Duke Failed to Vet Researcher's CV

Just as an astronomical white count is not an entity unto itself but a marker of a serious underlying disorder, like leukemia, so a big lie on a CV is an indicator of a grave underlying problem, like sociopathy.

Faculty members at the Duke Institute for Genome Sciences and Policy are learning this lesson the hard way thanks to the failure of someone at Duke to perform the basic HR duty of vetting Anil Potti's curriculum vitae 7 years ago. Potti, who was hired in 2003 as a physician-researcher by Duke, falsely claimed that he was a Rhodes Scholar on scientific grant proposals, according to a recent expose by Paul Goldberg in The Cancer Letter. That's a big lie and one suggesting that other big lies are possible, if not probable. Taking this very cue, Goldberg then questioned the integrity of Potti's research at Duke and found that 2 biostatisticians at M. D. Anderson discovered "a series of errors," including mislabeling errors, in a seminal article by Potti and others.

Consequently the biggest victim of Duke's remote HR lapse: cancer patients who enrolled in clinical trials, which were based on Potti's questionable work. According to the NYT, these trials have now been suspended (after stuttering efforts by Duke officials to reopen them, reported Goldberg). News coverage can also be found at NPR's Shots blog.

For yucks or groans, I performed a quick PubMed search: "A Potti" is the coauthor of 48 articles that were published within the last 5 years. Potti's articles appeared in, for example, PNAS, JAMA, JCO, PLoS One, Lancet Oncology, Nature Medicine, and the NEJM.

Photo of Anil Potti from Duke Institute for Genome Sciences and Policy.

10/24/10 addendum: As the AP reported yesterday, Lancet Oncology editor David Collingridge relayed an "expression of concern" from 15 European investigators who were coauthors with Potti and 3 other Duke researchers on a 2007 article in the journal. (The article validated the use of gene signatures to predict the response of breast cancers to neoadjuvant [perioperative] chemotherapy.) After the damning report from biostatisticians at M. D. Anderson about Potti's alleged errors in another article, the Lancet Oncology coauthors repeatedly attempted to contact their Duke colleagues, but they were ignored, wrote Collingridge.

The editor also revealed that "a large group of scientists" wrote to NCI director Harold Varmus on July 19th, expressing their concerns about the validity of a) Potti's cancer-treatment prediction models and b) 3 clinical trials that were based on these prediction models. Collingridge expressed his own concerns about the Lancet Oncology article given recent developments surrounding Potti. The journal has contacted the Duke coauthors—Anil Potti, Chaitanya Acharya, Sayan Mukherjee, and Joseph Nevins—and awaits their responses.