The editor of Clinical Medicine Insights: Oncology has issued an Expression of Concern over a 2008 paper by a group of authors in China after identifying “flaws” in one of the figures.
The article, “Exvivo experiments of human ovarian cancer ascites-derived exosomes presented by dendritic cells derived from umbilical cord blood for immunotherapy treatment,” purported to show that:
tumor-specific antigens present on exosomes can be presented by DCs [dendritic cells] derived from unrelated umbilical cord blood to induce tumor specific cytotoxicity and this may represent as a novel immunotherapy for ovarian cancer.
But according to William C. S. Cho, editor of the journal, there’s reason to doubt the conclusions. As the notice explains:
In 2012 the journal was informed of apparent flaws in Figure 2 of Qi-Ling Li, Ning Bu, Yue-Cheng Yu, Wei Hua and Xiao-Yan Xin. Exvivo experiments of human ovarian cancer ascites-derived exosomes presented by dendritic cells derived from umbilical cord blood for immunotherapy treatment. Clinical Medicine: Oncology. 2008;2;461–7. doi: 10.4137/CMO.S776. The journal commenced an investigation in accordance with the Committee on Publication Ethics (COPE) guidelines.
The journal was unable to complete a full investigation in accordance with the guidelines. In view of this, the journal must emphasise that no conclusions have been drawn on responsibility for the apparent flaws. In accordance with COPE guidelines the journal has published this expression of concern and recommends that readers interpret Figure 2 with caution.
The paper has yet to be cited, according to Thomson Scientific’s Web of Knowledge.
Although we don’t have enough data at this point, we seem to be covering more expressions of concern lately. Is it just our imagination, or are they become more common?
“Impressive” Photoshop (or Gimp, or whatever) job, that Figure 2. A band mirror-imaged and duplicated, plus bit of background the same in all images.
Mm… can’t see the duplication of bands. Although the malignant ascites derived exosome HSP90 band looks decidedly hand drawn.
compare the top row, second band, with the second row, second band. To me they are mirror images of each other. What hides it a bit is a spot of “noise” just above them, which is replicated in each and every blot…
If you enhance the contrast (using photoshop), it’s clear to the point of being funny that the background in all the panels of figure 2 is the exactly same. A nice dotted pattern appears very clearly, and it is identical in all the panels. Good stuff.
Nice find alterego……in addition see below:
Figure 3 FACS plots:
Compare the M1 plot profile of HLA-DR M1 on Day 3 with the M1 plot profile of CD86 day 12. Superimpose the two plots and move the CD86 plot slightly to the left.
Even with two very poor resolution profiles it can be seen they are identical.
Ok..so if you have an N=10 then why aren’t the author’s showing the “quantified” Western data vs. just a representative image. Duplication issues aside why wouldn’t a reviewer ask for this?