Maurie Markman, MD
Scientific misconduct in the oncology research field is a most difficult and concerning subject that warrants discussion, particularly in light of the apparent resolution of a distressing series of events that occurred at Duke University, where a laboratory- based researcher once considered a rising star was found to have extensively misrepresented data regarding predictive genomic signatures involving patients with cancer.1-3
The preclinical and clinical efforts of this individual, and the overall research program in which he was a major participant, continued for a considerable period of time after very serious concerns were raised both publicly by a highly regarded group of bioinformaticians at another nationally recognized cancer institution, and apparently far more quietly by some within the internal Duke University research establishment. 2,3
The major issues noted by the external reviewers included not only the published data but also the former Duke researcher’s disquieting responses to legitimate questions regarding the work.
Remarkably, the rogue activity of this individual ultimately was not halted as a result of any objectively mandated internal assessment by Duke University nor any external review by governmental funding or regulatory agencies based on these clearly articulated concerns, but rather because it was learned at some point in this distressing affair from a third party that this individual had made stunningly inaccurate statements in previous grant applications about his credentials.2-4
Unfortunately, the Office of Research Integrity at the US Department of Health & Human Services has “taken final action” in this highly distressing episode of scientific misconduct—which included the treatment of patients with cancer on a clinical trial based on false research data—without punishing any individual involved in required research oversight at Duke University.1,2
The researcher entered into a voluntary agreement in which he consented to extra oversight for a 5-year period in the event he engages in publicly supported research without admitting or denying the agency’s findings.1 However, in my opinion, the greatest concern with the profoundly distressing episode that has unfolded at Duke University in recent years is whether it will truly serve as a learning experience that would help prevent future misconduct—events that have the potential to result in serious harm to highly vulnerable patients with cancer.
There have been numerous reports of very disturbing scientific misconduct occurring within the laboratories of basic and translational laboratory scientists and clinical investigators during the past several decades involving multiple organizations and in many countries. In fact, a significant majority of retractions of previously published scientific papers have been found to be the result of scientific misconduct.5 A number of factors have been highlighted that may contribute to this disquieting state of affairs, the relevance of which will likely vary tremendously based on the specific circumstances of an individual or group such as the unique scientific environment, country, or current position of the individual.6
1. Strong incentive for laboratory or clinical results/ outcomes to be positive or even “very positive” to ensure publication in the highest-impact, peer-reviewed journal possible. Recognition that “negative” results or less than very positive outcomes would be more likely to lead to either no publication or ultimate publication in a far less prestigious venue.
2. An increasing number of biomedical students, PhD candidates, and junior nontenured faculty seeking initial jobs or more stable employment (“achieve tenure”), which in the academic world requires demonstrated success in obtaining grants and in publishing in the highest-impact scientific journals. Notably, this issue becomes more pronounced when organizations—including the government— substantially increase opportunities for students to pursue careers in biomedical research by providing essential funding for what is labeled “training,” but there is no meaningful change in the number of jobs available for these individuals in either the public or private sectors following the completion of this training. For example, from 2000 to 2012, the number of postdocs in the United States increased by 150% while there has been no overall increase in the number of full-time faculty positions available over this time period.7
3. Well-defined and recognized limited timelines for trainees or junior faculty to attain the required publications and grant funding to permit them to progress within the system (the academic “up or out” phenomenon).