The Paolo Macchiarini scandal draws wider circles, and unfortunately, it involves more unnecessary patient deaths though questionable tracheal transplants, this time at the University of Gothenburg in Sweden.
In 2011, scientists at the Sahlgrenska University Hospital of Gothenburg applied a method of bioreactor-engineered trachea (similar to that by Macchiarini) to operate on a 76-year old patient, possibly without a medical necessity (he had asthma since decades). The patient died 23 days later “due to cardiac arrest but with a patent, open, and stable tracheal transplant and intact anastomoses”. This is what the paper (Berg et al, Tissue Eng Part A. 2014) claims, which last author is Suchitra Sumitran-Holgersson, professor of transplantation biology at the Sahlgrenska Academy. The University of Gothenburg described Sumitran-Holgersson as “genius”, in an article presenting her rather unconventional method of purifying endothelial stem cells from patients’ blood. That work she did in collaboration with another Sahlgrenska professor and medical director of the Transplant Center, Michael Olausson.
Guest cartoon: Jill Howlin
Yet, it is far from certain that this elderly patient’s death was unrelated to the tracheal transplant, as authors claim. In fact, he died from an infection, according to Svenska Dagbladet. There is some debate of how life-threatening a condition the patient was in, but even Sumitran-Holgersson’s co-author Olausson stated in a comment on my blog:
“The patient transplanted acquired his problems 34 years previously after a trauma. The patient and the daughter wanted the procedure and the patient consented the procedure – he did not want live in the present conditions”.
First of all, patient’s and his family’s consent does not liberate Sumitran-Holgersson and Olausson from their responsibility to apply the best possible and the most reliable therapy, as opposed to testing novel experimental approaches despite the possibility of safer alternatives. Also, it is quite possible that after 34 years this patient was not a “compassionate case”, where no “conventional” therapies were applicable to save his life. This might have justified the testing of experimental therapies of a “bioreactor”-engineered trachea. In fact, Macchiarini is accused of having misrepresented the true medical state of his patients to justify compassionate case therapies. In one case he insisted his transplant recipients had cancer, while evidence of a stable and cancerless condition was seemingly suppressed. Thus, the Gothenburg case might even have been similar to Macchiarini’s operation on the patient Yulia T. in Krasnodar, whose clinical condition was anything but imminently life threatening and who, prior to giving her consent, apparently assumed to be undergoing a kind of cosmetically surgery, after a car accident left her with an painful hole in her throat (case summary in Russian here and SVT video in Swedish here). Also this decision to operate was apparently approved (via video-conference) by the Swedish doctors, from the Karolinska Institute (KI).
From Gothenburg Olausson declared that, despite patient’s death, the operation was a success, as a proof-of-principle:
“The patient died. The new information was the result from the autopsy which showed a very good recellulirized trachea (Berg et al). The paper was difficult to publish due to the Macchiarini group was reviewing the paper. It was considered unethical to publish this case since it might destroy the whole field of tissue engineering”.
Well, that depends on how reliable the presented data is. The image proving the recellularisation of trachea is found in two other, liver-related papers by Sumitran-Holgersson and Olausson, according to evidence on PubPeer. Even those who are not medical specialists like Olausson will know that trachea and liver are really different organs to confuse with each other:
Also, an apparent image duplication might be suggesting that the donor trachea was possibly not thoroughly decellurised by Sumitran-Holgersson and Olausson, before it was transplanted into the patient, who then died just 3 weeks later.
In fact, a growing number of publications by Sumitran-Holgersson as last and corresponding author are being flagged on PubPeer for image duplications and suspected data manipulation. Several of them feature Olausson as co-author.
As mentioned above, Sumitran-Holgersson and Olausson have already operated three children by implanting them with bioreactor-engineered veins. The University of Gothenburg described this as “stroke of genius”. The method has been patented by Sumitran-Holgersson and Olausson and commercialized by a company, NovaHep, which was founded by Sumitran-Holgersson husband, Jan Holgersson (who is also professor of Transplantation Immunology at the Sahlgrenska Academy). This is what readers noticed in the relevant papers by Sumitran-Holgersson, according to PubPeer:
Nowak at al, Circulation, 110 (2004)
“Expression of vascular endothelial growth factor receptor-2 or Tie-2 on peripheral blood cells defines functionally competent cell populations capable of reendothelialization”
Olausson et al, Lancet, 380 (2012)
“Transplantation of an allogeneic vein bioengineered with autologous stem cells: a proof-of-concept study”
As Swedish newspaper Dagens Medicin reported, the case is now being investigated, with the involvement of the original Macchiarini-detective Bengt Gerdin (whom The Lancet has attempted to discredit in 2015). Gerdin commented to me in an email:
“What is being done in Gothenburg is not an investigation due to suspicion of research misconduct, as was the case at KI, but an internal investigation based on a joint initiative between health care and university with the intention to assess in every detail what was done with respect to the transplanted trachea and the actual vein grafts, and secondly to analyze what to learn from that. In the end, the goal is to develop procedures for work in the boundary between clinical research and medical care. Professor Olle Lindvall, who is to lead the process to assess similar issues under the Royal Academy of Sciences, and professor Kjell Asplund, who is to lead the investigation on the parts of the Macchiarini affair that took place at the Karolinska Hospital, are loosely associated in order to validate the process”.
“It is incredibly important that we investigate all aspects of this. It is also important that the suspicions against one or more researchers are not the same as a conviction”.
According to Swedish regulations, after the ongoing internal investigation it will be up to Fredman to report the case to the Swedish Central Ethical Review Board (CEPN), which would then initiate an official external investigation. Then, science can start self-correcting itself.
Except that Swedish science already tried self-correcting with Sumitran-Holgersson and got a bloody nose, and the story of how this happened is much more bizarre that that of Macchiarini.
The latter was found guilty of misconduct by Gerdin’s institutional investigation at KI, while afterwards the (now sacked) Vice Chancellor Anders Hamsten chose to declare Macchiarini innocent and not to forward the case to CEPN for investigation. The Macchiarini status is of course different now, as I reported. Yet Sumitran-Holgersson was already extensively investigated for research misconduct between 2009 and 2011 by the highest Swedish investigative authority at that time (and now solely funding agency), the Swedish Research Council Vetenskapsrådet (VR). She was found guilty of research misconduct and data manipulation by the VR’s own investigative expert team, but nevertheless fully exonerated afterwards. The information below was provided to me by a reliable source (whose true identity I am aware of) and independently confirmed in comments and emails by other readers of my blog.
Sumitran-Holgersson used to work at Karolinska Institutet before moving in 2008 to University of Gothenburg. In 2009, three her former students reported her for misconduct to the then-KI Vice Chancellor Harriet Wallberg-Henriksson , who is also a member of the Nobel Assembly.
After an internal investigation, Wallberg forwarded the case to VR for investigation, which then appointed two scientists as investigators, Olle Kämpe, professor of molecular medicine, and Lena Claesson-Welsh, professor of vascular biology, both at Uppsala University.
As the investigation progressed, the two have found a number of data misrepresentations and falsifications. For example, they determined that Sumitran-Holgersson presented at a midterm PhD seminar a table from a manuscript, where the number of samples suddenly increased from n=1 to n=3. This happened overnight, simply by addition of standard deviations, while the mean value of every single one of the 17 measured flow cytometry (FACS) parameters stayed the same (original evidence available here).
Sumitran-Holgersson claimed that she did not manipulate this Table 1 prior the midterm PhD seminar, but that she instead herself performed the two additional experiments. In an email from 11.05.2009 she informed the Research Council (VR) that she has provided the raw data in question. On May 13th FACS histograms from two experiments labelled 080616 and 080628 arrived and were duly registered at VR.
However, the two FACS scatter-plots representing separate experiments were identical down to every single pixel. A strange date “090226” was stamped on the histogram analysis for the cell surface marker CD45, supposedly performed nine months earlier.
In a reply to the preliminary investigators’ report dated 18.11.2009, Sumitran-Holgersson added 102 different attachments. Among those were the attachments Nr. 31 and 33, which resembled the FACS histogram from May 2009, only that this time the date “090226” disappeared and the scatter plots now looked fuzzy.
Later on, on 18.05.2010, a former PhD-student of Sumitran-Holgersson’s discovered files named “VR1new”, “VR1” and “VR2”. The date stamps of these experiments did not match the dates when these files were first created, according to the file directory. Welsh and Kämpe were called to KI’s Department of Clinical Science, Intervention and Technology (CLINTEC) in Huddinge, where Sumitran-Holgersson used to work. There, the investigators asked the IT responsible to create a copy of the hard disk. The copy was left at Huddinge and the original hard disk brought to Uppsala to be kept in a safe in until further analysis. This analysis was later performed by IBAS, a Norwegian company specialized in IT forensics. IBAS were unable to detect any manipulations of the data on the hard disk. When the files were subsequently analysed, it emerged that the FACS files came from three completely different experiments for completely different stainings and conducted by an unrelated researcher on 18.11.2008, 24.03.2009 and 26.03.2009. When Sumitran-Holgersson originally submitted these files to VR, all the date stamps turned out to have been covered up by white rectangles, which also explained why some numbers were only partially visible in the histograms.
This information is evidenced in the report by Kämpe and Claesson-Welsh, which was submitted to VR and to the then-Dean of Research at KI, Martin Ingvar, on 07.11.2010.
The strange thing is, Ingvar cannot remember ever receiving such letter from Kämpe and Claesson-Welsh. What makes this even more strange, is this statement from the Registrar of the Karolinska Institutet, Lise-Lotte Wallin:
“We have searched our register and have not found any letter directed to Prof. Martin Ingvar the specific date from the two persons you were referring to.”
Well, whatever happened to this letter and its attachments at KI, here it is, at least in part.
Based on the investigative report, the Vice-Chancellor Wallberg-Henriksson decreed that Sumitran-Holgersson committed scientific fraud. The entire 27-pages-long final report from the VR (in Swedish), dated 9.09.2010, is available here. Sumitran-Holgersson was to be banned from VR funding for 10 years. A paper by Sumitran-Holgersson in (Elsheikh et al, Blood 106 (7), 2005) was retracted in May 2011, with the retraction note stating “that the results as published cannot be considered reliable”.
Yet some at the University of Gothenburg disagreed with this interpretation. Olausson still defends his collaborator Sumitran-Holgersson, calls the Kämpe-Welsh investigation as “below any standards” and insisted in his email to me that “the misconduct was performed inside KI after she left”. In fact, Olausson did his own analysis:
“I recruited Suchitra Holgersson from KI and wanted to be absolutely certain that she was not cheating. I succeeded in getting hold of backups from USB and computers and could see that the files had been edited. Using freeware I could see that files in a folder/catalogue had been renamed – but the time for their analyzes stated the right order. When the files are analyzed they are given a date and time stamp followed by an order .001, 002 and so on. I bought professional forensic computer software (Knoll) and put down more than 1000 hours of work”.
Yet, the direct evidence of deliberate data manipulation, namely data duplication and its re-use from an utterly unrelated experiment was provided by none other than Sumitran-Holgersson herself, upon request of the two investigators. Kämpe and Claesson-Welsh, never had any conflict of interest when investigating Sumitran-Holgersson, unlike her close collaborator Olausson.
Back in 2010, Olausson and his colleagues Elias Eriksson, Kristoffer Hellstrand, as well as Börje Haraldsson and Ola Stenqvist set to discredit the investigation. After they succeeded, they gave some of their professional opinions in University of Gothenborg’s own journal. Hellstrand stated, seconded by Haraldsson:
“Since I use similar devices on which the fraud was supposed to have been carried out, I became interested and began to examine the case. My conclusion was that there was no evidence that there ever was any cheating”.
A certain journalist named Sus Andersson also became involved. She got direct access at VR to all the documents and posted her investigation on the website FARAD!. Andersson and the Gothenburg professors insisted that the investigation did not suffice the Swedish guidelines for transparency, since some of the investigation’s documents have not yet been registered and made public right when they were first submitted to VR. This was their main argument, because Swedish law demands that every single official document must be publicized immediately. And we are talking about thousands of pages, with copious (yet, as I was told, hardly informative) statements by Sumitran-Holgersson and mutual accusations between opposing fractions at Karolinska and Gothenburg. Thus, the whole investigation became invalid regardless of its actual findings, just because not all of its documentation was stamped and published right away.
Also, in Sweden the whistleblower’s confidentiality is not protected by law, quite the opposite: the identity of the whistleblower must be released to the scientist accused of misconduct. Instead, the investigators tried to shield their informant, who was an employee of Sumitran-Holgersson, from repercussions. This was also held against the investigation as evidence for its allegedly insufficient transparency.
Some of the many Sumitran-Holgersson declarations, together with the whistleblowers’ names, were posted on a website by her loyal employee Pradeep Patil. Patil has a good reason to decry a “conspiracy in science” against his former and current boss, since several of his, also first-author, papers bear clear signs of image duplications, across unrelated papers. Here is an example:
As an interested Swedish researcher (whose identity is known to me), commented in an email:
“Suddenly whistleblowers had to prove that they are not guilty, that they had no personal reasons, that they did not get any benefit from the case and so on and so on.
Finally , when number of documents exceeded any reasonably readable amount, the case was done: it was indeed impossible to understand who did what and when and why”.
After the conclusion of the investigation the Vice-Chancellor Fredman initiated a procedure to sack Sumitran-Holgersson from the Gothenburg University. As Agnes Wold, professor at the Department of Infectious Medicine, University of Gotheburg, commented on my blog, this was met with the demands by Olausson, Stenqvist and Haraldsson that Fredman herself is to resign or be dismissed instead. They made their demands public, in the Swedish evening paper Expressen/GT.
The pressure from Gothenburg professors, plus Andersson’s reporting, made the Swedish authorities to turn against their own investigation. It was completely and utterly dismissed as nil and void in a decision from 24.10.2012 by VR’s investigative successor, CEPN, signed by Johan Munck, former chairman of the Supreme Court of Sweden. The file is available on FARAD!. The legally irrevocable and binding decision was obtained after only one and a half hour interview with Sumitran-Holgersson, seconded by Elias Eriksson, while neither the whistleblowers, nor the investigators were called to give evidence. In fact, CEPN did not even bother to obtain all the documents from VR, such as the binders containing the analysis of the FACS files in question or any other documentation. Ironically, CEPN acknowledged that Sumitran-Holgersson did indeed fabricate the evidence which she provided to the expert investigators, but this was deemed irrelevant since that fake research data was not intended for publication.
The VR Director-General Mille Millnert declared in Dagens Nyheter that the investigators “overstepped their mandate”. Hellstrand and Haraldsson made sure Sumitran-Holgersson got all the VR funding back, which was taken away from her, as Wold commented.
Now CEPN will be investigating and deciding about scientific misconduct and patient deaths connected to Macchiarini. In fact, the journal The Lancet announced to me in an email only to consider responding to harsh criticisms of Macchiarini’s publications, once CEPN has “reached a verdict”. Good news for Macchiarini, given the Sumitran-Holgersson precedent? Maybe not, since CEPN has a new leadership now, primarily Karin Almgren, the Justice of the Supreme Court,. Almgren used to be the chairwoman of the “Oredlighetsgrupp” at VR, which originally appointed Welsh and Kämpe and found Sumitran-Holgersson guilty of scientific misconduct.
Now, Sumitran-Holgersson actually swims in cash since the unpleasant incident. Due to her miraculous abilities to grow in her stem cell lab not only tracheas and veins, but also any other type of giblets like liver, intestine, kidneys, heart, lungs and pancreas, Sumitran-Holgersson has been literally showered with funding money by VR and other Swedish funding agencies:
- Ongoing grant of 1 Mio SEK from Hjärt-lungfonden (Heart-Lung foundation), apparently to grow an artificial heart from stem cells.
- Ongoing grant of 2.1 Mio SEK from VR, “Native extracellular matrix as scaffolds for tissue-engineering of pancreas with human stem cells”.
- Ongoing grant of 2.5 Mio SEK from the IngaBritt and Arne Lundberg Research Foundation, “to purchase a bioprinter that can recreate tissues such as blood vessels and liver tissue, as well as body parts such as ears and noses”.
- Ongoing grant from Kidney Research Fund Njurfonden (who removed this information from their website and then denied having ever funded Sumitran-Holgersson, but I saved a brochure with some info)
To round up the story, here is some blackmail:
In November 2009, VR received a letter from the company Adsorber, signed by the CEO Anders Karlsson (Swedish-language document available here). This company, as its website states, “was founded in 1999 based on inventions made by Associate Professor Jan Holgersson at the Department of Clinical Immunology and Professor Suchitra Holgersson at the Department of Transplantation Surgery at Sahlgrenska University Hospital, Gothenburg, Sweden”. Mr. Holgersson is still one of three members of Adsorber’s Board of Directors. This is what Karlsson wrote to Jan Stålhammar of the VR’s Legal Council:
“I write to the Research Council in my capacity as President of Absorber AB after having taken note of the inquiry professors Lena ClaessonWelsh and Olle Kämpe drawn up on behalf of the Research Council’s expert respect Professor Suchitra Holgersson’s research activities.[…]
It noted that Professor Holgersson coauthored in the journal Transplantation [Breimer et al, 87(4):549-56,2009; -LS] “the basis for registration with the Food and Drug Administration (FDA) for a new commercial test to regarding a study that is assess the risk of organ rejection in kidney transplant recipients.” The test is described is Absorber’s crossmatch test XM-ONE, for which the FDA announced on July 29, 2008 that the product is approved for marketing and sale in the United States. […]
In the event the FDA should act in error because of VR inquiry in terms of the XM-ONE’s approval, this could result in significant economic damage to Absorber, which made major investments in the product and the marketing and sale of same. If someone knowingly mislead the FDA in this question, Adsorber will be compelled to claim compensation against the person concerned. Such claims may amount to a very significant sum”.
Basically, Karlsson stated that Holgersson’s biotech company will sue everyone who dares to discredit his wife for all they have and much more, as “compensation” can easily mean millions of dollars. And he did not target his threats at the mighty Swedish authorities. In fact, as a VR lawyer told Kämpe and Welsh, they would be on their own against Absorber. No legal protection or assistance at all would be provided by VR. One can understand that Kämpe and ClaessonWelsh felt somewhat pressured in their investigations.
After the Macchiarini and Sumitran-Holgersson cases I really start to wonder about the feasibility of reporting and sanctioning of scientific misconduct in Sweden. A whistleblower must be career-suicidal, as his or her name will be officially served to the reported fraudster (who is often whistleblower’s own superior) on the silver platter. An investigator must be either a naïve idealist or a masochist, who does not mind seeing own name dragged through mud and threatened with lawsuits, all for the crime of actually doing one’s job thoroughly. Maybe this is why Sweden has such an admiringly low incidence of scientific misconduct.
Update 9.03.2016: I received today an email from Registrar with attached official statement by Vetenskapsradet (VR), where the Director General Mille Millnert declares (in Swedish):
“the reasons not endorse the expert group’s report are of two kinds:
• The lack of order in the handling of the investigation makes it impossible to follow afterwards• Expert Group goes beyond its mandate”
For the second argument Millnert explains that experts could not prove that Sumitran-Holgersson manipulated the data with the intention to deceive: “issues such as intent or purpose should not be treated by the expert group. These issues should instead be treated by the initiative universities or in court, which also gives the opportunity to appeal”. He attaches the work guidelines for expert group for investigation of suspected
- an explanation why […] the Vetenskapsradet apparently participated in the legal threatening of the investigators Olle Kämpe and Lena Welsh
- is Vetenskapsradet considering […] to withdraw at least the running funding to Prof. Sumitran-Holgersson?
Update 31.03.2016. I received from the VR’s Chefsjurist Anna Hörnlund a hard copy of the original Kämpe/Welsh report to VR (in Swedish), dated November 6th 2009. The scanned file is available here, I also attempted some OCR. Due to many technical OCR errors, which make automatic translation difficult, I offer below only my manually corrected Google translate version of the “Summary and conclusions of experts”:
“It is the responsibility of each researcher to demonstrate the original data to support des conclusions and claim it for the front .The documentation of the scientific work of Suchitra-Holgersson’s research group has been very weak and in several cases superficial and, to the extent it happened at all, often done with the use of loose-leafs and not in the prescribed bound laboratory diaries. These deviations from good research practice has prevented a meaningful analysis of the authenticity of some data, but the shortcomings are so serious that they must be considered as a form of scientific misconduct.
The paper in Blood (Elsheikh et al., 2005] contains fabricated data. This is scientific misconduct; and the article should be withdrawn. [which also was retracted, -LS]
The manuscript (Elsheikh et al. 2008) seems to have been improved with the addition of fabricated distribution measurements. This procedure involves scientific misconduct. To the manuscript originally compiled by Berglin a long list of improvements has been added, which cannot be substantiated with documentation, while it emerged that there never was a sufficient amount of serum available [to perform the described experiments, -LS]. This represents, in our opinion scientific misconduct.
Overall, we find serious deficiencies in the documentation, and lack of critical thinking and a sound scientific approach in Holgersson’s research, and the pattern has also continued in the ongoing investigation with the arrival of damning manipulations of research documentation. We also find it proven that in several cases have been clear fabrication of data. What we’ve found here is very serious deviations from good research. We therefore believe that there are strong reasons to investigate all research coming from Professor Suchitra Holgersson’s team over the years”.