Texas's $3 billion cancer research agency is in trouble again: Its peer review system appears to have come apart. The entire eight-member scientific review council of the Cancer Prevention and Research Institute of Texas (CPRIT) is stepping down, with most citing concerns about the integrity of the agency's peer review process. Many members of CPRIT's 100-strong roster of peer reviewers have begun to follow.
"Clearly there has been pressure at the board level to do things differently" than strictly by peer review, explains Nobel prize-winning biologist Phillip Sharp of the Massachusetts Institute of Technology (MIT) in Cambridge, who resigned as chair of the CPRIT scientific review council Monday. Despite changes made by CPRIT's leaders in response to criticism, such as hiring a compliance officer to make sure proper procedures are followed, "nothing has changed since last spring," when questions over CPRIT's review process first arose, Sharp says.
Approved by Texas voters in 2007, CPRIT expects to spend up to $3 billion from bond sales over 10 years on cancer research and prevention in the state. After disbursing $500 million in peer-reviewed research grants and recruiting many scientists to Texas, CPRIT made a decision that angered its top scientific advisers. The CPRIT board approved an "incubator" grant of up to $18 million to the University of Texas (UT) MD Anderson Cancer Center in Houston after a 3-week review that didn't involve scientific peer review.
The board also delayed a slate of grants, most intended for UT Southwestern Medical Center in Dallas, the former institution of CPRIT's chief scientific officer and Nobelist Alfred Gilman, according to Sharp and others, who also claim that CPRIT's board alleged the review rewarded favorites. Gilman announced plans to resign in protest.
Now CPRIT's scientific reviewers are following Gilman out the door. (He steps down today). Sharp writes in a letter to CPRIT that he is resigning for the same reasons as Gilman. The peer review system "was dishonored" because of how the MD Anderson incubator grant was reviewed and because grant awards were postponed over a "suspicion of favoritism," he writes. MIT's Tyler Jacks, also a member of the scientific review council, emphasizes the charges of bias in his letter: "These accusations [of favoritism], as well as the failure to mandate scientific review of so-called incubator grants during this period, served to undermine the careful work of my committee and the sanctity of the larger CPRIT scientific review process," he writes.
Council member Charles Sherr of St. Jude Children's Research Hospital in Memphis, Tennessee, adds a recent concern about the MD Anderson grant, which was pulled back for an additional round of review. New guidelines for incubators haven't yet been released, he notes. "I wonder whether some persons believe that forward movement in funding the [MD Anderson grant] would be facilitated by Dr. Gilman's departure and the possible elimination of other naysayers, myself included," he writes.
And council member William Kaelin of the Dana-Farber Cancer Institute in Boston cites a new problem with CPRIT's review process. He recently learned of at least two reviewers who rejected a commercialization project, then were asked by the CPRIT oversight committee to reconsider so their assessments "would be in harmony with" scores from commercialization reviewers that were "far more favorable." Kaelin adds: "In this environment I am not confident that scientific quality and rigor will triumph over grandiose promises and hucksterism."
Another council member, Clara Bloomfield of Ohio State University, Columbus, has not released her resignation letter. (Bloomfield was traveling and unavailable for comment.) The three remaining council members have indicated that they plan to resign over the peer review problems.
Many of the more than 100 scientists outside Texas who sit on an array of review panels overseen by the council are also quitting for similar reasons, according to several council members. In his resignation letter, William Hahn of Dana-Farber says he's "troubled" by the CPRIT request to reconsider scores for commercialization proposals without offering new information "to strengthen what were wholly naïve and underdeveloped applications. These actions make it clear that the CPRIT Oversight Committee has elected to disregard scientific review to pursue a different agenda."
"The whole thing is sorrowful," Gilman says. He adds, "The real problem is with the oversight committee," CPRIT's board, which consists of business leaders and others. Gilman says as new members come on, "Maybe the oversight committee can be dominated by people without self interests." If that happens, he says, "there would still be a very bright future for CPRIT."
Salvaging CPRIT will depend on who succeeds Gilman, Sharp says. "I think CPRIT is a wonderful statement by the state of Texas to advance cancer research and cancer care in the state and I wish it every luck and want it to continue. They're going to have to appoint someone to Al's position and reconstruct their review process."
CPRIT Executive Director William "Bill" Gimson was not available yesterday for comment because he was interviewing candidates to replace Gilman. But the agency released a statement downplaying the departures: "With the departure of Dr. Gilman, CPRIT is entering a new era. It is no surprise that some of the current reviewers have chosen to leave at this time." The institute has found "several exceptional candidates" for chief scientific officer whose first task will be to recruit "a full cadre of expert peer reviewers," the agency says.
The statement also defends CPRIT's decision to ask peer reviewers to reconsider their scores: "When there are divergent scores among peer reviewers, in fairness to the applicants, the process allows for further review or discussion." The statement says this allows scientific reviewers to take into account "new information" that arises during in-person reviews by commercialization reviewers.
*Correction 2:23 p.m., 12 October: CPRIT hired a compliance officer to make sure proper procedures are followed, not an ombudsman, as first reported.