Cancer Clinical Trials — A Chronic but Curable Crisis

Published at www.nejm.org June 16, 2010 (10.1056/NEJMp1005843)
Cancer Clinical Trials — A Chronic but Curable Crisis
Robert C. Young, M.D.

In April, the Institute of Medicine (IOM) released a comprehensive review of the clinical trials program of the National Cancer Institute (NCI), in which it concluded that "the system for conducting clinical trials in the United States is approaching a state of crisis."1 The report, which was commissioned by the National Cancer Institute, drew the attention of the New York Times, which published an editorial on April 25 highlighting cumbersome procedures, excessive bureaucracy, poor coordination, and failure to complete 40% of the trials.2 On May 5, during hearings of the Senate Appropriations Committee, which holds jurisdiction over funding for the NIH, committee member Tom Harkin (D-IA) cited the report and the editorial in expressing concern about the funding of the NCI's clinical trials network and the impending crisis.

Created 55 years ago and modified intermittently since then, the NCI's clinical trials program is by far the largest trials network in the country. The program includes 10 cooperative groups, more than 3100 institutions, and 14,000 investigators and enrolls more than 25,000 patients in trials each year. Cancer-related clinical trials are also performed by industry, individual institutions and cancer centers, international cooperative groups, and the Department of Veterans Affairs.

The program has been central to many important clinical advances in cancer in the past 50 years. Its trials have defined effective treatments for childhood cancers, improving survival rates from less than 10% to more than 80%. Its landmark trials in early breast cancer defined standard surgery for localized breast cancer. Other group-led trials have established the standard of care for many advanced cancers, demonstrated that adjuvant therapy improves survival in many cancers, and shown that certain drugs have preventive effects in patients at high risk for colon adenomas, breast cancer, and prostate cancer. With such a stream of accomplishments, the clinical trials program is crucial to continued progress in cancer treatment.

Unfortunately, the program is bloated, cumbersome, inefficient, slow-paced, overmanaged, and expensive. Layers of bureaucracy at participating institutions, the NCI, and other government agencies slow the review process, so it takes about 3 years, on average, from concept to protocol approval. Trials remain uncompleted because of poor enrollment, lack of interest, excessive administrative burdens, and inadequate funding. These abandoned trials are at best a terrible waste of resources and at worst unethical, since they subject patients to risks without producing definitive results. The program currently costs $145 million per year, but the budget has been flat in recent years and since 1999 has actually declined 20% when adjusted for inflation. Clinical investigators see this level of funding as insufficient to sustain a clinical trials infrastructure, support enrollment, and ensure proper follow-up.

These problems are not new. The last three NCI directors have struggled to identify strategies to address them. Each has commissioned an external review to provide recommendations, the latest being the IOM assessment.

The IOM analysis recommends changes in four areas: the speed and efficiency of the design, launch, and conduct of trials; innovation in science and trial design; trial prioritization, selection, support, and completion; and incentives for patient and physician participation. The IOM identifies the deficiencies and proposes solutions — some easy to implement, some much more difficult.

Current protocol initiation is plagued by inefficiencies, overlapping oversight, and lack of standardized collaboration between the research groups and industry. Many steps in the review process are redundant, and there are many oversight committees with different objectives and responsibilities. In one of its periodic modifications, the NCI created scientific steering committees to better define the trials with the highest priority. These committees, largely comprising physician–scientists who are not participating in the trials, have added complexity to concepts and slowed approval without improving approval rates. Having many layers of well-meaning reviewers decide what constitutes the best trial can result in trials that are of less interest to both patients and physician investigators. The rapid pace of change in science may render a delayed trial obsolete.

To address these problems, the IOM suggests consolidating the cooperative groups' "back-office" operations, including administrative functions, management of data and standardized forms, and patient registration. The institute also proposes consolidating the cooperative groups themselves, using a process of peer review. Although such consolidation would reduce the system's complexity, it would also limit the number of competing ideas that can be explored. The IOM points to the successful consolidation of pediatric cooperative groups, but most of the progress in childhood cancers mentioned above took place before that consolidation. Other recommendations include a Department of Health and Human Services effort to streamline and harmonize government oversight and regulation of cancer clinical trials — a laudable goal that will be difficult to achieve in the short run. One important IOM suggestion is that the Food and Drug Administration establish and implement a coordinated cancer program, a step that would benefit all cancer clinical trials, not just those of cooperative groups.

In its second set of recommendations, the IOM correctly notes that the clinical trial structure cannot remain frozen. Validated biomarkers, high-quality annotated biorepositories, and maintenance of tissue banks will be necessary for the personalized medicine of the 21st century. With many established standard-of-care therapies already in place and more than 800 potential therapies in the queue, innovative trial designs are mandatory (see table) — more clinical trials are required than could possibly be conducted using conventional trial designs. The cooperative groups should also take on comparative-effectiveness studies, as they have in the past. Long-term outcomes, quality of life, and survivorship issues are less likely to be addressed by industry-designed trials focused on drug or device approval. An earlier IOM study4 identified 100 important comparative-effectiveness studies needed in research on cancer.
open here to see the full-text:
http://content.nejm.org/cgi/content/full/NEJMp1005843