Grounds for dismissal

The Anil Potti affair took a new turn this Sunday when Joseph Nevins, Barbara Levine professor of cancer genomics and former mentor to Potti, stated publicly on the television program “60 Minutes” that it was “abundantly clear” that Potti had intentionally falsified his data. This statement, although an important and honest confessional on the part of Nevins, raises questions about the behavior of this professor who, until now, has played second fiddle to Potti in terms of his responsibility for the errors and has faced no repercussions from the University.

Nevins mentored Potti and collaborated as a senior author with Potti on the disputed research, allowing it to go to print and to be used as a basis for human clinical trials. The potential human cost of this error, in addition to Nevins’ status as a prominent, endowed professor affiliated with a prestigious University research group, means that he must be held accountable for these flaws. The magnitude of these errors is extremely large, and thus far, Potti seems to have taken the bulk of the repercussions.

First and foremost, we believe Duke should temporarily suspend Nevins’ professorship. At a minimum, Nevins’ behavior regarding this false research was negligent—at worst he was complicit in the falsification of results. Given the evidence available to the public on this matter, we consider that, as a senior author, Nevins should have conducted his own personal investigation into the validity of the results obtained as soon as they were called into question—that is, in 2009 when biostatisticians Keith Baggerly and Kevin Coombes of the University of Texas MD Anderson Cancer Center submitted their concerns about the research to the Annals of Applied Statistics.

Two reviews of the falsification of Potti’s research are currently in progress—an external review conducted by the Institute of Medicine continues and an internal review conducted by the University—to determine if research misconduct occurred. These reviews both remain confidential, in keeping with federal guidelines.

Both of these investigations should put Nevins on the hook and aim to determine his level of responsibility in the matter. As we see it, Nevins could have followed one of two paths.

The first possibility: After the initial questions raised by Baggerly and Coombes, Nevins should have scrutinized his and Potti’s work with great care to determine if errors had occurred. If he undertook this scrutiny but failed to produce any conclusions of falsification or did not endeavor to do so until later in 2010, when Potti’s qualifications were called into question, he was negligent and irresponsible as a researcher.

The second possibility: Nevins reviewed Potti’s data and discovered that it was falsified—a conclusion he managed to reach by the Feb. 12 “60 minutes” show. If he knew of the falsification before Potti’s suspension and allowed publications and clinical trials to go forward, he has committed an egregious violation of ethical standards.

Given either of these situations, we see grounds for the termination of Nevins’ position at the Duke University Medical Center. However, because of the secretive nature surrounding the investigation of this episode and the complicated workings of medical research, we defer to Duke’s institutional process in the hopes that the University will deliver sanctions to Nevins that reflect the severity of his involvement.