Conflict of interest, figure issues net retraction for cancer paper

Two major problems sunk this cancer paper.

For one, many of the images were copied from another paper. In addition, one of the authors did not disclose that he was the president of a related company, nor that his company provided reagents for the experiments.

It’s not clear when the paper was published, but The paper was published on October 16, 2014, and a withdrawal notice went up on January 16, 2015. Here’s the retraction for “Enhanced Detection and Phenotypic and Karyotypic in Situ Characterization of Circulating Tumor Cells”:

After manuscript acceptance and publication online as an uncorrected author proof, irregularities in conflict of interest disclosures and data presentation were noted in the article entitled “Enhanced Detection and Phenotypic and Karyotypic in Situ Characterization of Circulating Tumor Cells,” which appeared online October 16, 2014, in The American Journal of Pathology (http://dx.doi.org/10.1016/j.ajpath.2014.08.023). It was determined that although author employment by Cytelligen (San Diego, CA) was disclosed, Dr. Peter P. Lin did not disclose his role as President of Cytelligen, which according to the company website “specializes in circulating rare cell including circulating tumor cell (CTC) related clinical application and scientific research.” Cytelligen was also not disclosed as being a source of reagents in Materials and Methods. Additionally, figures present in the article had been previously published in other journals. Specifically, a figure labeled as a circulating colon cancer cell was previously published in Clinica Chimica Acta (2013, 419:57–61) labeled as lung cancer CTC. Another figure showing CellSearch data was previously published in Oncotarget (2014, 5:6594–6602). Following editorial communication with the authors, corresponding author Dr. Marc Shuman requested that the manuscript be withdrawn, with concurrence of the Editors. Acceptance of the manuscript is therefore being withdrawn from The American Journal of Pathology by the American Society for Investigative Pathology. All authors have been notified of this decision.

We should note that not all withdrawals at Elsevier are followed by clear retraction notices, so this seems a step in the right direction.

We reached out to Peter Lin, who sent us a long note of explanation. You can read the entire thing here. Lin had disclosed that he worked for Cytelligen, but not that he was president. According to Lin, after the manuscript was accepted, the editors contacted the authors about Lin’s position. The editors also asked if any of them owned stock in Cytelligen – according to Lin, they do not. As for not disclosing where the reagents came from:

We all thought the aim of this manuscript is to introduce the novel concept and technology to readers, and we were careful to not make Cytelligen a primary emphasis in this manuscript because it’s not acceptable to turn a very serious research report to a commercial advertisement-like report. Our another study previously published on Oncotarget clearly indicated that Cytelligen was the vendor for related reagents, and that paper was cited several times in this AJP manuscript.

We appreciate that the incorrect image of Fig. 2B was indicated by the editor in our this uncorrected proof of manuscript. That image is indeed the lung cancer CTC previously published by us in CCA. When one of coauthors prepared the Figure 2B for AJP manuscript, she mistook the image from the lung cancer folder which was shared with our collaborator, instead of our colon cancer folder…But frankly, editor of AJP also accidentally made a  mistake by mistranslating our  frequently used key abbreviation of in situ PKC (in situ Phenotyping and Karyotyping of CTC) into a well known but irrelevant  in situ “Protein Kinase C” assay, which indeed bothered many readers who read our manuscript on the AJP website. We indicated it to the Journal, and I thought we might have the equal opportunity to the Journal to correct whatever mistakes made by either the Journal editor or us in this author’s uncorrected proof.

Regarding Fig. 3C, according to reviewer’s suggestion showing comparison of sensitivity of the technology described in this paper vs CellSearch, we provided additional data obtained in our previous studies performed on gastric cancer patients which was published by us on Oncotarget and cited several time in this AJP manuscript. The reason for keeping the particular CellSearch image (the reference image) identical to the one previously shown in Oncotarget is that we don’t want to mislead people to think that the entire new experiment other than our previous study was conducted.

…

Because the unique advantage of the technology described in this manuscript has been confirmed on  many cancer patient samples, and it helped doctors a lot to identify circulating tumor cells with higher sensitivity and specificity comparing to the conventional techniques (some of these results have been published on different  journals), we would like to share our experience with  colleagues, and would like to re-submit  our manuscript revised according to AJP editor’s indication.

Editor Kevin Roth also gave us a brief note from his side of the aisle:

The concerns relating to the recently withdrawn article were noted by the Editorial Office during the normal course of compiled issue review and were followed up per standard operating procedure. The Retraction relays the results of that inquiry.