Journal of the National Cancer Institute Advance Access originally published online on August 28, 2007
JNCI Journal of the National Cancer Institute 2007 99(17):1282-1289; doi:10.1093/jnci/djm149
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© Oxford University Press 2007.
NEWS |
NCI Director Opts Against Breast Cancer Prevention Trial In Favor of Biomarkers
A major breast cancer prevention trial has fallen victim to the squeeze of the National Cancer Institute's belt-tightening. NCI head John Niederhuber, M.D., decided in June that the institute would not fund the third clinical trial in a series of breast cancer prevention studies, even after it had been approved in an extensive review process.
The trial, known as STELLAR (Study to Evaluate Letrozole and Raloxifene), had been embroiled in controversy since Niederhuber put it on hold in January. His decision comes after months of closed-door discussions and an ad hoc review process, which proponents of the trial called arbitrary and capricious. Many in the cancer prevention community speculate that this ruling will shape the future of breast cancer prevention research. Whether it is for better or worse remains a matter of opinion.
"We’re surprised that the director chose to put a hold on this and develop a process outside of the traditional peer-review system," said Larry Wickerham, M.D., associate chairman of the National Surgical Adjuvant Breast and Bowel Project (NSABP), which had proposed the trial. "It's kind of devastating, the whole thing."
|
STELLAR, also known as P-4, would have compared the drugs raloxifene and letrozole for their ability to prevent breast cancer in healthy postmenopausal women at high risk of developing the disease. STELLAR was intended to be a follow-up to the Study of Tamoxifen and Raloxifene (STAR) trial, which found that raloxifene is equal to tamoxifen in reducing the risk of invasive breast cancer in postmenopausal women but had fewer serious side effects (see J Natl Cancer Inst 2006;98:733–5).
Changing the Process
The NSABP trial had gone through a grant review process that began in the fall of 2005. (The review process was similar to the one that the STAR trial had undergone.) Since then seven review committees and the Food and Drug Administration have endorsed it. Following a final 7–2 vote of approval by NCI's executive committee last January, Niederhuber, who was one of the nay votes, shelved the trial. "I thought it was in the best interest of NCI, best interest of the community and patients, that we at this time, faced with this kind of financial decision, seek outside advice," he told The Cancer Letter in March. (Niederhuber declined to be interviewed for this article and did not allow other NCI employees to comment.)
The outside advice came in the form of a committee selected by Niederhuber and Anna Barker, Ph.D., the other nay vote on the executive committee. In a meeting held in March, which was closed to the public, the committee members debated the merits and the shortcomings of the trial—after signing confidentiality agreements barring them from discussing the meeting with anyone who did not attend. The role of the committee was purely advisory. "The meeting wasn’t set up to produce consensus. It was set up to share information and have an unconstrained discussion of the pros and cons of the trial," said Allen Lichter, M.D., executive vice president and CEO of the American Society of Clinical Oncology, who attended the meeting.
But supporters of the trial questioned the composition of the committee. "One of the ... strange things [about the March meeting] was that the NCI's division of cancer prevention was prohibited from attending. The people at NCI with the greatest expertise in the field were not present at this meeting," Wickerham said. He was referring to Peter Greenwald, M.D., Dr.P.H., and Leslie Ford, M.D., the director and associate director of the division of cancer prevention, who were both supporters of the trial. This action has led some people to accuse Niederhuber of creating a review process that would confirm his opinion about the trial. An editorial in the June 30 issue of The Lancet said that Niederhuber's "action clearly undermines the NCI's review process" and called for an investigation into this decision.
"Had they done this differently, I think both sides would have felt—they may not have been happy with the outcome—but they would have at least felt that it was reached with some degree of objectivity," said Jerome Yates, M.D., national vice president for research at the American Cancer Society, who was also at the meeting.
Some STELLAR proponents have expressed fear that this process could slow the momentum of prevention research if investigators feel discouraged from designing prevention trials in the future. Whether this unusual review process was an exception or whether it will become the rule is unclear. "To circumvent the peer-review system should be of concern to anybody applying to the NCI for grants," Wickerham said. The NSABP has since appealed the cancellation in a letter to the director of the National Institutes of Health.
Pros and Cons
After considering the arguments for and against the trial, three members of the National Cancer Advisory Board who were asked to attend the closed meeting—Bruce Chabner, M.D.; Kenneth Cowan, M.D.; Ph.D.; and Diana Lopez, Ph.D.—presented a report at the June 14 advisory board meeting, which outlined their own concerns about the trial and was not intended to reflect any formal consensus by the attendees of the March meeting. "While P-4 is a well-designed, interesting, and relevant clinical trial, even if positive, it is unlikely to change the practice of preventative oncology. In view of the cost of the trial, the expiration of the patent for letrozole, and the need for expanded research on biomarkers for risk, we cannot offer strong endorsement of this trial as it was presented for current funding," Chabner, clinical director of the Massachusetts General Hospital Cancer Center, told the advisory board. Further discussion of the trial was vague as the advisory board members were told that specific grant applications could be discussed only during the closed session of the meeting.
Five days later, Niederhuber announced that the trial would not be funded. A letter was sent to the NSABP informing them that Niederhuber "has determined, after receiving considerable community input—both scientific and public—and after much deliberation, that the numerous scientific concerns about the P-4 trial are sufficiently formidable that the NCI will not commit to the funding of this particular trial."
Opponents of the trial cited several reasons against proceeding with the study, but it boiled down to three major faults. First, prevention clinical trials of this nature are long and expensive. Because only a small percentage of women in the trial will get breast cancer, nearly 13,000 women would have to be monitored for at least 5 years. While there was debate over the actual cost of the trial, estimates ranged from around $50 million to more than $100 million.
Many opponents wondered whether the results of the trial would be worth such a large investment. Even if the trial showed that letrozole could reduce breast cancer risk more than raloxifene, many believe that it would not change how physicians treat high-risk women. For one, critics point to tamoxifen, the first drug to be approved by the FDA for reducing breast cancer risk, which some studies suggest has not been widely adopted. They presume that letrozole may be unpopular as well. However, this position itself is controversial, as there are few data on tamoxifen use for prevention, and many feel that raloxifene would be more widely accepted because it causes fewer side effects (see J Natl Cancer Inst 2007;99:913–4). Raloxifene was recently recommended for use in these high-risk women by an FDA advisory committee, but it has not yet gotten full approval so there are no data on whether it will be more widely used.
Also, letrozole will be going off patent before the trial would have been over. So Novartis, the company that makes letrozole, would have little incentive to file for FDA approval or market the drug to women who could benefit from it, according to Chabner and others. This line of thinking assumes that unless the drug companies market the drug, high-risk women and their doctors will not use it. But, Wickerham says, if the drug turns out to reduce breast cancer risk by 70%–75%, as it is predicted to do, it's hard to believe that somebody wouldn’t try to get it approved.
Finally, many at Niederhuber's meeting said that biomarkers should be the highest priority for cancer prevention research, and before any more large prevention trials are conducted, researchers need to identify biomarkers that predict breast cancer risk. "In this new world of biomarkers ... I think we should focus a great deal of our efforts on figuring out who really is at higher risk of breast cancer before we subject healthy women to drugs that have serious side effects," said Fran Visco, president of the National Breast Cancer Coalition, which also opposed the two NSABP breast cancer prevention trials that preceded the STELLAR trial.
In Pursuit of Biomarkers
However, researchers have been searching for biomarkers of risk for more than a decade without much to show for it. "Aside from the Mitchell Gail's formula [a model that uses personal and family history to predict breast cancer risk], which is imprecise, I don’t think we’ve moved forward in risk identification a whole lot in the last 10 years, or 15 years for that matter," said Paul Goss, M.D., Ph.D., director of the breast cancer research center at Massachusetts General Hospital Cancer Center and principal investigator of a large breast cancer prevention clinical trial based in Canada.
How to move the field forward is at the heart of the controversy over the STELLAR trial. The debate ultimately comes down to a difference of opinion on the best approach to prevention research. One side believes that with technological advances and the hope of a better understanding of cancer genetics, the next decade holds more promise for biomarkers than the last. In the June 26 edition of the NCI Cancer Bulletin, Niederhuber wrote, "We are beginning to view our approach to prevention research differently, and it's an approach that NCI hopes will lead to more dramatic advances." He suggested an approach that moves prevention research out of the clinic and into the lab. "[It] is defined by the use of advanced tools and technologies—such as those employed in genomics, proteomics, and metabolomics—to pursue the molecular events associated with the mechanisms and early signs of cancer development." Identifying biomarkers of risk is a large part of this approach.
Prevention researchers on the other side say that biomarkers could be a few years or a few decades away. But in the meantime, clinical trials are necessary, they say, to determine which drugs reduce breast cancer risk. "When the biomarkers are discovered, we’re going to want to have as rich a therapeutic set of choices as possible, and this [STELLAR] trial would have helped enrich this knowledge base," Lichter said. In any case, large trials will be necessary because biomarkers will have to be validated in a clinical trial before they are adopted, and that testing could be incorporated into a drug prevention trial like STELLAR. Supporters of the trial agree that finding biomarkers is important, but not to the exclusion of prevention clinical trials. "To do the basic science and nothing else, while many women get a diagnosis of breast cancer while we’re waiting" is irresponsible, said Joan James, program coordinator of the STAR trial at Fox Chase Cancer Center in Philadelphia.
Both approaches—either looking for biomarkers before doing clinical trials or as part of the trials—are valid, Goss says. Both have their merits and pitfalls. It's impossible to know which team will win, but as the institute nears the end of its fifth year without a major budget increase, NCI is forced to pick a side. "We just don’t have the money to do both," Chabner said.
"I’m sure this was not an easy decision for John [Niederhuber], although it has all the appearances of an easy decision. He had to look at what he thought were the tradeoffs," Yates said. "If this was the late ‘90s when the NCI budget was doubled, this probably would never have happened."
Related Articles in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2006 98: 733-735.
J Natl Cancer Inst 2007 99: 913-914.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||