Critics say Dr. Anil Potti was arguably misleading in describing the errors in his cancer research when applying for a medical license in South Carolina, where the practice he now works for is headquartered.
In the past year, the former Duke cancer researcher and his co-authors have retracted eight academic papers, citing data corruption issues and the inability of other experts to replicate the results. At the time of its publishing, the research was considered potentially groundbreaking because it claimed to demonstrate the ability to genetically predict which cancer patients would respond to specific types of chemotherapy.
When Potti applied to practice medicine in South Carolina in January 2011, he described the controversy surrounding his research in a letter to the South Carolina Board of Medical Examiners as involving “two academic rival groups,” one at Duke and another at the MD Anderson Cancer Center, which is located in Texas and first questioned the validity of Potti’s research.
Keith Baggerly, one of the MD Anderson biostatisticians who discovered the errors in the Potti papers, said this characterization is potentially misleading because it makes clear-cut problems with the research sound like an unresolved scuffle.
“It was being described as an academic debate when we thought that the main issues were ones of, ‘Were the labels correct?’ which is not really all that abstract of an issue, and it’s also an issue about which there isn’t legitimate debate,” Baggerly said. “It’s right or wrong. You got it right or you didn’t.... If one side is retracting papers, that’s pretty clearly an acknowledgment that that one was wrong.”
Baggerly and his collaborator and fellow biostatician, Kevin Coombes, began checking the research’s methods because doctors at MD Anderson found the results of one of the first papers promising and hoped to use the findings to treat their own patients once their validity had been confirmed.
In the January 2011 letter, Potti wrote that the “rivalry” dated back more than four years. This was around the time that Coombes and Baggerly first began checking the work following the publication of a study in Nature Medicine.
“We never viewed this as a rivalry,” Coombes said. “They published papers, we wanted to use their methods. We were trying to see if we could reproduce their results—we failed.”
Potti could not be reached for comment.
Patient care
Potti filed his application before beginning to work with Coastal Cancer Center, which has four locations in South Carolina and one in North Carolina. In the letter he attached, he wrote that the issues raised about his work were unrelated to the care that he provided to patients. Depending on how the events of the last year are interpreted, this claim is debatable.
Potti wrote that the controversy surrounding his work did not involve patients. In the letter, Potti references a misconduct review that Duke is conducting in order to determine whether the errors in the papers were made intentionally.
“It is important to note that during all of this, there were never any questions or concerns raised regarding the care I provided to my patients or the support offered to their families,” Potti wrote.
The division between his clinical record and genomics research may not be that clear.
Potti and his co-authors’ research led to the creation of three clinical trials at Duke intended to put the methods described in the papers into practice. The University terminated those trials—including one for which Potti was once the principal investigator—when it became clear that the findings they were based on were flawed.
Dr. Michael Cuffe, former Duke University Health System vice president for medical affairs, told The Chronicle shortly after Potti’s November 2010 resignation that the trials “should not have been done.”
Dr. Victor Dzau, chancellor for health affairs and president and CEO of Duke University Health System, confirmed Tuesday that the University has reached settlements with a number of the families of patients who were involved in those trials.
Baggerly said he disagrees with Potti’s comment that no issues were raised about his clinical work.
“Certainly near the end of [the trials] in 2010, many of the objections we were raising were based on concerns that patients might not be being treated appropriately,” Baggerly said. “So those were concerns associated with patient care, and those were ones we definitely had.”
Misconduct review ongoing
Duke’s evaluation of Potti’s research and the grants he obtained continues to move forward.
As Potti noted in his letter to the South Carolina Medical Board, this investigation, which began in the summer of 2010, is expected to take between two and three years to complete.
Dzau said he has kept a distance from the committee in order to allow it to function without his influence but that he received an update that the members are making good progress and are getting closer to making potential recommendations.
“Timeline-wise, I don’t know, but I understand they are making good progress, which means it might not be as long as one would think,” he said.
When the committee presents its findings to Dzau, he will then decide how Duke should proceed.
“I can pledge you an absolute commitment that we will do the right thing,” Dzau said.
Get The Chronicle straight to your inbox
Signup for our weekly newsletter. Cancel at any time.