NCCN Unveils Preliminary Guidelines

FORT LAUDERDALE, Fla--The National Comprehensive Cancer Network(NCCN) unveiled preliminary clinical practice guidelines for eightdiseases at its first annual conference (see photograph below).The network now includes 14 institutions nationwide with the additionof its newest member, Roswell Park Cancer Institute.

In an interview, Stanford University's Robert Carlson, MD, chairof the breast cancer panel, said that "there was remarkableagreement among our committee in nearly all aspects of the guidelines."He cited two primary areas of controversy.

In the committee's initial draft, routine chest x-rays were notincluded in the follow-up of patients after adjuvant chemotherapy."However, there was a ground swell from people external tothe committee that made it clear to us that the guidelines hadto include routine surveillance chest x-rays."

The other topic of greatest discussion was whether to includehigh-dose intensive chemotherapy with bone marrow/stem cell supportin the guidelines as a routine treatment, an issue that pittedwhat Dr. Carlson called "disease-oriented specialists"against "modality-oriented specialists."

"The committee felt, without exception, that high-dose intensivetherapy was investigational," he said. Consequently, in afootnote to the breast cancer guidelines, the committee statedthat "based on current evidence and NCCN expertise, dose-intensivechemotherapy is not appropriate outside the confines of an appropriatelydesigned, peer-reviewed prospective clinical trial." Peerreview in this context includes formal NCI review or NCCN institutionalsci,entific review committees.

Dr. Carlson called the footnote "a positive statement signifyingthat we need more information regarding high-dose intense therapy;we certainly encourage participation in clinical trials."

At the meeting, Dr. Carlson asked the audience for a show of handsas to whether dose-intensive chemotherapy should be restrictedto clinical trials or included in the guidelines for use in high-riskpatients (10 or more positive nodes) or patients with respondingmetastatic disease in first relapse.

Dr. Carlson expressed surprise at the degree of unanimity. Ineach case, only 4 or 5 hands went up in favor of inclusion inthe guidelines.

David S. Ettinger, MD, of the Johns Hopkins Oncology Center, whoheaded the non-small-cell lung cancer committee, told OncologyNews International that the most contentious point with his panelwas whether to use surgery or radiotherapy for M1 disease (solitarybrain or adrenal metastasis with a resectable lung lesion).

"We took an aggressive approach in favor of resection ofthe metastatic lesion (plus or minus radiation for brain lesions),with the lung cancer treated according to the TNM classification"he said.

Dr. Ettinger added that in patients with T1N0 or T2N0 diseasewith a single lesion in the brain, resection can produce 5-yearsurvival rates of up to 10%.

Panel member Mark Kris, MD, of Memorial Sloan-Kettering, pointedout that a large number of adrenal lesions in otherwise resectablepatients are not due to cancer, and a pathological diagnosis ofthe adrenal lesion needs to be made "so as not to consigna curable patient to an incurable category."

The lung cancer committee also dealt with the question of whento evaluate response to chemotherapy. The guidelines recommendevaluation with CT scan after the first course. "We feltthat 25% of patients would have progressive disease after thefirst cycle," Dr. Ettinger said, "and since the costof chemotherapy is significantly higher than the cost of a CTscan, there would a significant saving from evaluating after onecycle."

Moreover, he said, if the patient is not responding, and has agood performance status, he or she is then eligible to receiveother chemotherapy that may be effective and, if second-line treatmentalso fails and performance status remains good, to proceed toa phase I/II trial.

Dr. Ettinger also mentioned a controversy that arose over nodalresection and mapping in N1 or N2 disease. "We first suggestedsampling a minimum of three nodes; Dr. Kris suggested four. ThenRobert Ginsberg, of Sloan-Kettering, said that all nodes shouldbe sampled."

Dr. Ettinger noted, however, that any recommendation the committeeultimately makes "whether three, four, or all nodes, is morethan we're getting now from some of our colleagues doing lungresections in the community."

Both Drs. Carlson and Ettinger stressed that the underlying assumptionof all the treatment recommendations is that, whenever possible,patients should be considered for enrollment in clinical trials.

[Editors' note: Look for additional reports on the NCCN guidelinesin next month's issue.]