Sonia Melo case: PhD advisor Esteller sham-investigated, postdoc PI Kalluri with $ 80Mio COI

Sonia Melo, Portuguese cancer researcher and recipient of the prestigious EMBO Installation Grant, now has her publications investigated by EMBO for suspected image manipulations. Her current and former research institutions are apparently actively avoiding any attempts to scrutinise her papers, some due to very heavy financial conflicts of interest. Yet my information suggests that Melo’s former PhD advisor and co-author, Manel Esteller, is being presently investigated by his research centre in Barcelona.

Melo already had to retract a first-author publication from her PhD period with Esteller from the journal Nature Genetics. The retraction notice admits image duplications:

“We have recently become aware of the presence of duplicated images in the Figures 3 and 4 and Supplementary Figures 5 and 6 in our publication Nat. Genet. 41, 365–370, 2009, that were assembled according to the specified author contributions. We therefore retract the publication for the sake of the high standards we expect for research and scientific journals. All the authors have signed this statement”.

1skhskq

Melo herself told RetractionWatch:

“As for further comments/concerns, which arose on PubPeer, the authors and respective institutions have not ignored them. All concerns were addressed, raw data registered on notebooks was analysed and reports developed for each case. The data was shown to be correct so no further action was taken”.

It appears, she spoke the truth. These are three important stations of Melo’s career in academic research institutions, and none of them is apparently interested in taking a closer look at her papers:

  1. IPATIMUP, University of Porto

Here, Melo has only recently established her own research lab, at the Instituto de Patologia e Imunologia Molecular da Universidade do Porto (Ipatimup). The Ipatimup scientists and directors remained unresponsive to my inquiries. One could therefore speculate that Melo’s research is not being investigated there. Why? Generally, scientists under ongoing investigation do not communicate with media, they are often specifically forbidden to do so.  Yet Melo was able to freely share her views with RetractionWatch, even to publicly declare the absence of institutional investigations at her current and former places of work.

 2. University of Texas MD Anderson Cancer Center

Melo used to work as postdoctoral scientist at the lab of Raghu Kalluri, Chair of the Department of  Cancer Biology and specialist for exosome (extracellular secreted vesicles) research. Kalluri has five publications featuring Melo as first or second author listed on PubPeer, including in prestigious journals Nature and Cancer Cell. Here an example from the latter paper:

f5krhwn

However, also some other Kalluri papers, where Melo was not co-author, were flagged. Earlier corrections acknowledged inappropriate figure assembly which has happened “accidentally”, e.g. this LeBleu et al 2013 paper in Nature Medicine:

u8lucbj

Most importantly, there are reasons to worry that any MD Anderson investigation into Kalluri’s publications with Melo might be compromised by a large financial conflict of interest. Melo’s and Kalluri’s 2015 common paper in Nature is the basis of an $80 million investment into a biotech company Codiak Biosciences, as Forbes reports:

“The little company is announcing today it has raised $80 million in committed venture capital spread over two installments. Two venture firms known for creating ambitious biotech companies, ARCH Venture Partners and Flagship Ventures, pieced it together to build on exosome biology from the lab of Raghu Kalluri at MD Anderson Cancer Center in Houston, Tex. Fidelity Management and Research, the Alaska Permanent Fund and Alexandria Venture Investments joined the financing […]

The initial focus of Codiak will be pancreatic cancer. It’s a notoriously tough-to-treat cancer, and it’s often diagnosed late in the game. People usually only have a few months to live after diagnosis. The company will seek to build on research Kalluri’s team published in Nature in June. At that time, Kalluri said his team had identified a proteoglycan (a protein covered with lots of carbohydrate molecules) called Glypican-1, which was present on the surface of exosomes.  […]

Gillis [managing director at Arch Venture Partners, -LS] said Kalluri’s group has generated more data from animal tests, which hasn’t yet been published. “It’s about being able to target various oncogene targets that have previously failed to be targetable via small molecules,” Gillis said. Kalluri, he said, “has gotten some dramatic data in animal models. If we can rapidly exploit that clinically, it would be a big win””.

How exactly is MD Anderson involved into Codiak Biosciences research on endosomes? According to information available to the industry news site FierceBiotech, rather directly:

“MD Anderson–which has been busily spinning out new biotechs–signed off on an IP [Intellectual Property Rights, -LS] license and a sponsored research agreement, keeping the big cancer center directly engaged in the R&D work ahead”.

In fact, Codiak Biosciences was apparently (officially or unofficially) founded by Raghu Kalluri himself, as the website GenomeWeb reports:

“The firm has licensed exosome-related technology that Kalluri developed at MD Anderson, including using the small, extra-cellular vesicles to deliver therapies and analyzing their nucleic acid and protein contents for diagnostic purposes”.

The other, official, company founder is none other than Eric Lander, president of the Broad Institute of MIT and Harvard. Lander is a top scientist and also well versed in money issues: he was recently widely criticized for his article in Cell , where Lander attempted to exclusively assign the CRISPR discovery credit to his Broad Institute, which accidentally is also a key contender for the potentially billions-dollar worth patent on CRISPR genome editing technology.

Should this Nature paper (image below) or other Kalluri publications, especially those featuring Melo, be retracted, the big investors will likely lose trust in Codiak Biosciences and might be asking MD Anderson for their money back, with interest. Neither the Texan Cancer Center nor the biotech company can afford to be caught with wasting investors’ money on any questionable research, especially if tainted by misconduct and data manipulation.

y3qyeiz

Thus, Melo was probably telling the truth that MD Anderson decided not to investigate her publications with Kalluri. With $80 Million at stake, it was probably a financially very wise decision, maybe less so in the sense of academic research integrity.

 3. IDIBELL in Barcelona

Melo performed her PhD studies in the lab of Manel Esteller, director of strategic projects at the Institut d’Investigació Biomèdica de Bellvitge (IDIBELL). Three of Melo’s papers with Esteller were flagged on PubPeer. About the now retracted paper in Nature Genetics, Esteller announced on PubPeer in November 2015:

“Dear PubPeer Colleagues,

Thanks for your informative messages and insightful comments. I acknowledge the obviously serious nature of the case. The matter will be promptly and thoroughly investigated and appropriate action will be taken. The first author, that left the lab years ago, has been contacted to ask for a formal and detailed explanation. The conclusions will be disclosed in the journal and in this forum. I also take the opportunity to stress that the collaborators that coauthor the manuscript or myself as corresponding author have absolutely nothing to do with the issues raised.

As I have always done in my twenty-three years research career, truth and science should always come first.

Keep the good work,

Dr. Manel Esteller”

When asked about any institutional investigations happening, Esteller wrote to me in an email in December 2015:

“We do not have currently a formal ombudsman for this issues in our place, it is in the making”.

It seems now, Esteller himself will be the first case whom IDIBELL’s new Ombudsmen will be investigating.

Like Kalluri, he also has a number of problematic papers where Melo is not an author. Band duplication’s were suspected by PubPeer commenters in Esteller’s papers in Cancer Research, Oncogene and Nature Communications, like this here:

smkxhiv

Now, the director of IDIBELL, Jaume Reventós, wrote to me:

“As you know, since many years, Sonia Melo is not any longer in our institution and she is pursuing her career elsewhere. Concerning the publications in which Dr. Melo is not an author, I want to let you know that we will pursue our inquiry following our standard procedures for those matters.

We support a fair and clean science from our scientists and believe that fraud doesn’t not have place in our institution”. 

This on-going IDIBELL investigation might also explain why Esteller has not replied to my emails since. I asked Reventós for clarifications, but received none. My interpretation is then, that IDIBELL will not be investigating Melo’s papers from Esteller lab, since those are likely already being analysed by the EMBO commission, as EMBO director Maria Leptin announced. Instead, IDIBELL will probably focus their investigative efforts on other papers by Esteller, sans Melo, which were flagged on PubPeer. Thus, EMBO is seemingly indeed the only institution investigating Melo’s publications for suspected data manipulations.


Update 29.08.2016. In fact, the IDIBELL investigation of Manel Esteller’s work was a sham (hence correction to the title of this article). As described in my follow-up article, IDIBELL declared Esteller as not responsible for data integrity in his lab’s papers  and tasked him with investigating himself.

Recent news may explain such peculiar attitude of IDIBELL. There was a major financial COI hindering any attempt at investigation: just as MD Anderson with Kalluri, the Spanish institute has similar business interests with Esteller. Together with the also Barcelona-based pharma company Ferrer, Esteller has developed a cancer diagnostics test called EPICUP (see press release and The Lancet Oncology paper Moran et al 2016). A patent was filed in 2012 and approved in January 2016, the inventor is Esteller and the applicant is IDIBELL. Unlike with Kalluri and his Codiak Biosciences, Sonia Melo was not part of this research by Esteller and IDIBELL.

58 thoughts on “Sonia Melo case: PhD advisor Esteller sham-investigated, postdoc PI Kalluri with $ 80Mio COI

  1. The Codiak link makes me wonder whether there isn’t room for civil lawsuits against MD Anderson for fraud (this may go beyond “losing trust” to “feeling defrauded”). Faking documents to obtain 80 Million is possibly even criminal.

    Like

  2. Accidental error when the IDIBELL Director Jaume Reventós used the double negative “doesn’t not”? Two negatives make a positive, and this is common in Portuguese and Spanish, so, was he really trying to say “We support a fair and clean science from our scientists and believe that fraud does have place in our institution”? Oh, the irony of the double negative!

    The University of Porto is one of Europe’s oldest and most reputable academic institutions, and most likely one of the premier ones in Portugal, alongside University of Coimbra, Lisbon University and Evora University, both in terms of historical reputation and academic excellence.

    If there is anything Portuguese hate more than a scandal, it’s a cover-up. So, if there is any hint at a cover-up or insufficient accountability or transparency, please report to other higher education authorities in Portugal, including the Rector of University of Porto.

    There are enough hints here for those wishing to follow-up:
    https://en.wikipedia.org/wiki/Higher_education_in_Portugal

    Like

  3. Melo / Esteller entries at PubPeer based on COPE member analysis.

    The following are all COPE member journals:

    Melo and/or Melo+Esteller
    Oncogene
    https://pubpeer.com/publications/20802525

    Cancer Cell
    https://pubpeer.com/publications/EB26CFF7EAE4F3E766EF305622527B
    https://pubpeer.com/publications/C729E05C2FCE20FCBF28D64F4B494F

    Nature
    https://pubpeer.com/publications/26106858

    Journal of the American Society of Nephrology
    https://pubpeer.com/publications/23274427
    https://pubpeer.com/publications/2971EE7D7625B040F9E6EC82E0B8EE

    PNAS
    https://pubpeer.com/publications/21368194

    Esteller (without Melo)

    The EMBO Journal
    https://pubpeer.com/publications/70F59BBD8E2CE2F8E24CE176A57615
    https://pubpeer.com/publications/15719023

    Nature
    https://pubpeer.com/publications/16357870

    Carcinogenesis
    https://pubpeer.com/publications/17449905

    Oncogene
    https://pubpeer.com/publications/18223687

    Nucleic Acids Research
    https://pubpeer.com/publications/23558747

    British Journal of Cancer
    https://pubpeer.com/publications/18087279

    Nature Communications
    https://pubpeer.com/publications/6359D689D6E4C8E6F8C7F5FEC310A4

    Clinical Cancer Research
    https://pubpeer.com/publications/19088041

    Carcinogenesis
    https://pubpeer.com/publications/16377805

    Molecular Cell
    https://pubpeer.com/publications/24582497

    Cancer Research
    https://pubpeer.com/publications/19861541

    Journal of Cell Science
    https://pubpeer.com/publications/19638407
    https://pubpeer.com/publications/12508111

    Non-COPE member journal queries (Esteller)

    RNA Biology
    https://pubpeer.com/publications/23611944

    Please observe the ratio of COPE to non-COPE member journals:
    21 : 1

    Like

    1. What is the ratio supposed to signify?

      Do you even know how a chi-squared test is supposed to work? You need to supply expected values for your “ratio” to mean anything or be worth “observing”.

      Like

  4. Honestly, I have been working at IDIBELL, and as Schneider said IDIBELL is never going to investigate Manel Esteller due to a heavy conflict of interest. Dr. Esteller is like a demigod over there, has brought millions of euros to the institution, he has the largest lab, and he is also a key figure in the establishment of the institution and politically well supported, with the information provided here an investigation will never go to anywhere.

    Like

    1. Anon anon, alot more than just explaining required. The validity of this Keystone meeting may be in question (see list that follows next).

      Raghu Kalluri, a co-author on many Melo/Esteller papers, will also be attending:
      https://pubpeer.com/search?q=Kalluri&sessionid=43BB398907927EE739F2

      On PubPeer, and also attending the same Keystone meeting:

      Napoleone Ferrara:
      https://pubpeer.com/publications/2FEACDED255496DC95EBA283E7A5F4

      David Lyden/Johan Skog:
      https://pubpeer.com/publications/22635005
      https://pubpeer.com/publications/25985394
      https://pubpeer.com/publications/22891351
      https://pubpeer.com/publications/DCC833B438230851C78D20E7D78A81

      Louise Laurent
      https://pubpeer.com/publications/25008523

      Hakho Lee
      https://pubpeer.com/publications/9A5889FAF668706D9B3091A6D16DE0

      Maybe Dr. Schneider could contact the Keystone Meeting organizers urgently to alert them to the Melo “controversy”? A word of caution, though: we all know Keystone’s defensive posture, as was clearly displayed in the Voinnet scandal, and documented at Retraction Watch.

      Like

      1. All of the scientists mentioned above by Plantarum must be urgently investigated if they are not being already. They all distort science as mentioned by Prof. Randy Schkeman.

        Like

      2. Is not really a big surprise that Susana Gonzalez has got a new tenured job…the same way it will be very hard that M. Esteller will be fully investigated at any time….
        Plantarum, by the way it seems David Lyden and Napoleone Ferrara won’t be anymore at the Keystone Symposia: http://www.keystonesymposia.org/16E4
        Nevertheless, we still can ask the remaining scientists mentioned above and Ayuko Hoshino (Lyden lab) if they have something to tell us about the PubPeer comments relative to their respective papers

        Like

  5. Given the way science works in Spain as it is visible by the Susana Gonzalez case, I would investigate specially the responsibility of the Spanish scientist Manel Esteller in the Sonia Melo/ Esteller/ Kalluri case.

    Liked by 1 person

    1. There is still to much water to go under the bridge in what relates to the Brexit, but one of the things for sure British people would be happy for is not to have to contribute for funding grants such as the Susana
      Gonzalez ERC grant or Sonia Melo EMBO funding.

      Like

  6. Dodgy science is everywhere, in every country, science is very international nowadays…in Spain, in USA, Germany, China, Malaysia, UK….we just have to combat it….given the Brexit subject is in the order of the day I think governments from European countries should take the chance to find new ways for governing science in what concerns science integrity and publish, so that cases like Sonia Melo, Susana Gonzalez and other ones can be fairly solved with the adequate penalties so that science can become more fair…..in this point hopefully Leonid Schneider journalism work and his blog might a very good help.

    Like

  7. Not surprising. Let’s just say Esteller has a number of papers that made untruthful or incompetent statements but are somewhat in the gray or ambiguous area so they weren’t targeted for investigation. These untruths add up, though, and really makes me wonder how his papers are being approved for publication. What happened to solid peer review? Then again, see below…

    Disclaimer: I am biased against Esteller due to the fact, as a peer reviewer, he rejected a paper submitted by my lab I used to work at, claiming the findings were not significant. We resubmitted the paper to another journal, which ended up having 100+ citations and a multitude of papers afterwards corroborating our findings and data. Needless to say, this made us question Esteller’s true expertise in his field. It is sad that he has so much influence in the scientific research community in Spain.

    Liked by 1 person

    1. How do you know that Esteller served as the reviewer? Unless it was an open peer reviewed journal. Can you provide proof of your claim?

      Like

  8. UPDATE relative to the comments above and realtive to the link above: Tactic Or Strategy: Missiles Into Syria, DePinho Out Of Office – Dr. Len Zwelling

    Recently, several papers were published that do not confirm at all the Melo GPC1 paper. Please, see: https://forbetterscience.com/2016/11/24/post-publication-peer-review-of-a-multimillion-dollar-heavy-nature-paper-by-ana-pedro/. Also, as I mentioned before my own original mass spectrometry data does not confirm the findings of Melo et al., 2015 GPC1 paper.

    If Dr. Kalluri will be investigated or not? Probably to start an investigation on Dr. Kalluri’ s work someone anonymously or not anonymously will have to make a complaint relative to misconduct on Dr. Kalluri’s lab to the MD Anderson RIO and provide very clear and irrefutable proofs of misconduct. I challenge those ones who may have those proofs to present them at the MD Anderson RIO.

    Finally, I am presently trying to rescue my unfortunately preliminary, although 100% data from my postdoc and waiting for the elife replication study of Peinado et al., 2012

    Like

  9. UPDATE relative to the comments above and realtive to the link above: Tactic Or Strategy: Missiles Into Syria, DePinho Out Of Office – Dr. Len Zwelling

    Recently, several papers were published that do not confirm at all the Melo GPC1 paper. Please, see: https://forbetterscience.com/2016/11/24/post-publication-peer-review-of-a-multimillion-dollar-heavy-nature-paper-by-ana-pedro/. Also, as I mentioned before my own original mass spectrometry data does not confirm the findings of Melo et al., 2015 GPC1 paper.

    If Dr. Kalluri will be investigated or not? Probably to start an investigation on Dr. Kalluri’ s work someone anonymously or not anonymously will have to make a complaint relative to misconduct on Dr. Kalluri’s lab to the MD Anderson RIO and provide very clear and irrefutable proofs of misconduct. I challenge those ones who may have those proofs to present them at the MD Anderson RIO.

    Finally, I am presently trying to rescue my unfortunately preliminary, although 100% honest data from my postdoc and waiting for the elife replication study of Peinado et al., 2012

    Like

    1. Additionally, a senior editor from a prestigious journal considered very adequately and as expected my data as too preliminary: “For such a study to be published in …. we would expect to see a much more convincing examination of these potential biomarkers from both a mechanistic and clinical perspective, including robust replication in independent data sets.” I therefore ask myself if in the studies published bellow is done a convincing examination of potential biomarkers including robust replication in independent data sets?
      https://pubpeer.com/publications/26106858 – Melo et al. 2015 (GPC1)
      https://pubpeer.com/publications/22635005 – Peinado et al. 2012
      https://pubpeer.com/publications/25985394 – Costa-Silva et al., 2015
      https://pubpeer.com/publications/8F2A681C84FB1D9FC6A9D0B46F5E0A – Hoshino et al., 2015

      Like

  10. Polemic ideas about research funding and scientific career:

    no one should have an 100% research job
    no one should be paid just to perform research
    Postdocs must be abolished
    any scientist should have a job not necessarily related with research and should apply for research funding
    the results of any funded research should be openly-accessed and their application in society should be clearly visible

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s