The authors of a 2009 study in the Proceedings of the National Academy of Sciences (PNAS) have retracted the paper, which found a particular molecule could make breast tumors respond to a drug to which they’re not normally susceptible.
The paper — which has been cited five times, according to Thomson Scientific’s Web of Knowledge — was the subject of a fair amount of press coverage, although the molecule is not yet in clinical trials. In a Reuters story, lead author Caroline Ford said of the alleged tamoxifen-sensitizing compound, Foxy-5:
“It flips the switch basically,” Ford said in a telephone interview. “It makes breast cancer cells respond to tamoxifen in women who cannot be treated with the drug,” she added. “If you don’t have that molecule you can’t get tamoxifen because there is no target.”
According to the retraction notice, signed by all three of the study’s authors:
The authors wish to note the following: “During efforts to extend this work, we re-examined the laboratory records for all figures and found that the Excel files on which Fig. 4C was based contained serious calculation errors; the first author of the paper takes full responsibility for these inaccuracies. Considering the importance of this figure for the conclusions drawn, the authors hereby retract the work. We apologize for any inconvenience this may have caused.”
A review in Cell Cycle by the same three authors and another colleague relies heavily on the PNAS paper, and will also be retracted, Ford co-author Tommy Andersson tells Retraction Watch:
The retraction to the PNAS was sent by me on the 15th of November. Yes, you are right, our Cell Cycle review relies heavily on the retracted PNAS paper and therefore I sent a retraction request of this review to the Cell Cycle journal on the 17th of November. I have during these retraction processes been corresponding with both journals, and on the 26th of November I e-mailed the Cell Cycle journal the proof of the PNAS retraction, and on the 29th of November I got an email from the Cell Cycle journal that the retraction of our review was ok with them, and they would publish it as soon as possible.
I think the situation that led to the retraction is covered by the text, in that we, during the continuation of our work, re-examined the lab books and data files for the PNAS paper and in doing so we found serious miscalculations in a figure that is really central for the entire paper.
Of note: Ford and Andersson disclosed in the PNAS paper that they had filed for a patent on Foxy-5. The work is part of the basis of WntResearch, a company linked to Lund University, where Andersson works. We asked him what impact the retraction might have:
One should remember that the major part of my research, and the scope of the company, is focused on the potential use of Foxy-5 (and Box-5) in the development of anti-metastatic drugs. We have both in vitro and in vivo data for the anti-metastatic effect of Foxy-5. The possible effect of Wnt5a and Foxy-5 in combination with tamoxifen was only derived from in vitro experiments (and now these are retracted). Of course, the retraction will have an effect on my research activities in that I cannot use the findings in the PNAS paper as a basis for future studies.
Also of note: PNAS did not send out a press release about this retraction, which is an omission we’ve said before suggests a lack of transparency. That’s despite the fact that the paper was highlighted in their press materials when it was originally published, and was actually featured on the cover.
Shouldn’t that merit a release?
Please see a new post about Wnt Research’s initial public offering (IPO).
Interesting that this was a PNAS Direct Submission. Another failure of that system and another black mark on PNAS for not being actually peer reviewed?
Ryan, not to defend PNAS policies too strongly, I think you may be confusing ‘direct submission’ with ‘contributed articles’. Direct submission (fka track II) is how they get a majority of submissions, which go through peer review much like at any journal. Authors can suggest NAS members to edit (and referees), but are not guaranteed that they’ll get them.
Most folks bristle at NAS member ‘contributed’ articles (fka Track I) and they still seem to have something like what used to be called Track III in which an NAS member can shepherd a friend’s article through … I think they call them ‘features’ now.
So, I understand the confusion, but think you’re picking at the wrong thing. It’s not clear how peer review would catch serious miscalculations (which a cynic like me might take to mean data manipulation)
See PNAS’s author policies. They’re a trip. http://www.pnas.org/site/misc/iforc.shtml#submission
There is a company, Wnt research, founded largely based on these findings that recently had an emission of new stocks….