NCCN Guidelines Inform Decisions Across the Continuum of Cancer Care

July 2011, Vol 2, No 4

Philadelphia, PA—The National Com - prehensive Cancer Network (NCCN) is an alliance of academic cancer centers in the United States that seeks to optimize decision-making and policies for improving the delivery of appropriate and effective cancer care, said Al B. Benson III, MD, FACP, Associate Director for Clinical Investigations, Robert H. Lurie Comprehensive Can - cer Center and Northwestern Univer - sity, Chicago, immediate Past Presi - dent of the Association of Community Cancer Centers, and Past Chair of the Board of Directors of NCCN.

The 21-member NCCN represents clinical communities across the United States. As one of its main missions, the NCCN evaluates scientific information to inform and improve the decisionmaking between patients and physicians. The major way it achieves this mission is through the development of evidence-based guidelines.

One type of guideline is a process map of integrated interventions over time (Figure). The map addresses coordination of care and the continuum of care, offering references for each decision pathway. For example, there are more than 1500 decision pathways in the guideline for management of breast cancer. “The intent is to deal with reallife clinical situations” that are applicable to most patients, said Dr Benson.

“We do follow a 5% rule; if there’s a situation that occurs less than 5% of the time, it becomes very difficult to create a guideline. For much of what affects patients on a routine basis, we try to incorporate guidance in terms of decision-making.”

The second type of guidance is a systematic review of a single issue. The systematic review is a comprehensive review and analysis of the literature that can address a single decision point. “These aren’t mutually exclusive; systematic reviews are considered strongly in the NCCN guideline development,” said Dr Benson.

Evidence-Based Clinical Decision-Making
The various NCCN clinical guidelines are created by multidisciplinary panels for cancer screening, diagnosis, treatment, and supportive care. Each recommendation is supported by 1 of 4 categories of evidence ranging from category 1 (based on high level of evidence and uniform consensus) to category 3 (any level of evidence but major disagreement that the recommendation is appropriate). Most of the recommendations are supported by category 2A evidence (based on a lower level of evidence and uniform consensus that it constitutes appropriate care), usually data from clinical trials.

“The guidelines are not prescriptive” and are not meant to dictate decision- making but, rather, to inform decision-making. “We contend that something is wrong if a center is 100% concordant with guidelines. It’s impossible with the variability of clinical situations seen on a daily basis. However, they are applicable in cancer medicine to the majority of our patients,” Dr Benson said.

Each guidelines panel has on average 25 members. The evidence is reviewed continuously and guidelines are updated accordingly; new studies change the standard of care over time.

The entire process is transparent. “We spend a great deal of effort to minimize bias,” he said. About 900 clinicians serve on the review panels, which represent different geographical areas and specialties.

No industry or other interest group funds are used to support panel meetings, and no industry representatives are allowed at the meetings. Potential conflicts of interest are declared formally (verbal and written), and members can be removed from a panel permanently if their association with industry is too pervasive. Financial conflicts of interest for individuals are published on

The NCCN Compendium™ supports decision-making on appropriate use of drugs and biologics and is used for coverage policy by the Centers for Medicare & Medicaid Services and many private health plans.

“Our precertification folks use the compendium, because virtually all our chemotherapy regimens are precertified, even oral regimens,” Dr Benson said. “This process is a continuum, so if the guidelines are revised and it affects the compendium, the compendium is revised at the same time to reflect the guideline changes.”

Benchmarking Concordance with Guidelines

Another critical component of NCCN is the creation of an outcomes database in 1997. The initial purpose was to monitor and benchmark concordance of practice in NCCN institutions to the guidelines and to describe the patterns and outcomes of care in the member institutions.

All clinical interventions are collected, including specific regimens, lines of therapy, and oral agents. More than 270 separate data elements are collected.

The database also establishes a major research repository of data relevant to patient situations. As oncology moves to personalized medicine, data that reflect subsets of patients based on tumor biology will be important.

“When there are biological profiles such as predictive or prognostic markers available, they get integrated into the database and are integrated into the guidelines,” said Dr Benson.

The reports from the database inform discussion of individual recommendations at NCCN guidelines panel meetings. If a recommendation is not being followed, it can be for one of several reasons:


  • The recommendation is unclear, perhaps because of the variability of evidence
  • The recommendation is wrong
  • Doctors are practicing in advance of guidelines.

“When we send concordance data to our institutions, we expect a response,” he said. “We need to understand when there is a nonconcordance. So if you have a category 1 level of evidence, you would expect concordance to be around the 90% mark. If it’s not, we need to know why.”

The NCCN is interested in developing a compendium of predictive and prognostic testing. Molecular testing has the potential to identify:

  • Patients with more aggressive variants of disease.
  • Those whose disease may (or may not) respond to specific interventions.
  • Select systemic therapies based on the individual patient and the molecular profile.

“We are looking for better ways to utilize guidelines and decision-making,” said Dr Benson. “We’re calling it at this moment a therapeutic index. We would like to look at each recommendation in light of both safety and efficacy. In a potentially curative situation, more toxicity may be tolerated for good efficacy, whereas in a palliative situation, less toxicity is acceptable.”

Related Articles

Subscribe Today!

To sign up for our newsletter or print publications, please enter your contact information below.

I'd like to receive: