Inefficiencies in the NCI-funded cancer clinical trials endeavor prompted the NCI Clinical Trials and Translational Research Advisory Committee (CTAC) to establish the Operational Efficiency Working Group (OEWG). This body was charged with identifying barriers to speed and efficiency in protocol development and approval-and devising ways to eliminate them.

NCI approved the first phase of OEWG's work in March, and it went into effect this January. The 14 recommendations (see sidebar, OEWG Recommendations for Trial Activation on page 8) contain enforceable timelines for protocol development and implementation, including a "drop dead" date, at which time it will be terminated.

Draconian? Not really, said James H. Doroshow, MD, Director, NCI Division of Cancer Therapy Evaluation Program (CTEP) and Chair of the 63-member working group. "The report was unanimously endorsed, and everyone agreed on the timelines."

How did inefficiencies in the clinical trial system evolve? Gabriel Hortobagyi, MD, Professor of Medicine and Chair, Department of Breast Medical Oncology, MD Anderson Cancer Center, and Co-Chair of OEWG, said there are several major reasons.

First are the overwhelming regulations. "Thirty years ago, it took less than a month to get a trial going, from an investigator's idea to accrual of the first patient, and that included writing the protocol, IRB submission and approval, and calling a pharmaceutical company for sponsorship. Now all that takes a year or more-not including all the bureaucratic rigmarole of reporting adverse events, modifying the proposal, and filling out more forms."

And that's only when a single institution is doing the study. When it's a multisite affair, such as with a cooperative group, the time, expense, bureaucratic intervention, and aggravation are multiplied exponentially.

Dr. Doroshow agrees. "The minutiae of bureaucracy eat up so much time and so many resources that it has become impossible to control. When I first came to work at NCI in 1975, a protocol was 5 or 10 pages long. A consent form was a page or so. Now each file is inches thick," he said.

"The boilerplate is excessive, consent forms are so long and involved that a lawyer can barely understand them, let alone a patient," he continued. "Everything has more steps, involves more people, takes more time, and costs more money. There are too many rules and regulations."

Second is the health-care economy in general. There are many well-known reasons why everything costs more each year, and clinical trials are no exception.

Despite challenges, the federally funded clincal trials system has led to countless breakthroughs that have transformed cancer care. Removing bureaucratic impediments to trial activation will unleash the potential of the system to move quickly to investigate the latest ideas for new treatment options, Lawrence Wickerham, MD, Associate Chairman, National Surgical Adjuvant Breast and Bowel Project, Associate Professor of Human Oncology, Drexel University College of Medicine, Pittsburgh, and OEWG member, told The ASCO Post. "We at NSABP design trials that not only can meet accrual goals but can answer the scientific questions as well. We've had considerable success, and so have many other institutions and cooperative groups."

Time to Trial Activation and Accrual

Before improvements were made to the system, some trials took so long to get started that investigators lost interest before the first patient was recruited. Cooperative group phase III trials have a current median time to activation of 830 days-more than 2 years. It is 550 days for phase II trials and 200 days for investigator-initiated cancer center trials (see Fig. 1).

Many trials accrue only a small percentage of the required subjects, and some accrue none. In fact, the longer the process takes, the less likely full accrual will be reached. A recent study sponsored by CTEP found that only 18.5% of all trials achieved their minimum accrual goals, and 54.2% of 2,685 trials at 14 NCI comprehensive cancer centers did not accrue any patients at all. Only 26.6% accrued more than two patients.

The situation is not all that dire, however, said Dr. Wickerham. "Of 10 NSABP breast cancer trials activated between 2000 and 2007, only 1 was stopped because of inadequate accrual, and only 1 of 4 colorectal studies failed to meet its goal. Of the more than 34,000 patients in these trials, only 0.5% entered a study that was closed due to low accrual. We try to avoid enrolling any patient in a trial that will not be completed," he said.

Other trial groups have equally positive numbers; nevertheless, the clinical trial community must develop better ways to choose trials, get them going with minimal delay, accrue patients-and stop them when it is pointless to continue.

The Remedies

In addition to OEWG's 14 initiatives, the committee established targets for trial activation: 300 days for phase III, 210 days for phase II, and 90 days for investigator-initiated trials. Protocols that have been in the review pipeline for more than 24 months (for phase III) or 18 months (for phase II) will be terminated, retroactive to January 2011.
Cancer centers that conduct NCI-funded grants can improve internal management to increase efficiency (see sidebar, Recommendations to Increase Efficiency at NCI-funded Cancer Centers), but some are reluctant to participate in multisite trials because collaboration in such efforts is not recognized as a legitimate academic activity, nor is it seen as an important service. As a remedy, OEWG recommended the following measures:

• Including trial leadership, as well as accrual of patients in trials led by others, as a criterion for academic tenure and promotion
• Encouraging deans and department chairs to honor trial participation
• Enhancing training and mentorship for junior investigators as incentives to remain in academic medicine and clinical research (NCI will create new training programs to further this goal)

OEWG also recommended that CTEP and cooperative groups work with FDA, industry, and NCI to define standards about registration studies, establish consistency in protocol development, and resolve issues of trial objectives and design.

Collaborative Efforts

The OEWG initiatives will be "taken seriously by everyone involved," said Dr. Wickerham, "and I'm sure that people will put forth their best effort to make them work."

Yes, everyone agrees that clinical trials need to be faster and more efficient, said Dr. Hortobagyi. And if it were up to only those conducting them, it wouldn't be such a problem. But the system is enormous, unwieldy, and fractured.

There are too many government agencies, private industry members, politicians, and advocacy groups for it to work even reasonably smoothly. "It takes 800 steps to perform one trial, so we must change the systemic problems."

It's Starting to Work

Between April 1 and December 1, 2010, CTEP received 13 phase III proposals, 45 unsolicited phase I, I/II, or II proposals from cooperative groups, and a number of intramural ones. Of those approved, all are on schedule.
"People respond well to targets," said Dr. Doroshow. "This new system will get clinical trials moving again, increase transparency, and improve accrual."

NCI has instituted initiatives to help achieve OEWG goals. Two are especially helpful. During the course of getting a trial up and running, innumerable questions and problems arise. They have traditionally been resolved by sending e-mails, twiddling thumbs while they languish in in-boxes, waiting for replies, and then asking someone else the same or related questions. This takes weeks. A much better way is to conduct a conference call and hash things out in a matter of minutes. "These calls save enormous amounts of time and frustration," said Dr. Doroshow.

The other initiative is establishment of a website to track timelines. Steve Friedman, MHSA, Chief, Clinical Trials Operations and Informatics Branch, explained that the website provides access to NCI employees and investigators who need to follow a protocol's progress. All steps are listed in a timeline (how much time each took, how much time is left, etc), which can be compared to those of other protocols.

"It works like the FedEx tracking system: You know where your package is, when it's on the truck, and when the driver is going to ring your doorbell," he said. ■