JPET peeves: Paper withdrawn after drug company won’t disclose chemical structure

A group of researchers at the drug company ChemoCentryx is withdrawing a 2012 paper in the Journal of Pharmacology and Experimental Therapeutics over failure to appropriately identify the molecule they describe in the article.

The withdrawal notice tells the story:

The following manuscript was published as a Fast Forward article on February 29, 2012:

Sullivan TJ, Dairaghi DJ, Krasinski A, Miao Z, Wang Y, Zhao BN, Baumgart T, Berahovich R, Ertl LS, Pennell A, Seitz L, Miao S, Ungashe S, Wei Z, Johnson D, Boring L, Tsou C-L, Charo IF, Bekker P, Schall TJ, and Jaen JC, Characterization of CCX140-B, an orally bioavailable antagonist of the CCR2 chemokine receptor, for the treatment of type 2 diabetes and associated complications. J Pharmacol Exp Ther jpet.111.190918; doi:10.1124/jpet.111.190918.

It was later found that the chemical identity of a compound cited in the article, CCX140-B, was not sufficiently disclosed. The authors are unable, at this time, to provide the chemical identity of CCX140-B in accordance with the editorial policies of The Journal of Pharmacology and Experimental Therapeutics. As a result, the authors have voluntarily withdrawn this manuscript from publication.

We apologize for any inconvenience this may cause JPET‘s readers.

We spoke with Richard Dodenhoff, journals director for the American Society for Pharmacology and Experimental Therapeutics, which publishes JPET, and he confirmed our suspicions:

We have a policy that chemical structures have to be revealed in a paper. You can’t just use a code name.

In this case, Dodenhoff said, the manuscript “slipped through” the review process and was published as a fast-tracked online article. Recognizing the mistake, the editor contacted the authors, who tried to get permission from the company to disclose the chemical compound but failed.

Without the identity of the compound,

you can’t tell anything from that name what that is; somebody trying to replicate the experiment couldn’t do it.

Which, in the world of drug development, is precisely the point.

We note that GlaxoSmithKline is ChemoCentryx’s development partner:

Our lead independent drug candidate, CCX140, which targets the chemokine receptor CCR2, successfully completed a Phase II clinical trial in type 2 diabetics and is currently in two Phase II clinical trials in patients with diabetic nephropathy, a form of kidney disease. The CCR2 receptor is believed to be of central importance to certain inflammatory diseases, such as diabetic nephropathy and Type 2 diabetes. CCX354, which targets the chemokine receptor CCR1, successfully completed a Phase II proof-of-concept clinical trial for the treatment of rheumatoid arthritis, or RA . This successful Phase II proof-of-concept clinical trial triggered GSK’s option rights under our collaboration agreement. GSK exercised its option to further develop and commercialize CCX354 in November of 2011 and has an exclusive right to initiate a Phase IIb clinical trial for CCX354 in RA. CCR1 plays a significant role in rheumatoid arthritis. CCX168, a complement 5a receptor antagonist, is currently in a Phase II proof-of-concept clinical trial for the treatment of anti-neutrophil cytoplasmic antibody, or ANCA-associated vasculitis, or AAV, and is subject to GSK’s option.

JPET also is retracting an unrelated paper by a group of Chinese researchers who evidently used boluses of the same data in three other publications.

According to the notice:

Re: Ping J, Gao A, Qin H, Wei X, Bai J, Liu L, Li X, Li R, Ao Y, and Wang H (2011) Indole-3-carbinol enhances the resolution of rat liver fibrosis and stimulates hepatic stellate cell apoptosis by blocking the inhibitor of κB kinase α/inhibitor of κB-α/nuclear factor-κB pathway. J Pharmacol Exp Ther 339:694–703; doi:10.1124/jpet.111.179820

The article referenced above contains significant data that have been published in multiple articles in other scientific journals. Portions of the data shown in Figures 1, 2, and 3 and in Table 1 from this paper represent either exact duplication or very close replication of similar experiments reported in at least three other published articles from the authors’ laboratory. The data were republished in JPET without citation to this other work. This is in direct violation of JPET‘s editorial policies and in contradiction to the assurances and statements provided by the authors in the authorship responsibility form that was signed by all of the authors and submitted with their manuscript.

One of those prior publications might be this paper from 2011 in Yao Xue Xue Bao.

Dodenhoff said a reader alerted the journal to the recycled data. The authors, he said,

 felt that it wasn’t duplication but couldn’t really offer any reason why it wasn’t.

Hat tip: Clare Francis

0 thoughts on “JPET peeves: Paper withdrawn after drug company won’t disclose chemical structure”

  1. As a scientist and reviewer, I hate it when people try to do this. Thanks are due to JPET for standing up against this nonsense. I’ve stood up to it as a reviewer, pointing out that the journal’s own guidelines don’t allow such publications and have been ignored multiple times. Every paper that refuses to disclose the chemical identity of any compound used should be rejected BEFORE review. Otherwise it is just wasting time.

  2. Nice job by JPET. Recently I asked for some researchers where they publsihed some agonists against one of the receptor which I am also interseted. Their reply is that “legal bandwitdth of their company” is not allowing the researcher to share this compund with any one. One researcher was from Glaxo and I forggot the second company which is based in california.

  3. In reply to aktfamily June 20, 2012 at 11:51 pm

    I wonder if

    “The actual immunogens are considered proprietary”

    is part of the same world as

    “legal bandwitdth of their company”.

    Antibodies are componds too. Often you find papers where the only reference to the antibodies will be a company name, and perhaps a dilution to use. When you go to look up the antibody described in the paper you find many sold by the company against the same antigen. Knowing the immunogen against which the antibody was raised would be helpful, if only to identify the antibody used in the paper. Ultimately some might like to make the antibody anew, in which case knowing the immungen would be essential, but that would be in the perfect world.

    PRS4627, Anti-SATB2 (ab1) antibody

    Sigma-Aldrich Technical Service Answer
    Scientist’s Response Via Email (Jessica) 06/14/2012 01:56 PM
    Hello Clare,

    Thank you for contacting Sigma-Aldrich Technical Service. It is possible that these antibodies are cross reacting with other related proteins or non-specifically. How similar are the SABT1 and SABT2? Product PRS4627, Anti-SATB2 (ab1) antibody is made with an immunogen near the N-terminus of the protein. Product PRS4629, Anti-SATB2 (ab2) antibody is made with an immunogen near the C-terminus. (The actual immunogens are considered proprietary.) Do SABT1/2 have high homology at the N or C-terminus? Have the authors checked SABT1 mRNA expression in SABT-/- mice? You can test the antibodies on cell lysates the express either SABT1/2 recombinantly, but lack endogenous SABT1/2.

    I hope that you find this information helpful. If you have further questions, please reply to this email.

    Sigma-Aldrich Technical Service
    Your Question Via Email (Entered by Jessica) 06/14/2012 01:56 PM


    clare francis
    Jun 12 (9 days ago)

    to cssorders

    1. PRS4627 Sigma Anti-SATB2 (ab1) antibody produced in rabbit

    2. PRS4629 Sigma Anti-SATB2 (ab2) antibody produced in rabbit

    I have some concerns about the specificity of the 2 anti-satb2
    antibodies (Ab1 and Ab2) used in the three publications:-
    1. Neuron. 2008 Feb 7;57(3):378-92.
    2. Eur J Neurosci. 2005 Feb;21(3):658-68.
    3. Am J Hum Genet. 2006 Oct;79(4):668-78.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.