The history of the National Comprehensive Cancer Network is a saga of the obstacles encountered by America’s cancer centers in their efforts to demonstrate that they provide better outcomes that justify higher prices.

Also it’s a story of successes, many of them unplanned and unforeseen.

Two decades ago—at the onset of the discussions that led to formation of NCCN—the top executives of comprehensive cancer centers had a lot to worry about.

The Clinton administration was ushering in an era of managed care, threatening to eliminate all but the most efficient of health care providers.

Of course, many academic oncologists believed that they provided the best care available, but they lacked data to demonstrate this.

Survival was at stake. Centers made money on patient care and spent it on research. Therefore, the loss of patient care dollars would extinguish research.

“Capitation” and “oncology carve-outs” were the buzzwords heard at oncology meetings in the early 1990s. In that dystopia, insurers would pay providers to assume all cancer risk within their portfolios.

Now, turn the clock two decades forward.

In 2012, the buzzword du jour is “comparative effectiveness,” and cancer centers are feeling even more threatened, this time by declining federal funding and the continuing onslaught of lower-cost providers of cancer care.

While the Clinton-era threat of capitation hasn’t materialized, the Obama administration’s health care reform plan includes provisions that make cancer centers feel less than secure.

The Affordable Care Act requires the Department of Health and Human Services to “report quality measures of process, structure, outcome, patients’ perspective on care, efficiency, and costs of care” at hospitals that were exempt from Prospective Payment System reimbursement limits. (The PPS exemption was crafted specifically to protect the centers from the impact of the 1982 reimbursement structure based on “Diagnosis-Related Groups” of medical services.)

Alas, one thing hasn’t changed in that time: there is still no way to demonstrate conclusively that cancer centers do better for their patients than other providers.

“It’s a story of people trying to guess what health care reform is going to do and how oncology is going to evolve—and they are all guesses,” said Robert Young, former director of the Fox Chase Cancer Center and chair of the NCCN board from 1997 to 2001. “Sometimes you guess right—sometimes you don’t.”

Audio files of The Cancer Letter’s interviews with Young and the group’s first executive director, Catherine Harvey, can be found at:

The Threat of Capitation

“When the Clinton health plan seemed likely to come down in 1992-1993, a group of [cancer center leaders] got together and said, ‘Holy shit, they are going to start sending our patients out to all these community hospitals because it’s cheaper,’” said Joseph Simone, who had been physician-in-chief at Memorial Sloan-Kettering Cancer Center during the early 1990s, and would become the NCCN’s first chairman of the board.

The network was formed by five centers: MSKCC, MD Anderson Cancer Center, Fred Hutchinson Cancer Center, Fox Chase Cancer Center and City of Hope.

Simone remembers discussions where his colleagues stated flatly that centers provided better outcomes than local doctors and community hospitals.

“We may be better than they are, but what evidence do you have?” Simone asked at one of the initial meetings.

“Of course, there was no evidence,” Simone recalled. “That started us thinking where we could get to the point where we could get that kind of information.”

Conversations at preliminary meetings quickly turned to the question of measuring quality—and stayed there.

“The question of higher quality is a debatable one, of which there is—as in most things in oncology—no evidence,” Young recalled. “That was the centerpiece of the issue that got everybody talking about how we could band together to (1) protect ourselves, and (2) begin to demonstrate that institutions of our type actually did deliver a high-quality product that justified the increasing cost.”

The questions were urgent: How would centers remain competitive?

What kinds of patients would continue to seek care at centers?

Could centers combine to launch a business-generating effort? Would this offshoot organization be a for-profit or a non-profit?

The group first turned to entrepreneur Michael Goldberg, whose specialty was applying emerging informatics technology to oncology.

“They hired Michael to create an organization, and then, as a result, to have products that the market would want,” Harvey said. Goldberg’s South San Francisco-based company, Axion Health Care Inc., was also running a drug distribution business called Oncology Therapeutics Network.

OTN would later be sold to Bristol-Myers Squibb. Later still, Goldberg would run an unsuccessful venture to set up a network of oncology practices organized around information systems and clinical pathways for managing cancer care.

As he explored possible directions for NCCN, Goldberg recruited Catherine Harvey, then an administrator at the Medical University of South Carolina, to run day-to-day operations of the emerging organization. Harvey came on board in February 1994.

In the beginning, five centers had committed to join and sponsor the organization. Their initial contributions paid Harvey’s salary, which was floated through Axion.

Working from her home in Charleston, S.C., Harvey helped put together a proposal that was presented at the 1994 annual meeting of the American Society of Clinical Oncology.

At the meeting, there were 17 centers in the room, Harvey recalled. The centers would have to join before the end of the year, contributing $100,000 each.

“There were three things that we proposed to the group, where there would be minimal competitive angst among us,” Harvey said.

Their three goals were contracting, clinical guidelines and outcomes measurement.

“The one that was driving it more than the clinical guidelines at that point was contracting,” Harvey said.

“It was to create some kind of a centers-of-excellence model, where people like Blue Cross/Blue Shield would want to have the ability to send people to these centers. That was particularly a driver for people like Bob Day and Peggy Means out at Fred Hutch. They wanted bone marrow transplants to be a major focus of that.”

There was also talk about second opinion services to be provided by cancer centers.

“They realized pretty quickly that if you are going to run either a second opinion service or be a center of excellence, you’ve got to have guidelines that prove that your care is superior and that it’s following a pathway,” Harvey said.

The third goal was outcomes measurement.

“After that meeting, they sent me out on a road trip, and I went and visited all the centers that were at the table,” Harvey said. “In the beginning they were trying to figure out why they needed each other, and then the other challenge was getting their administration to ante up enough money to get us going.”

Meanwhile, the group that was exploring the contracting possibilities ran into resistance from payers as well as internal resistance.

“There was a series of steps we went through and a lot of discussion about whether this was a business-generating effort or not, and we were split about half-and-half in those meetings,” Simone recalled.

While some wanted to approach national and international companies and managed-care organizations, others were skeptical.

“Some of us were not confident that that was worthwhile,” Simone said. “We didn’t fully realize that most insurance is regional and local. It’s not national. So no one could commit their national patients to you.”

Young was similarly skeptical. “I didn’t think that market would develop in any major way,” he said. “And to a certain extent it never has. There is certainly more flying or more referral to centers than there was then, but not a lot.

“The reality is, 85 percent of patients get cared for on an outpatient basis, and they get cared for near home. And that hasn’t changed a lot.”

There was discussion of the form the offshoot organization would take, and the idea that it could be a for-profit company ran into opposition.

“These were all not-for-profit organizations,” Simone said. “There was major concern that they would have to be part of a business venture that they had no control over. And it would be a for-profit.

“People looked at it at institutions and said, ‘No, it’s not going to work. It’s too complicated, and we lose control of our own destiny.’ So that’s why we ended up doing guidelines.”

The pursuit of contracting continued.

“We danced with that for over a year,” Harvey said. By December 1994, the organization had 13 members, and therefore a budget of $1.3 million or so. The committees were working in all three areas. “That was my main job, to keep those committees moving,” Harvey said.

The guidelines committee accomplished more than others. To Simone, focusing on standards was simple common sense.

“We have no product. What do we have to sell?” he said, recapping one of these early discussions. “We ought to have some standards. I didn’t call them guidelines; I just called them standards. We have all these smart people around. We ought to develop standards.

“Ultimately, we decided that we would be better off doing that.”

A focus on guidelines meant that there would be no reason for Goldberg to stay in the job of CEO of the new organization. Also, some centers said that his business ventures—setting up the Oncology Therapeutics Network—constituted conflicts of interest.

He resigned as CEO in November 1994.

“It wasn’t an easy, comfortable fit for people having an outside person that had never worked in the cancer centers being the CEO,” Harvey said. “That was hard for some of the leadership to reconcile. And he had his other business that he was running at the same time. He was growing a big pharmaceutical distribution business.

“There was also, on both sides, a question early on of what were we really going to be. Were we going to be another not-for-profit on steroids that never had income or just a vehicle to facilitate things like contracting? Or were going to have a business strategy of our own?”

Goldberg was replaced by Bruce Ross, a retired Bristol-Myers Squibb executive, who lived in the Philadelphia area. Ross ran NCCN from an office that moved to Fox Chase Cancer Center. Harvey stayed on, continuing to work from her home in Charleston. Ross took the job on condition that he would stay long enough to get the organization going. He came on as a favor to Simone and Young, making it clear that he wouldn’t stay in the job long.

During his time at BMS, Ross developed strong connections in the academia. He had coordinated some key projects, including the development of the drug Taxol.

However, working as the CEO of an organization funded by the centers was a new experience for Ross.

“It was a shock for me to go from the industry into academia, because in academia everything is done by consensus,” Ross said. “It’s hard to make decisions. These were all academic cancer centers. I answered to a board, and I was used to making decisions. I would hear people out and say, okay, I’ve heard you out, and this is the way we are going to do it.

“That’s not the way it works in these organizations.”

Ross ran the organization as it focused on guideline-making.

The Guidelines Format Emerges

Disputation is an endeavor in which academics excel.

“The idea was that we used free horsepower—all the docs in these institutions were asked to serve on committees,” Simone recalled. “We didn’t have to pay them. We had to pay the travel, but we didn’t have to pay them. And we could build a reputation for the organization based on having established standards.”

In March 1995, the organization presented guidelines that covered cancers of the breast, colon, prostate and the lung. Also covered were acute leukemia and pediatric cancers.

Altogether, these diseases accounted for 51 percent of all cancers.

Using templates designed primarily by Rodger Winn, chairman of the NCCN Guidelines Steering Committee and an oncologist at MD Anderson at the time, the organization came up with guidelines that were both brief and easy to change.

For example, the most controversial of the guidelines at the time—the first version of the guideline for breast cancer—fit on just 25 pages. The guideline addressed standard—as opposed to investigational—treatments. This allowed the breast cancer committee to avoid a fight over bone marrow transplantation.

“We assumed that investigational care will always take precedence over the guidelines,” Robert Carlson, a Stanford University oncologist who headed the breast cancer panel, said at the time (The Cancer Letter, March 15, 1996; posted at

The committee didn’t recognize bone marrow transplants as standard practice, and a fight was averted.

The unveiling of the guidelines established the centers as the leading voice in oncology, Simone said.

“It was a hit,” he said. The first meeting of NCCN sent a message to the field: the cancer centers are setting the standards for quality care. The centers would do well by doing good.

“You would send your complex cases there, ones that you couldn’t handle,” Simone said.

“The biggest reason we thought it would help us is that it would give us an opportunity to look at what we were doing,” said Young, who first served as vice chairman of the NCCN board and chairman of the outcomes committee. “We were pretty confident that, measured by our own guidelines of what constitutes state of the art care, we would do reasonably well.

The guidelines were unique in medicine.

“You have guidelines written by state-of-the-art experts in their fields, covering all cancers, updated on the monthly basis,” Young said. “And that doesn’t exist anywhere, and probably couldn’t, with any other group. And it certainly doesn’t exist in any other disease.”

It would be impossible to accomplish this breadth with evidence-based guidelines. Such guidelines “tended to be developed, published, and ended up on shelves gathering dust,” Young said. “And they lacked a number of things we believed were critical. One, they ought to be guidelines that could be read in a very short period of time. Simple guidelines. They ought to build in the capacity for encouraging patients to enter clinical trials.

“And two, they needed to be comprehensive, because it wasn’t very helpful to have one guideline on breast cancer treatment and one on diffuse lymphoma and one on colon cancer—and nothing else.

“I think oncologists realize that if you just used evidence-based medicine, you wouldn’t have guidelines on much of anything. A few in adjuvant therapy, and breast cancer, and colon cancer, and that’s about it.

“We said, ‘Look, we want evidence-based medicine as much as we can use it, we want criteria for the strength of particular recommendations, but we want recommendations of what a group of experts believe is state of the art care that day.’

“The other thing that we built into it early on is the capacity to provide feedback to the institutions. If they found themselves either as a group or individual doctors within their institutions lacking adherence to our own agreed-upon guidelines, they would be encouraged either to conform or explain why the guideline is inappropriate and thereby encourage having it be changed.

“It was a continuous quality improvement strategy.”

McGivney Tapped to Lead NCCN

Ross left NCCN at the end of 1996. This was an orderly changing of the guard.

Ross, who took part in choosing his successor, believes McGivney was an excellent candidate.

“We interviewed a whole bunch of people,” Ross said to The Cancer Letter. “At the time, he was with Aetna, and he knew a lot about the health care industry. He seemed to be quite well qualified.

“I think, for the most part, he did extremely well. I think what did him in was his inability to keep making money for the organization. These are tough times for cancer centers.”

At the time he was hired, McGivney was the vice president for clinical and coverage policy at Aetna Health Plans.

Soon after taking the job, McGivney told The Cancer Letter that academic cancer centers and managed care companies have a common interest: developing the capability to make decisions based on outcomes data.

“Both communities, in terms of decision-making, are interested in basing their decisions on outcomes data, and that’s a common theme we need to sit down and talk about more,” he said at the time.

“There is a potential for a better relationship, if you can get by some of the standoffishness and doubts about the other side.”

The profile of McGivney, which appeared in the Jan. 24, 1997, issue of The Cancer Letter, is posted at

A timeline provided by NCCN follows:

NCCN Clinical Practice Guidelines in Oncology:

  • NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) are the recognized standard for clinical policy in oncology in both the academic and community settings and significantly influence appropriate practice patterns and prescribing behavior. The evidence-based NCCN Guidelines are developed and updated by 47 individual panels, composed of over 900 multidisciplinary clinicians and oncology researchers from the 21 NCCN Member Institutions geographically dispersed across the country. Panel members possess in-depth knowledge of the biomedical literature and awareness of, if not actual leadership and/or participation in, the trials that provide the evidence for the NCCN Guidelines. NCCN annually surveys clinicians to determine how they use the guidelines.
  • The first seven NCCN Guidelines, covering breast, lung, ovarian, prostate, colon, rectal, and pediatric cancers were released in 1996. NCCN Guidelines covering melanoma, sarcomas, lymphomas, bladder, central nervous system, head and neck, and pancreatic cancer were first developed in 1997. The first NCCN supportive care guidelines covering use of anti-emetics also were introduced in 1997. In 1998, the NCCN Guidelines for Hodgkin’s disease, endometrial and cervical cancers, esophageal and stomach cancers, testicular and renal cancers, myeloma, chronic myelogenous leukemia, and neutropenic sepsis were first published. NCCN also introduced its first screening guidelines in 1998 covering breast, prostate, and colon cancers screening, including the use of genetic testing.
  • As of 2012, NCCN has produced 55 clinical practice guidelines covering 97% of all malignancies affecting individuals with cancer, including separate guidelines for prevention and screening as well as supportive care for patients. There are 136 separate algorithms covered within the guidelines mentioned above.
  • The quantity and sophistication of information in the NCCN Guidelines have become much greater over the years. The guidelines have moved from generic recommendations to detailed recommendations for the multidisciplinary management of cancer. In 2003, Principles of Chemotherapy, Radiation, and Surgery were first included in the guidelines, and this information has become widespread throughout the program. The NCCN Guidelines have become increasingly rich with discussions of pathology, biomarker testing, and imaging. NCCN now also offers NCCN Chemotherapy Order Templates (NCCN Templates) for regimens recommended in the guidelines to improve the safe and effective use of drugs and biologics in cancer care.
  • NCCN has been a leader in identifying arrangements involving NCCN Guidelines Panel Members who have a significant financial or fiduciary interest in an outside entity where such arrangements may represent a source of conflict or an appearance of conflict in the participation in the development of the NCCN Guidelines. The NCCN disclosure policy requires disclosure of external relationships and recusal of NCCN Guidelines Panel Members with significant conflicting interests so that the integrity of the NCCN Guidelines is not compromised or diminished by conflicts or by the perception of conflicts. NCCN began collecting this information from NCCN Guidelines Panel Members, NCCN Guidelines staff, and the NCCN management team in 2006, and the results are posted on
  • In response to the Centers for Medicare and Medicaid Services (CMS) requirements for compendia to have a publicly transparent process for evaluating therapies and for identifying potential conflicts of interest, NCCN has become more transparent with respect to how decisions are made and what data support those decisions. NCCN has published the criteria used for evaluating NCCN Guideline recommendations, the names of the NCCN Guidelines Panel Members who participated in the development of the guideline recommendations, submissions of data to the panel, the votes for recommendations, and the panel’s disclosures.
  • A number of the NCCN Guidelines are available in Chinese, Japanese, Korean, Portuguese, and Spanish. Approximately 47% registered users on who access the NCCN Guidelines, NCCN Drugs & Biologics Compendium (NCCN Compendium), and NCCN Templates reside outside the United States.
  • NCCN’s continuing medical education programs began with the NCCN Annual Conference in February 1996 and expanded in 2002 to include NCCN Regional Guidelines Symposia. These half-day programs aimed to make physicians aware of the data upon which the guidelines were based. The new NCCN Guidelines Update Webinar Series is designed to quickly communicate significant updates to the NCCN Guidelines. During these disease-specific webinars, NCCN Guidelines Panel Members discuss the issues considered by the panels that resulted in changes to the NCCN Guidelines, new data is compared with existing standards of care, and significant studies that supported modification of the recommendations are reviewed. All of these programs inform clinicians of which data NCCN Guidelines Panels regarded as persuasive and why, increasing the transparency of decision-making.

NCCN Drugs & Biologics Compendium:

  • The NCCN Drugs & Biologics Compendium, derived directly from the NCCN Clinical Practice Guidelines in Oncology, provides authoritative, scientifically derived information designed to support decision-making about the appropriate use of drugs and biologics in patients with cancer.
  • On August 9, 2005, NCCN petitioned the CMS Administrator for CMS to officially recognize and use the NCCN Compendium as one mandated reference among others for coverage decisions about the appropriate use of drugs and biologics in cancer care. CMS convened a meeting on March 30, 2006 of the Medicare Evidence Development and Coverage Advisory Committee that identified 13 basic characteristics of a good compendium and then voted giving the NCCN Compendium the best score on each characteristic.
  • The Secretary of Health and Human Services has the authority to add compendia to Part B. NCCN petitioned the Secretary on October 19, 2006 to do so with supporting letters from many major national organizations (e.g., American Medical Association, American Society of Clinical Oncology, Association of Community Cancer Centers, Oncology Nursing Society, National Patient Advocate Foundation, Cancer Leadership Council).
  • Private payors such as the Blue Cross Blue Shield plans, UnitedHealthcare, Aetna, Cigna, and Humana utilize the NCCN Guidelines to make coverage determinations for drugs and biologicals used in an anti-cancer chemotherapeutic regimen. On January 16, 2008, UnitedHealthcare announced that it would be the first payor to base its benefit coverage for chemotherapy drugs used in outpatient settings on the NCCN Compendium, effective March 15, 2008.
  • On June 5, 2008, CMS recognized the NCCN Compendium as a mandated reference for establishment of coverage policy and coverage decisions regarding
    the use of drugs and biologics in cancer care. The NCCN Compendium will be utilized by CMS for national coverage determinations and by intermediaries
    and carriers for locoregional determinations. The major application will be in determinations about coverage for the use of drugs and biologics beyond
    the FDA-approved indication. The use of drugs and biologics for indications in cancer care beyond the FDA-approved label is a common, appropriate, and important mechanism to provide the most effective care to patients.

NCCN Guidelines for Patients:

  • Since September 23, 2010, NCCN has developed eight NCCN Guidelines for Patients, consumer-friendly translations of the NCCN Guidelines, on breast cancer, chronic myelogenous leukemia, malignant pleural mesothelioma, melanoma, multiple myeloma, non-small cell lung cancer, ovarian cancer, and prostate cancer. NCCN Guidelines for Patients are now available in a flip-book format on and Many of these guidelines are also available in print copy.

Collaboration with National Business Group on Health:

  • On December 16, 2010, the National Business Group on Health (NBGH) announced its collaboration with NCCN on a three-year project to develop An Employer’s Guide to Cancer Treatment & Prevention – resources and tools in benefit design to be utilized for employers to address cancer care in the workplace. The first project deliverable, the Quick Reference Guide and Assessment Tool, was released in April 2011. The second resource, the Benefit Design and Assessment Tool, was released in July 2011. Additional tools include the Request for Proposal and Proposal Scoring Tool, the Vendor Contracting and Administration Tool, and the Vendor and Program Evaluation Tool.

NCCN Oncology Research Program:

  • The NCCN Oncology Research Program (ORP) is organized to obtain funding to support scientifically meritorious research projects at NCCN Member Institutions. The NCCN ORP has received more than $34 million in research grants to support investigator-initiated trials. These trials explore new venues of clinical investigation that answer important scientific questions. Studies evaluate innovative combinations and sequencing regimens of drugs, drug resistance, mechanisms of action of specific agents or explore extended uses for specific agents.

NCCN Oncology Outcomes Database:

  • The NCCN Oncology Outcomes Database is a network-based data collection, reporting, and analytic system that describes the patterns and outcomes of care delivered in the management of patients with cancer. The concept for the Database was established in 1996, and the operation of the first database in breast cancer was initiated in July 1997. With the NCCN Oncology Outcomes Database, NCCN seeks to implement the NCCN Clinical Practice Guidelines in Oncology through performance measurement. Presently, the NCCN Oncology Outcomes Database has five active database components: breast, colon/rectal, non-small cell lung, and ovarian cancers as well as non-Hodgkin’s lymphomas. The Database follows approximately 85,000 patients with data elements collected on each patient in areas of socio-demographics, clinical interventions, and clinical and non-clinical outcomes. The data is high-quality and research-worthy as onsite audits of data occur within three months of a site joining the database and on an annual basis thereafter.