Administrators commission internal, external investigations of Potti

Updated 2:09 p.m. July 30 to include information contained in a memo sent Friday by Dr. Victor Dzau to The School of Medicine.

The University will launch two investigations into cancer researcher Anil Potti and his work as a result of allegations that he embellished his resume.

Potti, an associate professor of medicine, is accused of falsely claiming to be a Rhodes Scholar on applications for federal funding and producing flawed research. Provost Peter Lange will lead a faculty misconduct investigation and review Potti’s credentials, said Dr. Victor Dzau, chancellor for health affairs and president and chief executive officer for the Duke University Health System.

“We want to be transparent, open and supportive—and objective, most importantly,” Dzau said.

“I’m working to get an external organization... [with] high-credibility to take it over and do all of it without any Duke involvement,” Dzau said. “In other words, we’ll provide everything they need to make it transparent and we will not be engaged in that review in order for the scientific body to be able to do it in an open and transparent and objective fashion.”

Friday, the University released a memo from Dzau addressed to the faculty of the School of Medicine. Dzau wrote that Duke will permanently divest all equity and potential royalties related to the science, which will be investigated by an external firm. The divestiture intends to eliminate a financial conflict of interest during the review. Dzau added that he will continue to provide updates as the situation warrants.

‘A personnel matter’

The University has placed Potti on administrative leave while Lange leads a review of all of Potti’s resumes and grant applications. But Lange said the specifics of the investigation cannot currently be disclosed, such as whether Potti claimed to be a Rhodes Scholar when applying to join the University.

“The process will take a bit longer than it would otherwise because we wanted to ensure that he did have legal representation,” Lange said. “And, in addition, yes, he will be offered the opportunity to present his case with the respect to the issues that we identify, if any.”

Lange said he will be assisted by a “variety of different administrative offices” at Duke but declined to name specific individuals involved in the investigation. He added that the review should be completed “promptly.”

Although Dzau is not directly involved in the investigation, he said it is “close to being completed.”

The University has not determined how it will announce the outcome of the investigation to the public, Lange said.

“This is a personnel matter and in general personnel matters are not discussed publicly, and yet there is great public interest in this matter and it has been publicly voiced already,” Lange said. “We’re going to have to work that out [and] we’re still in the process of working that through.”

The American Cancer Society has suspended payments on the $729,000 grant that Potti received for his research. In his application for the grant, Potti is listed as a “Rhodes Scholar (Australia).”

“We are extremely concerned about these allegations as we take accuracy in the scientific process very seriously,” wrote David Ringer, vice president for extramural grants for the American Cancer Society in a letter to the dean of the School of Medicine that was obtained by The Chronicle. “We ask that you provide us with an official curriculum vitae for Dr. Potti and an explanation of any variance it may have from the one included in his grant application (copy attached). We are also interested in what action you are taking to assure that the funded work has been done to the highest standards of science, accuracy and integrity.”

The Rhodes Scholarship brings students from around the world to study at the University of Oxford in England, according to the organization’s website. But the Rhodes Trust has no record of Potti ever having received a Rhodes Scholarship, according to The Cancer Letter, the newsletter that first alleged that Potti had made false claims on his funding applications.

The meaning of “(Australia)” remains unclear because some Rhodes Scholars come from Australia but they do not pursue their studies there.

Lange noted that President Richard Brodhead will oversee the review process but will not have an active role in reviewing Potti’s resumes and applications.

“To be clear, Duke immediately began an internal process to investigate the allegations as soon as we became aware of them,” Brodhead wrote in an e-mail Tuesday. “That process is still underway, and until it is completed we need to respect both Duke policies and legal requirements for privacy. We take matters like this very seriously and are moving as quickly and thoroughly as possible to understand the facts.”

Protecting patients

Last week, the University shut down three clinical trials that depend on Potti’s work.

Researchers stopped admitting new patients into programs that rely on Potti’s research—genomic technology used to match specific chemotherapy treatments with patients. Individuals whose treatment already relies on Potti’s work are not at risk because they are receiving chemotherapy as it is commonly administered, Dzau said. Potti’s study was not about establishing new methods of chemotherapy, but rather about using genetic markers in order to predict which of the currently used chemotherapy regiments is most effective for individuals, he added. That choice is currently made through “trial and error.”

“Therefore, they are not taking anything new that is putting them at risk,” Dzau said. “Stopping [the treatments] would be much worse than continuing them.”

This is not Duke’s first review of Potti’s work since his arrival in 2003. Administrators previously investigated research connected to Potti after biostatisticians Keith Baggerly and Kevin Coombes of MD Anderson Cancer Center found errors in the researcher’s work, according to The Cancer Letter. Although clinical trials were briefly suspended, Duke officials eventually allowed the trials to be restarted after completing the investigation. But after reviewing the reports of the University’s review, Baggerly and Coombs said the investigation’s documents do not adequately validate the study’s work, The Cancer Letter added.

More recently, a July 19 letter signed by 33 biostatisticians addressed to Harold Varmus, director for the National Cancer Institute, urged the organization to suspend the trials until a more rigorous investigation of Potti’s work is completed.

“Recently, published and peer-reviewed re-analyses of the work done by Potti and [Joseph] Nevins [director for the Center for Applied Genomics and Technology] revealed serious errors that questioned the validity of the prediction models upon which these ongoing clinical trials are based,” the letter said. “This led to a temporary suspension of the trials and a Duke-led review involving independent statistical experts. However, despite written statements from the external experts, who uniformly stated they were not given sufficient information to confirm the validity of the models, the trials have been reinitiated.”

The letter’s authors requested that more detailed information be released so that the findings can be validated. Only after a more rigorous investigation should the clinical trials be resumed, the letter added.

Paul Goldberg, Trinity ’81 who is covering the story for The Cancer Letter, said the situation is now an institutional problem for Duke and no longer the matter of a single dishonest researcher. He said he would like to know who was on the committee that originally reviewed Potti’s work, but that information has not been released.

"It could have been the problem of a rogue researcher based on silence—now it's the problem of the rogue researcher and the administrators that protected him,” Goldberg said. "This is a question of the administration. They are taking a problem that they could have solved, and turning it into a problem that is structural to the institution."

But when asked why the names of the first review committee could not be released, Dzau said the University is legally obligated to keep their identities private. Duke cooperated with the standard practices of the The Institutional Review Board, he added.

“When they do any review, it is what they call ‘peer-reviewed,’ which means that it has to be done with attorney-client privilege, and so under the process, that’s why we are not able to release the names,” Dzau said. “And no, we’re not trying to hide them, but it is a legal process, and it’s hard to explain that, of course, with sound bites.”