The journal Nature is revising its policy on publishing animal experiments after a study it ran in 2011 received criticism because the authors allowed tumors to grow excessively large in mice. The paper reported that a compound isolated from a pepper plant killed cancer cells without harming healthy cells.
Yesterday, the journal published a correction to the study (the paper’s second), which noted that “some tumors on some of the animals exceeded the maximum size … permitted by the Institutional Animal Care and Use Committee.” The tumors were only supposed to grow to a maximum of 1.5 cubic centimeters, but some reached 7 cubic centimeters, according to David Vaux, a cell biologist at the Walter and Eliza Hall Institute of Medical Research in Melbourne, Australia, who first raised concerns about the paper in 2012. (Vaux spoke to Retraction Watch, which first reported the correction.)
In an editorial published yesterday, Nature calls the large tumors “a breach of experimental protocol,” one that could have caused the mice to “have experienced more pain and suffering than originally allowed for.” The journal also noted the lapse could have implications beyond the one study, saying that “cases such as this could provoke a justifiable backlash against animal research.”
Nature says it will now require authors to include the maximum tumor size allowed by its institutional animal-use committee, and to state that this size was not exceeded during the experiments. The journal does say, however, that it is not retracting the paper, and that the study remains “valid and useful.”
Science Executive Editor Monica Bradford says that during the manuscript submission process, authors must agree to a list of conditions, one of which is that experimental animals have been handled in accordance with the authors’ institutional guidelines. Furthermore, she says, if during the evaluation of a manuscript, a reviewer requests more information about the handling of experimental animals, Science’s editorial team will follow up with the authors to satisfy the reviewer query.