More than five years after comments appeared on PubPeer about a 2012 paper in PLoS ONE with a raft of problematic images — and a deceased member of the group whom the corresponding author suggests might have been able to support the validity of the data — the journal has retracted the article.
The article, “Placental expression of CD100, CD72 and CD45 is dysregulated in human miscarriage,” was written by a team of researchers at the Università Politecnica delle Marche, in Ancona, Italy. The first, and corresponding, author of the paper was Teresa Lorenzi, of the school’s Division of Neuroscience and Cell Biology.
The paper has 19 citations, including two in 2019, according to Clarivate Analytics’ Web of Science. The lengthy notice begins with a rundown of 14 questions about three of the paper’s figures. We’ll spare you the entire catalog of ships, but here are a few examples:
After this article [1] was published, the following concerns were raised about the results reported in Figs 5, 7 and 9:
Fig 5: There appear to be vertical discontinuities after lanes 2, 4 of the first trimester nCD100 panel, after lane 3 of the first MC nCD100 panel, after lanes 2, 3, 4, 5 of the second MC nCD100 panel.
Fig 9, first trimester CD45 blot: There are similarities in background signal above and below the bands in lanes 1, 3, and there are similarities in background signal below the bands in lanes 2, 4.
The notice continues:
The authors provided the available original blot data to support the nCD100, CD72, and CD45 data reported in Figs 5, 7 and 9, but the data provided did not resolve the concerns and did not include matching data for all results in question. Corresponding author TL indicated that other underlying data for the reported experiments are unavailable due to issues including the amount of time that has passed (seven years) since the article’s publication, as well as the death (ALT), retirement (AT, AM, MC), and inability to contact (FP, LL, FM) some of the article’s authors.
In regards to the β-actin duplications, the authors commented that the same samples were analyzed in Figs 5, 7 and 9 and so the same β-actin controls were used for the corresponding panels of these figures. However, the β-actin data were not provided and we were unable to verify the authors’ claim based on the available data.
The authors disagreed with similarities of β-actin data across panels within figures (points 11–14, above). In regards to points 7–10 the authors raised that there are some differences between the data for which similarity concerns were raised. However, per our editorial assessment, the pixel similarities in the areas mentioned in points 7–10 are more similar than would be expected for different experimental results.
Overall the comments and data provided by the authors did not resolve the issues outlined above. The PLOS ONE Editors retract this article due to concerns about the validity and reliability of the reported results. We regret that the above issues were not identified and addressed during the article’s pre-publication peer review.
TL did not agree with the retraction and stands by the reported results. ML and DM disagree with the retraction. AT, FP, FM, LLS, MM, PC, AM and MC did not respond or could not be reached. ALT is deceased.
We emailed Lorenzi for comment but have not heard back.
PLOS picks up the pace
As we noted in April, PLOS ONE has become a “major retraction engine,” with 91 so far this year. That seems to be due to the fact that the journal has hired new staff to work through a backlog of allegations brought by various readers, in particular Elisabeth Bik.
Pseudonymous whistleblower Claire Francis emailed PLOS with a link to the PubPeer comments on March 15 of this year. “Note that many of the problematic data were pointed out in words 5 years ago,” Francis wrote.
David Knutson, the senior communications manager at PLOS, told us that the clock didn’t start running on the retraction until March:
We were notified directly of the concerns about this article in March, 2019, although issues had been raised earlier on Pubpeer. Per our policy, readers should notify the journal directly of any concerns about published work, and should not rely solely on Pubpeer, social media, or other third party sites.
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].