Anil PottiThe University voluntarily canceled three clinical trials that drew from the research of Dr. Anil Potti, a cancer researcher whose research is currently under investigation.

The trials had previously been suspended. Duke researchers stopped admitting new patients to the two studies on lung cancer and one study on breast cancer July 18. The principal investigators made the decision to permanently end the trials following the Oct. 22 retraction of a key paper published by Potti and his mentor Dr. Joseph Nevins, said Dr. Michael Cuffe, vice president for medical affairs. Clinical oncologists and principal investigators are in the process of contacting about 100 patients who were enrolled in the three trials, Cuffe added.

“That request to retract represents a retraction of some of the foundational science for these trials.... It became less appropriate to move forward,” Cuffe said. “The question of patient safety is always a top priority—that’s something that has been looked at repeatedly by a lot of parties and continues to be examined.”

Duke and the Institute of Medicine are each conducting investigations into Potti’s research, which concerns the use of gene-based models in predicting patient response to chemotherapy drugs. Potti is currently on paid administrative leave.

Two biostatisticians from M.D. Anderson Cancer Center, Keith Baggerly and Kevin Coombes, first approached Duke with concerns regarding the research about a year ago. In light of the questions raised about the research, the three trials were suspended in October 2009, but the University restarted them in January after an internal review did not find problems meriting their termination.

“Do we wish that we had known this earlier? Of course,” Cuffe said. “I think everyone involved would have liked to understand whether that paper was correct or not, but that’s why we set up all trials with great caution.”

In late October, Nevins requested the retraction of a paper that influenced the two lung cancer trials which he published with Potti in the Journal of Clinical Oncology. Nevins is director for the Center for Applied Genomics and Technology and the Barbara Levine Professor of Breast Cancer Genomics. All three canceled trials, however, cited Potti’s paper, “Genomic signatures to guide the use of chemotherapeutics,” which was published in Nature Medicine, possibly opening the door to another retraction.

“It leads me to believe that either retraction or some definite clarification of [that paper] is underway,” Baggerly said. “Either it means they don’t trust the paper or they trust only bits of it, in which case they need to be very explicit about which bits of it they trust.”

A clinical trial that lists Potti as a sub-investigator is currently recruiting patients, according to the National Institutes of Health database. Although that trial also cites the Nature Medicine paper, the questions surrounding the research have not posed problems for the study because the paper is not central to the inquiry, Cuffe said.

Dr. Paul Kelly Marcom, the principal investigator for the breast cancer trial that has been canceled, said he is still in talks with the Department of Defense to determine what will be done with a $7 million grant that supported his work. In the study, Marcom and his team assigned patients in the experimental group to one of two standard chemotherapy options depending on the genetic signatures identified by Potti, while the control group was randomly assigned. Marcom said he believes the risk of harm to patients involved in the study is minimal.

“If the science is invalid, then the likelihood of harm to patients is extremely low,” Marcom said. “To say that the signatures would have assigned them to the wrong chemotherapy is to say they had some kind of validity.”

Marcom acknowledged that he is concerned that the controversy will undermine patients’ trust in clinical trials. But he does not believe that his study—which began in July 2008—was a worthless effort.

“It’s certainly not time or resources well spent to answer a question that’s not... valid,” he said. “I still wouldn’t say it’s a waste of time because science is a very start-andstop process. We’ve learned a great deal about how to conduct a trial like this, so that is still a very valuable experience.”

Marcom added that patients who have been notified about the cancellation of the trial have taken the news well so far.

“For the most part, they have understood that the goal of the trial was to see if these genomic signatures worked, but we were picking between standard therapies that they would have received anyway,” he said.

But he noted that he is concerned that the scandal will pose a setback for the field as a whole.

Baggerly said it may be challenging for scientists to reconsider their understanding of the role of genomics in chemotherapy, which was influenced by Potti’s publications.

“These papers have had major effects, and if they are wrong, it’s going to require a reset of quite a few things,” he said.

Yet Marcom said he is confident that his colleagues’ naturally doubtful posture will help them move forward.

“The papers may have influenced other groups’ research, but I don’t think it had a major steering effect,” he said. “ I think the papers raised people’s hopes that genomics would be able to help us with something that we had strived for for a long time—a way to individually guide chemotherapy research.... It’s been tried over the last 30 to 40 years in oncology. There’s always a great deal of skepticism when someone says we have a test because we all know how many things haven’t worked.”

But the scandal may raise doubts about the University, Baggerly said.

“There is a piece of this story that does question Duke’s actions as an institution,” he said. “This is a big deal. This is one of the bigger scientific controversies in this rather abstract field that I’ve seen.”