is so enormous. This will be another major blow against an emerging field (cell-based therapies), which did raise tremendous expectations, but appears to be crippled by irresponsible ‘clinical scientists,’ who appear to be focused exclusively on their own agenda.”
Other stem-cell researchers, cardiologists, and clinical trial researchers had similar thoughts about the glaring deficiencies of the Strauer papers cited by Francis et al., although several said they were unable to fully judge the nature and extent of the misconduct.
In an accompanying editorial, Peter Wilmshurst, a cardiologist who has been involved in several previous scientific controversies, writes:
“The analysis by Francis et al. suggests that there have been numerous publications of the same sets of data from Strauer’s group. Moreover what initially appear to be duplicate publications provide contradictory descriptions of experimental designs and of results. At the same time there are implausible agreements between data in populations of different sizes as well as statistical and arithmetic impossibilities in other datasets. I can see no logical option but to suspend belief in any of the data reported in these publications until an appropriate investigation has been performed.”
Some of the details cited by Francis et al. appear to be trivial, but some of the errors are startling. Francis et al. cite one example of a figure that appears unchanged in 2 separate publications. The two figures, reprinted below, showing the change in ejection fraction over time in the treatment and control groups, are from 2 separate publications about 2 different clinical trials with different numbers of enrolled patients. The first figure is reprinted from a chapter about the BEST-Heart study by Strauer and 2 colleagues in a book edited by Strauer and 2 colleagues (one of whom was also a coauthor of the chapter). The second figure is from the European Journal of Heart Failure publication of the STAR-heart study. Click to enlarge each figure.
Francis et al. also report that there were “many more patients being treated than bone marrow aspirations actually being processed.” In other words, the number of patients reported to have been treated with stem cells in the trials was greater than the number of stem cell preparations. In addition, they report:
Extraordinarily unlikely occurrences were reported. In one case two samples of very different sizes and different means had identical standard deviations.
Kaplan–Meier plots proved an unexpected quagmire. Some failed to start at time 0, whilst others failed to show the vertical steps at event times, and some did not have horizontal plateaus between events… Presumably through incorrect software use, curves sometimes looked almost manually-sketched … Authors did not always realise that Kaplan–Meier analysis and survival regression are contrary approaches.
The lead author of the new report is Darrel Francis, an academic research cardiologist at the National Heart and Lung Institute in London. His own research is far removed from the stem-cell arena. He became interested in the field when one of his patients travelled to Germany for stem-cell treatment from Strauer’s group. When the STAR trial was published, he “noticed that the numbers didn’t add up.” When he failed to receive an adequate answer to his questions from Strauer or from the journal, he began a more systematic examination of papers from the group. He was shocked by the findings.
The Francis paper in International Journal of Cardiology has itself been criticized by academics and clinical researchers who have seen it. For one, Francis overstates the prominence of Strauer and the impact of his work. Although the paper describes Strauer’s group as the “foremost group in this field,” other researchers in the field strongly disagree with this opinion. Strauer appears to have been an active researcher with many publications, but he is not considered a thought leader or a highly influential figure.
Another more serious critique of the Francis paper is that it fails to sort out minor or trivial errors from major flaws that raise the possibility of serious scientific misconduct or fraud. The overwhelming amount of detail contained in the paper makes it difficult to assess the situation without an enormous commitment of time and energy. But the experts who spoke to CardioExchange agreed that there were enough instances of serious flaws to warrant concern.
Beyond Strauer: Journals and the Scientific Community
A second major issue raised by Francis et al. is the unwillingness of journal editors to investigate articles they have published. In the online supplement to their article, Francis et al. reprint their correspondence with the editors of the Journal of the American College of Cardiology and the European Journal of Heart Failure (published by the European Society of Cardiology). The editors dismiss the issues raised by Francis et al., although they do not provide a substantive response to the issues. Zeieher was equally critical of journal editors for being “unwilling to take the necessary steps to ensure scientific validity of their own publications, even if confronted with obvious discrepancies in some of the papers published.”
Francis et al. also point out that although publications from Strauer’s group have been frequently cited, and also included in meta-analyses, the authors have failed to uncover many of the glaring errors in the papers.