Retraction Watch

Tracking retractions as a window into the scientific process

Mystery: PLOS One seeks investigation after publishing two papers with “substantial overlap”

with 10 comments

Screen Shot 2015-10-01 at 4.27.21 PMPLOS One has retracted one of two cancer papers with “substantial overlap” that were reviewed simultaneously by different editors.

This one’s a bit of a mystery — neither of the papers share an author, and no authors share institutions. Once the editors discovered the overlap, they contacted the authors. One group of authors provided the requested documentation for the experiments. The other did not — so the editors retracted that article, even though it was published months before the other one.

In the meantime, the editors have asked the authors’ institutions investigate how the articles — which contain entire identical sentences, and some extremely similar figures — were put together. According to a statement from the editors:

We are awaiting an update from both institutions.

The first paper, about microRNAs in colon cancer, was published in December, 2014 by authors based at the General Hospital of People’s Liberation Army in China. It has not been cited, according to Thomson Scientific’s Web of Knowledge. The second paper, about microRNAs in prostate cancer, was published in April 2015 by authors based at institutions in Shanghai. It’s been cited once.

Despite the mystery surrounding these papers, some things are clear — the text definitely overlaps. The first sentence of both abstracts is identical, and here’s the last sentence of the abstract of “MicroRNA-302a Functions as a Putative Tumor Suppressor in Colon Cancer by Targeting Akt“:

These results reveal miRNA-302a as a tumor suppressor in colon cancer, suggesting that miRNA-302a may be used as a potential target for therapeutic intervention in colon cancer.

The experiments in the prostate cancer paper, “MicroRNA-302a Suppresses Tumor Cell Proliferation by Inhibiting AKT in Prostate Cancer,” do…pretty much the same thing:

These results reveal miRNA-302a as a tumor suppressor in PCa, suggesting that miRNA-302a may be used as a potential target for therapeutic intervention in PCa.

The retraction note for the 2014 paper elaborates:

There is a substantial overlap in the text, and multiple figures appear similar or identical despite labels and legends indicating that the data were obtained from different cell and tissue types (prostate cancer cell lines and tissues in the above-mentioned article versus colon cancer cell lines and tissues in the retracted article).

Our follow-up in relation to how the overlap with the article by Zhang et al. (DOI:10.1371/journal.pone.0124410) [2] arose is ongoing. Upon our follow-up with the authors of the current article by Sun et al., we have been unable to obtain the underlying data and approval documents related to the study, and as a result, the PLOS ONE editors have been unable to verify the integrity of the work. In light of these concerns, the PLOS ONE editors retract this article.

It has been brought to the attention of the PLOS ONE editors that there is substantial overlap in content between this article and another article published in PLOS ONE by different authors: “MicroRNA-302a Suppresses Tumor Cell Proliferation by Inhibiting AKT in Prostate Cancer” (DOI: 10.1371/journal.pone.0124410) [2].

For an example of figure overlap, take a look at these two images, which are nearly identical, except for the labels. (The faded image is from the retracted paper):

A PLOS spokesperson passed on a statement from the editors. The statement explains that they are not sure how the overlap occurred — since the manuscripts were edited simultaneously:

A reader raised the similarities between the two articles to the attention of the editorial office. Our follow up involved steps aimed at obtaining information on whether and how the authors of ‘MicroRNA-302a Functions as a Putative Tumor Suppressor in Colon Cancer by Targeting Akt’ could have obtained access to the manuscript by Zhang et al. prior to its submission to PLOS ONE. Unfortunately, after following up on different possible avenues, we have not been able to establish with certainty how the duplication occurred.

The two manuscripts were submitted within a short period of time and were in review simultaneously, handled by different Academic Editors. Since December last year, PLOS ONE automatically screens all manuscripts against earlier submissions. This functionality is designed to prevent this type of situation arising again in future.

To find out which article was plagiarized, they requested documentation from both sets of authors:

We have contacted the authors of the two articles. As stated in the retraction notice, we contacted the authors of ‘MicroRNA-302a Functions as a Putative Tumor Suppressor in Colon Cancer by Targeting Akt’  in relation to the planning and oversight for the study and the underlying data but we did not receive the requested documentation. Zhanget al. have, on the other hand, provided the documentation requested by the editorial office. We have contacted the General Hospital of People’s Liberation Army and the Fudan University Shanghai Cancer Center, the institutions where the authors of the two articles are based, and we have requested that they investigate the circumstances that led to the publication of the two articles. We are awaiting an update from both institutions.

The retracted paper was received in September 2014, and accepted in early December. The other paper was received in October, and accepted in March. We don’t know whether it had been submitted elsewhere first.

We’ve emailed corresponding author of the retracted paper — Yi Ling, of the General Hospital of People’s Liberation Army in China — and the corresponding authors of the prostate cancer paper, to ask if they had previously submitted the paper to another journal. We will update this post if we hear back.

 Hat tip: Rolf Degen

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.

Written by Shannon Palus

October 6th, 2015 at 2:30 pm

Comments
  • Query October 6, 2015 at 2:53 pm

    Amazing. Do you think that you will receive a response from the General Hospital of People’s Liberation Army in China?

  • Sylvain Bernès October 6, 2015 at 4:23 pm

    Perhaps the 44 male patients included in these studies were diagnosed with both colon and prostate cancer? Poor guys!

  • EDit.or October 6, 2015 at 4:34 pm

    Can PLoS please name the two editors who oversaw each of the papers?

    • Bobo October 6, 2015 at 4:36 pm

      Umm, editor names are already publicly available on every single PLOS article.

    • QStel October 6, 2015 at 4:47 pm

      The retracted paper was edited by Chun-Ming Wong, University of Hong Kong, Hong Kong. The other paper was edited by Jun Li, Sun Yat-sen University Medical School, China.

    • Raymond Wan October 6, 2015 at 8:34 pm

      Not too sure about the relevance of the editors’ names or the reviewers’ — I don’t think it’s their fault. For better or for worse, the reviewing system now takes the approach of “innocent until proven guilty”. Whether it is a high impact journal or a low impact one, editors and reviewers assume that the authors have done honest research when they first look at the paper.

      Maybe this approach needs to change and when a paper is submitted for review, the submission system somehow checks the paper for plagiarism automatically? I think this is somehow inevitable…hopefully it includes automatic blacklist of authors, too!

      • SB October 7, 2015 at 4:14 am

        I agree that papers should be checked for plagiarism automatically but I suspect that as they were reviewed simultaenously this would have escaped detection. Most plagiarism detections packages would only capture already published work in their checks.

  • QStel October 6, 2015 at 4:45 pm

    Are the tumours on those mice in line with regulations on animal experiments? They look huge.

    • ohman October 9, 2015 at 11:36 am

      Each institution’s Animal Care and Use Cmte (ACUC) will establish their own policies for things like maximum tumor burden, but there tend to be similarities among such policies. For instance, a tumor (or tumors in total) that is equivalent to 10% of the body weight of the mouse (after subtracting the tumor weight), or a tumor that impedes/interferes with normal behavior (feeding, drinking, ambulation) are not uncommon criteria for removal of the mouse from the study. It must be kept in mind however that tumor burden should not be looked at in isolation. For instance, a tumor that is ulcerated, such as that on the rightmost mouse, is often a criterion for removal regardless of tumor size. Also, bear in mind that these limits can be exceeded if the investigator can scientifically justify such exceptions to the satisfaction of the ACUC, which would then formally approve the exception.

      As for the mice in the figure, it is difficult to tell without first-hand examination, unlike the mice of concern here: http://retractionwatch.com/2015/09/16/animal-welfare-breach-prompts-nature-correction/. Nonetheless, the mouse with the ulcerated tumor would be of potential concern. But, of course it is dead and thus not in the study at the time the photograph was taken; so perhaps it was removed from the study in an appropriately timely manner.

      • QStel October 10, 2015 at 5:42 am

        Thanks ohman for your answer!

  • Post a comment

    Threaded commenting powered by interconnect/it code.