Stem cell retraction leaves grad student in limbo, reveals tangled web of industry-academic ties

stem cells developmentA contested retraction in Stem Cells and Development has left the Queensland University of Technology (QUT) graduate student who fought for it in limbo, uncertain if he will earn his PhD. And many of those who didn’t want the paper retracted have a significant financial interest in a company whose work was promoted by the research — despite any lack of disclosure in the now-retracted paper.

Here’s the notice:

Concerns were raised by one of the co-authors, Luke Cormack, and after an investigation was conducted by the authors’ institution, The Queensland University of Technology, Stem Cells and Development is officially retracting the paper, A Chimeric Vitronectin: IGF-1 Protein Supports Feeder-Cell-Free and Serum-Free Culture of Human Embryonic Stem Cells, by Manton et al., from volume 19, issue 9 (pages 1298-1305).

The concerns center primarily on whether the images provided in Figure 1 show colonies of the stem cell line described at the passage stated stained for the marker listed. There were also related concerns with the PCR data included.

It is important to note that the corresponding author, KJ Manton, has denied deliberate wrongdoing. Stem Cells and Development acknowledges and appreciates the thoroughness of the investigation undertaken by Queensland University of Technology.

Stem Cells and Development believes that had the peer reviewers of the paper been aware of the extent and nature of the mistakes in the paper that apparently went undetected by the authors through an initial submission, two revisions, as well as galley proof, they would not have deemed the paper acceptable.

Stem Cells and Development is dedicated to the highest ethical standards of scientific publishing.

Although Cormack, a QUT grad student, was not the lead author on the paper, published in February 2010, it “formed the entire basis of my PhD,” he tells Retraction Watch. He submitted that PhD in December of last year, only to have it rejected in March. He didn’t have enough data, nor enough cells to do the necessary assays, and the committee felt the literature review wasn’t sufficient.

But there were other findings worth publishing, and QUT gave Cormack a scholarship to write manuscripts. Cormack went back to the original data from the now-retracted paper. The media had been developed in June 2007, before he joined the lab.

It was during the writeup of some manuscripts that I found that the paper was so full of errors.

That prompted an internal investigation, and Cormack’s co-authors wanted an erratum, but Cormack didn’t think that was enough. Neither did Graham Parker, who edits the journal. Parker tells Retraction Watch:

There were problems with the paper that I believe had the reviewers known they would not have accepted as it stood. Those problems must have been known to at least one of the authors. After revisions and galley proof approvals no-one except Cormack alerted the journal to those problems. Following an investigation by the host institution those problems were acknowledged yet no explanation was offered as to why the publishers had not been previously informed.

The paper has been cited six times, according to Thomson Scientific’s Web of Knowledge.

What also caught our notice is that senior author Zee Upton is consulting chief scientific officer of Tissue Therapies, which QUT spun off to develop technologies based on vitronectin and other compounds. She and other investors co-founded the company in 2003, approached QUT for a license on the intellectual property, and had Tissue Therapies listed on the Australian Stock Exchange in 2004, according to an Upton presentation available on the QUT website. An eprint of another paper by Upton and co-authors available on the QUT site notes that several authors had bought stock in the company:

The Authors have purchased shares in Tissue Therapies Ltd., an enterprise spun-out from the Queensland University of Technology, Brisbane, to commercialize some of the technology described in this manuscript.

That all seems to us like a conflict of interest for both the authors and for the university investigating the research. None of that was disclosed in the now-retracted paper, however.

Upton, Manton, and the head of the QUT press office are on holiday, according to their out-of-office replies. We’ll update with anything we learn.

So what next for Cormack?

It’s been a horrendous year. I haven’t been in the lab since March.

Cormack tells Retraction Watch he’s written a letter to the dean of research and training, asking if he can revise his thesis to include more background and describe the problems in the paper and why the retraction is a significant contribution to the literature.

I’d talk about my outcomes, what I have, and if I had been privy to the truth, what outcomes I should have had.

He remains optimistic.

I think things are a lot clearer now, now that it’s retracted. I’m moving forward.

53 thoughts on “Stem cell retraction leaves grad student in limbo, reveals tangled web of industry-academic ties”

  1. But he didn’t write a complete and acceptable thesis based on original, novel research… No real surprise that he’s been told he didn’t do enough work for a doctorate.

    1. It appears that the graduate student involved should be viewed as a whistleblower (by his own account anyway). Shouldn’t there be some acknowledge meant of the time and effort involved in cleaning up someone else’s mess?
      Perhaps if he his not deemed to have achieved work suitable for a PhD then the QUT owes him a considerable financial compensation for providing a highly deficient scientific environment for his study?

  2. I think identifying serious errors in published work counts as original novel research, especially given that it led to a retraction.

    1. You mean, once you publish a paper and then retract it, it constitutes a contribution to science? Doubtful, to put it mildly.

    2. Although identifying errors and correcting the literature is very important and the student deserves credit, calling it original novel research is a stretch.

    1. This (White Bull) article is really just an opinion piece. It basically says that senior tenured academics muscle into the publications of their junior collaborators without making a substantial contribution to the manuscript. It also says they do it as a result of some personality defect but this is just speculation. It’s just as likely in my opinion to be a structural defect in the academic system whereby most senior authors spend most of their time writing grant applications and participating in administration, so have little or no time for “real science” let alone being in the lab to see it happen. But getting funding is dependent on track record, so having authorships is essential. Catch 22. That doesn’t deny that there are some mean and vain academics around – nor that these people often climb the greasy pole of academia over the bodies of less ambitious (but perhaps more talented) others. But are we surprised? And is that any different to how it’s always been?

  3. “KJ Manton, has denied deliberate wrongdoing” Interesting, let’s have a look at other publications.
    She is first author on this paper (Br J Cancer. 2005 PMID: 15685234) and take a look at Figure 2A. The tumour (T) normal (N) bands have been reused 5 times!!!

    1. Wow, you are right. Just zoom in on the bands and you can see the pixels match exactly..with the naked eye. Looks like another retraction will be coming soon, should be coming soon.

      1. At first glance they all look very similar. However, after zooming in and looking at higher magnification, I dont think it is possible to say with confidence. Some may have been re-used but some could be different. 438 and 795 look very similar at higher magnification also. The others not necessarily so.

        1. The pattern extends beyond the bands over the length of the lanes. A two-lane image repeat seems obvious. Perhaps authors could publish high resolution images on RW.

      2. There is no doubt about this, this is the highest magnification that the journal gives (ie I have not magnified it further)
        http://img24.imageshack.us/img24/6996/tempgwe.jpg
        This is matching down to the last pixel – you can even see the slither of the next band cut off during the cut and paste appearing between each set of two wells.

        At risk of sounding like a broken record I seriously don’t understand how people can be so lazy. How hard is it to produce a gel with this pattern on it? I used to work in a lab where results were routinely falsified but we always maintained the highest standards. The idea of using photoshop in such a crude fashion is just an insult to the entire Guild of Thieves.

        1. Littlegreyrabbit, interesting figure. But not sure how you got that, I see something different when I look at the paper.

          1. James
            Go to this link
            http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2361880/figure/fig2/
            Click on it, it should appear in another window. This window says double click to magnify – you can do this twice to get the highest level available from the journal. This is what I took as screen capture – I made no further magnification than the journal provided.

            And I should add I didn’t spot this – I was only providing it for those who were disputing what Junk Science had said.

      3. This is the homepage of Kerry Manton
        http://staff.qut.edu.au/staff/manton
        I have a feeling she will be allowed to go on splicing to her hearts content – she has the look of an untouchable.
        I would also estimate the chances of Luke Cormack being awarded a PhD as about zero unless he brings in lawyers and QUT decide that it would save them vexation.
        Thats science.

      4. Ooops. Put it in the wrong place. LGR, can you reveal which lab, or what kind of results were faked? Dying of curiosity.

      5. @JonH It’s an exact match, as littlegrayrabbit eluded to: it’s so obvious it’s sad. Do you want something simpler? Stem Cells. 2007 PMID: 17702986. Fig 1. CD row, Chondroitin sulfate chain expression column, Day 7 and 14 are identical. And by the way the Norm panel of Heparan sulfate chain Expression is reused in Fig 2C of the following paper J. Mol. Histol. 2007 PMID: 17653826. If you want to test your skills, have a look at J Cell Physiol paper from 2006, it also has some funny stuff.

      6. In reply to Junk Science, yes J Cell Physiol 2006 Oct;209(1):219-29 Fig. 1F lanes Syd 2 and Syd 4 are the same; actins in Syd1 and Gpc3 are probably the same. Some of the pairs of actins in Fig. 5 look very similar too.

      7. Nice Michael! I also had some difficulties with the actin bands in Fig 5., but it sure looks like several pairs are identical, just some copy and paste with variation added by stretching, cutting etc. The funny thing is that grey squares and even a green-blue one pops up in Fig 1F when you lower the contrast and brightness and increase the saturation I never seen that before, some New Year fireworks for us sleuthing geeks!
        I assume that Luke Cormack and hopefully some QUT significants are following this post. The main conclusion so far should be that Kerry J Manton’s (aka “have denied deliberate wrongdoing”) first author publications should be investigated further by QUT and from my standpoint it’s obvious what the final verdict should be.

      8. Go to the online web version of the figure, click to make the image larger and zoom in. Look at the perfect repeating pixel pattern … I’m guessing that the statistical chance of pixels matching like that five times in a row at random would be will beyond the Avogadro constant.

    2. You are right, 5 times. Impossible not to see this now that JS pointed that out (maybe a blur-effect from your displaying software is distorting the image, if you do not see the replication).

    3. WOW! Good eye! You can see that the same 2 lanes are cut and pasted across the panel. Most of the rest of the paper is bar graphs which could be faked even more easily. Has anyone notified British Journal of Cancer yet?!?!

      1. Thanks for pointing that out. There is also splicing of the left lanes in figure 7A. Tilt the screen back and you can see all the specks in the right lanes of figure 7A. They start about halfway up and go to the top. The specks all line up in both lanes.

        In the left-most lane of 7A there is something like these specks, but they are right at the top. At about 1/3rd the way down there is a break in the specks which has zone without background with straight horizontal boundaries (I can see faint grey lines). Somehorizontal splicing.

        Figure 6B is problematic. In the left-most 2 lanes there is a gently sloping straight line, which marks an abrupt change in signal. Just above this where you might expect the free probe (it is an EMSA) the bands in the other lanes are missing. I think that there is also splicing between the 4th and 5th lanes. Thin, grey/light streak between the bands just over 1/3rd the way up,and then more streaks further up.

        Two of the names, first and last authors, on this publications are present faculty.

        http://www.med.unsw.edu.au/CVRWeb.nsf/page/research

        In the other 3 retractions the first authors are different. 2 of the 3 have the same first author, 1 has another.
        4 retractions. 3 different first authors. Lightning 3 times?

    1. Reply to michaelbriggs December 31, 2012 at

      Thanks for pointing me in the right direction on J Biol Chem. 2010 February 5; 285(6): 4038–4048.

      Images in figure 5D, PCNA and MMP-2 panels overlap.

      The right half of the PCNA panel can be found in the left half of the MMP-2 panel.

      The MMP-2 panel goes a bit deeper, but the overlap is quite subtantial.

  4. I’ve been following Tissue Therapies for a number of years and there seems to be a string of events designed to promote the VitroGro products.

    see http://www.awtrs.org/media/newsletters/AWTRS%20Newsletter%20Vol%202(1).pdf

    “Zee’s most significant research contribution to date is the development and commercialisation of Vitro-
    Gro®, a vitronectin:growth factor protein complex. Zee is the technical founder and Consulting Chief Scientific Officer of Tissue Therapies limited, a biotech spin-out, now ASX-listed that was formed to commercialise the VitroGro® platform. This technology is facilitating the cultivation of human adult and embryonic stem/progenitor cells in defined media for the first time and was recently licensed to Invitrogen
    Corporation. The technology also holds exceptional promise as a wound therapeutic and clinical trials
    are about to commence in Canada in conjunction with Prof Gary Sibbald a world-leading wound clinician.”

    – In 2009 Upton published another paper on media for stem cells. Richards et al. Development of defined media for the serum-free expansion of primary keratinocytes and human embryonic stem cells. Tissue Eng Part C Methods. 2008 Sep;14(3):221-32. doi: 10.1089/ten.tec.2007.0428.

    -This media formulation was then licensed to Invitrogen as described above.

    -No product was released by Invitrogen. Did the media not work? I can find no explanation on the ASX.

    -Is the Richards et al paper (doi: 10.1089/ten.tec) valid?

    -Upton published: IGF-1 Protein Supports Feeder-Cell-Free and Serum-Free Culture of Human Embryonic Stem Cells, by Manton et al., from volume 19, issue 9 (pages 1298-1305). This paper was retracted as described above (contained multiple errors).

    -Was the Manton media formulation tested in other laboratories? What happened in these instances? It looks like there are other medium formulations from the Upton lab (Parker et al and Dawson et al)? Do any of these work?

    -Tissue Therapies was not awarded a CE mark for the Vitrogro product. They have provided little explanation for the delay. Is there a problem that they are not sharing with us?

    -My conclusion is that Tissue Therapies is promoting a shady product.

    1. @AM
      TIS core business is wound healing. They have shown definitively that Vitrogro promotes the healing of chronic wounds (Upton et al. 2011 doi: 10.1111/j.1742-481X.2011.00859.x.). Looks promising to me 🙂

      1. As far as I can tell (after a quick scan of the publication) there are no controls for the treatment, so this study has no practical value.

  5. The uptown presentation on the QUT website, about the ASX company Tissue Therapies, has been removed, sometime between when this article was published and now.

  6. Upton and Manton received an NHMRC (Australian Government) grant for this work in 2009 (Improved culture of cells for human therapies, #553028). It’s possible that this grant application contains questionable figures from the retracted paper. Does anyone know if past NHMRC grant applications are available on request? Like in the US, would having fraudulent figures in an NHMRC grant application be a criminal matter in Australia?

    1. Any fraud may be a criminal matter, defrauding the government certainly is.

      You could do a FOI /Access request of the NHMRC (send it to the top of the chain though, and make sure its electronic)

    2. If a grant reviewer’s comments might be construed as implying research misconduct, the NHMRC has told review panels they cannot take those comments into consideration when scoring the grant for funding.

      1. The NHMRC can tell the review panels whatever they wish until fraud is proven, and the advice seems logical. It is inconsequential until there is evidence to defraud a government of taxpayers money.

        If evidence does come to light there may be a change of policy.

  7. I have seen many people accepted as doctoral candidates that perhaps should not be. What I do find interesting about this story is that someone, who was apparently engrossed in the work and a current author of the paper, did not realize the mistakes until his thesis was externally rejected. It seems that this group is under a bit of pressure now, and was perhaps naive to have published this data prematurely. Like so many scientific studies, is that they may work in the lab under certain conditions, but they are not suitable for industry or medical purposes. Does that make the data less valid?

  8. Courtesy of hotcopper.com. A list of questions addressed to TIS, the company linked to the retracted paper described above.

    re: stem cell retraction (cormackl) Trade TIS

    I would like to clarify some information between Tissue Therapies and shareholders. Can tissue therapies please verify each one of these points?

    1. Did Tissue Therapies claim to have made the worlds first serum free and feeder cell free embryonic stem cell media using the VitroGro formulation?
    2. Did Tissue Therapies licence this formulation to Invitrogen?
    3. Did Invitrogen fail to release a product based on this formulation?
    4. Did you have two external companies (Australian Stem cell Centre and Stem Cell Sciences) test the VitroGro media formulation and did they find that it did not support embryonic stem cells?
    5. Did tissue therapies scientists submit a paper to stem cells and development that contained errors in almost every figure in the Manton et al. paper (2010)?
    6. Was this paper retracted by the Journal?
    7. In the paper review process of the Manton et al. (2010) paper, did tissue therapies scientists tell stem cells and development editors that no one else had tested the media with other cell lines when in fact other labs had?
    8. Was this untrue at the time?
    9. Are you planning to republish the Manton et al. (2010) data set?
    10. Did you claim to have cultured two cells lines when in fact you knew at best you had only cultured one?
    Item 10 is in ref to ASX published document Tissue Therapies Ltd annual general meeting 27 November 2008.
    11.Did your 2009 NHMRC grant application (553028) contain modified data in its only two figures, which you changed during your editing?

    I’m really looking forward to the answers, as I’m sure all shareholders are.

    1. @kelly
      Faked data in paper +
      Faked data in a grant +
      Faked data in stock exchange release to share holders +
      Failure to provide training environment for phd student +
      Errors found in 5 additional publications

      =..errr…wait..does not equal misconduct..err…= inadvertent???

        1. I have read through the paper you are talking about. It seems straight forward. In fact, with the relatively low “passage number” published, it is a surprise that a major company like Invitrogen was interested in a licensing deal in the first place.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.