“I do wish that journal editors would not take six years to perform an investigation and to retract.”

In July 2014, Elisabeth Bik notified PLOS ONE that she’d found three papers in the journal by a group of researchers who had clearly manipulated figures in the articles. 

More than six years later, the journal has finally retracted the publications. 

The authors were affiliated with the Fourth Military Medical University in Shaanxi, China. The notice for 2013’s “Hyperthermia-induced NDRG2 upregulation inhibits the invasion of human hepatocellular carcinoma via suppressing ERK1/2 signaling pathway” refers to overlapping data in Figure 1A, similarities between data in Figure 1B and Figure 6C despite the fact that lanes “represent samples exposed to different experimental conditions,” and more:

The corresponding authors commented that the invasion results for HepG2 and Huh7 cells were highly similar in the 37 and 39 degree groups. The original data underlying the results reported in Fig 1A are no longer available.

The authors provided available image data to support the results reported in Figs 1B and 6C, but these data did not resolve the concerns. The MMP-2, MMP-9, and Tubulin data reported in Fig 6C appear to match image data provided in support of Fig 1B, albeit with different blot and lane labels, suggesting that all three panels of Fig 6 may have reported the wrong data. None of the image data provided appear to match the results reported in Fig 1B. The authors commented that no additional western blot data are available currently.

In light of the unresolved concerns, the PLOS ONE Editors retract this article.

LY, JZ and WL agreed with retraction and apologize for the issues with the published article. The other authors either could not be reached or did not respond directly.

That’s similar to the notice for another 2013 article, “Role of activated Rac1/Cdc42 in mediating endothelial cell proliferation and tumor angiogenesis in breast cancer.” Reference 3 in the statement is “Inhibition of endothelial cell proliferation and tumor angiogenesis by up-regulating NDRG2 expression in breast cancer cells,” which was published in 2012 — and has now also been retracted for similar reasons. 

We emailed Wenchao Liu, a corresponding author on two of the articles, for comment but did not receive a reply. 

David Knutson, the senior manager of communications for PLOS ONE, told us that the delay resulted from a backlog of retraction investigations at the journal: 

Elizabeth Bik notified PLOS ONE of concerns about these articles in December 2014, amidst a large volume of other incoming image cases that exceeded the journal’s editorial capacity for case handling. PLOS’ Publication Ethics team is continuing to work through these older cases, prioritizing those in which more serious issues were raised. In this context, we recently concluded this case.

We asked Knutson for an estimate of how many potential retraction cases the journal was currently investigating, but he demurred: 

We cannot discuss unresolved cases, as any answer would be fairly inaccurate.

Bik told us: 

Two of these PLOS ONE papers, published in April and June 2013, respectively, were found during our screen of 20K papers for image duplications (published here: https://mbio.asm.org/content/7/3/e00809-16). The third paper, published in PLOS ONE in February 2012, came up after I checked more publications from these authors. All three papers were reported to the journal editors in July 2014, with a reminder to the editors in November 2016. It is sad that it took more than six years to investigate and retract these papers.

Bik noted that journals generally don’t retract for figures with overlapping panels, which are present in the two 2013 articles. 

They will often accept apologies by the authors and correct the paper. In this case however, the editors might have looked at the overall situation, with three papers by the same group displaying similar problems, and they might have decided that the problems were too serious to address with just a correction. 

The decision to retract is in my opinion the correct one. I do wish that journal editors would not take six years to perform an investigation and to retract. According to Google Scholar, these papers have been cited 42, 27, and 56, respectively. That means they have been used by many other researchers to build their studies on. If papers would be retracted faster, that would send out a signal to other researchers sooner that those papers have serious concerns. That would result in much less wasted time and effort. Still, it is a good decision that these papers are now corrected.

Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].