Scientist Scrutinized

Cancer researcher may have misrepresented research results
May 20, 2014

This summer, allegations surfaced that Duke cancer researcher Anil Potti had misrepresented his academic credentials on research grant applications. In response to this and to questions about his research methods, the university took three immediate actions: a “complete review of the credentials and claims,” led by Provost Peter Lange, that Potti made about his background; a research misconduct inquiry “to be conducted as specified by Duke policies and federal law”; and efforts to facilitate “an independent, external investigation of the science in question by one of the country’s leading research bodies, to which Duke would supply any and all data and information, but would otherwise have no involvement.”

In addition, Duke suspended new enrollment in three clinical cancer studies with which Potti was associated.

According to a university statement issued in late August, “The first part of the investigation—the review of credentials—has now been completed. Issues of substantial concern were identified, and have resulted in corresponding sanctions. However, a final decision about Dr. Potti’s future status as a Duke employee and faculty member will also be informed by the results of the research misconduct inquiry and the independent external evaluation of the science. Until such time, he will remain on administrative leave from his research, teaching, and clinical responsibilities.”

Potti, an associate professor of medicine, is accused of claiming he was a Rhodes Scholar in multiple grant applications when, in fact, he was not.

Several biostatisticians and cancer researchers have called into question the validity of Potti’s findings on the relationship between genetics and cancer treatment. Potti had claimed that genetic analysis of tissue from early stage lung cancer tumors could be used to predict which treatments would work best to combat the disease.

Administrators investigated research connected to Potti last year after biostatisticians at the University of Texas MD Anderson Cancer Center, in a journal article, said they had found errors in his work and had not been able to reproduce his results. Although clinical trials were briefly suspended at that time, Duke officials allowed the trials to be restarted following a subsequent investigation.