New Imaging, ChemoRT Recommendations From NCCN

HOLLYWOOD, Florida—Changes in the 2006 National Comprehensive Cancer Network (NCCN) guidelines for non-small-cell lung cancer (NSCLC) reflect fast-moving changes in imaging, in the availability of treatment agents, and in data supporting concurrent chemoradiotherapy (CRT) over sequential chemotherapy plus radiation. David S. Ettinger, MD, of Johns Hopkins Kimmel Comprehensive Cancer Center, and Mark G. Kris, MD, of Memorial Sloan-Kettering Cancer Center, discussed the guidelines changes at the NCCN 11th Annual Conference.

Dr. Ettinger outlined the main changes to the guidelines, which include:

Deletion of chest x-ray and deletion of lactate dehydrogenase (LDH) as part of the chemistry profile in initial evaluation of patients with NSCLC.

Separation of clinical stage I NSCLC into peripheral T1 and peripheral T2, with the latter added to stage I central and stage II disease.

Use of brain MRI for nonsquamous stage II NSCLC.

Adding adjuvant treatment for stage IA NSCLC in high-risk patients such as those with poorly differentiated tumor, vascular invasion, wedge resection, or minimal margins (category 2B recommendation).

Adding pulmonary function tests (PFTs) to pretreatment evaluations if PFTs were not done previously.

Adding MRI of the spine and thoracic inlet for T3 superior sulcus lesions abutting the spine or superior subclavian vessels.

Dropping surgery as recommended initial treatment for superior sulcus tumors of the lung.

Node-Positive Disease

Dr. Ettinger also discussed changes to the guidelines for node-positive NSCLC, which include:

For N2 positive nodes on mediastinal biopsy, replacing bone scan with PET scan. Dr. Ettinger said, "The use of CT scans to stage the mediastinum is associated with false-positive and false-negative rates of 10% to 20%. PET has been proposed to replace mediastinoscopy in staging the mediastinum. . . . When PET is positive, histologic confirmation of the nodal status is required. However, PET is not 100% accurate. When PET is negative, there is still a significant possibility that N2 or N3 disease is present."

Changing recommended initial treatment for N2 positive disease to definitive concurrent CRT.

Adding concurrent CRT as a treatment option for N2 resectable disease with positive margins.

Changing initial treatment for resectable stage IIIB (T4, N0-1) disease to concurrent CRT.

For stage IIIB (T1-3, N3), adding consolidation chemotherapy following concurrent CRT.

For stage IIIB (T4, N2-3), the initial recommended treatment is now concurrent CRT.

The surveillance recommendation is now spiral CT at 4 to 6 months after surgery, at 1 year after surgery, then annually. In addition, the definition of locoregional recurrence of NSCLC was changed to include mediastinal lymph node recurrence, with recommended treatment by concurrent CRT if the patient has not previously been treated with radiation, Dr. Ettinger said.

The guidelines working group also added the recommendation that for patients with symptomatic bony metastasis, clinicians consider adding bisphosphonate therapy.

Gefitinib (Iressa) was deleted from the recommendations for third-line therapy for patients with progressive disease and performance status 0-2 because of lack of data availability. Recommended second-line therapy is docetaxel (Taxotere) or pemetrexed (Alimta) or erlotinib (Tarceva). Recommended third-line therapy is erlotinib, according to Dr. Kris.

Bevacizumab (Avastin) with chemotherapy is now the preferred treatment for recurrent and metastatic NSCLC if the patient meets the treatment criteria. "Any chemotherapy can be used except gemcitabine [Gemzar]/carboplatin or any regimen that has more than a 10% risk of causing grade 4 thrombocytopenia," Dr. Ettinger said.

Measuring Response

In his presentation, Dr. Kris addressed what he called a "stealth" recommendation in the NSCLC guidelines, which advocates checking for response after a single course of therapy. "This is probably the most overlooked part of the guideline," he said. "The last thing we want to do is treat a patient with a potentially toxic regimen if the patient is not going to benefit. By checking earlier, we can potentially spare the patient who progresses further exposure to futile therapy. Also, if you have other alternatives, you can give them sooner. We used to argue about whether there was a second-line regimen, and now we are talking about third-line agents."

Dr. Kris noted that clinical trial data have shown a 17% increase in response rates from the addition of bevacizumab to chemotherapy. "There were more deaths on the combination regimen, mainly from bleeding, but I believe these deaths are balanced by the improvement in survival and response rates. But you have to make decisions for the individual patient," he said.

For adjuvant treatment of stage IB-IIIA NSCLC, Dr. Kris said that recent data from Winton et al (N Engl J Med 32:2589-2597, 2005) and Douillard et al (ASCO 2005, abstract 7013) "solidifies our recommendation that adjuvant vinorelbine [Navelbine]/cisplatin is the treatment of choice." With regard to neoadjuvant or induction chemotherapy for stage III NSCLC, Dr. Kris acknowledged that it produces "extraordinary rates of response, clearly better than historic comparisons," but cautioned that these findings are not based on randomized, controlled clinical trials.

During the question-and-answer period, the panel was asked about the use of carboplatin rather than cisplatin in adjuvant treatment of other than stage IB NSCLC.

"This is a data-free zone," Dr. Kris said. "We have only stage IB data and only for 4 years." Dr. Ettinger said that he does use carboplatin more widely in NSCLC cases. "It's hard to believe that it could be less effective than cisplatin," he commented.