Commentary (Bonomi): First-Line Treatment for Advanced Non–Small-Cell Lung Cancer

November 1, 2005

Drs. Laskin and Sandler havedone an excellent job of summarizingthe results of chemotherapyin the treatment of stageIV non–small-cell lung cancer. Theyhave pointed out that a meta-analysispublished in 1995 showed that chemotherapyprovided a modest survivaladvantage compared to supportivecare alone. In addition, they have citedstudies that have shown the treatmentis cost-effective and that it relieveslung cancer–related symptoms.They have thoroughly discussed thefact that multiple phase III trials testingnewer chemotherapy doubletsshow slightly different toxicity profiles with virtually identical efficacy,and that giving more than four coursesof therapy with regimens that consistof three or more drugs does notprovide significant benefit.

Drs. Laskin and Sandler have done an excellent job of summarizing the results of chemotherapy in the treatment of stage IV non-small-cell lung cancer. They have pointed out that a meta-analysis published in 1995 showed that chemotherapy provided a modest survival advantage compared to supportive care alone. In addition, they have cited studies that have shown the treatment is cost-effective and that it relieves lung cancer-related symptoms. They have thoroughly discussed the fact that multiple phase III trials testing newer chemotherapy doublets show slightly different toxicity profiles with virtually identical efficacy, and that giving more than four courses of therapy with regimens that consist of three or more drugs does not provide significant benefit. Phase III Trials
The authors have also reviewed three different classes of biologic targeted therapies (anti-epidermal growth factor receptor [EGFR] tyrosine kinase inhibitors [TKIs], a rexinoid, and an antiangiogenesis monoclonal antibody). Each of these agents was tested in a phase III trial that compared chemotherapy to chemotherapy combined with the biologic targeted agent. The strategies for conducting these phase III trials were all different. The EGFR tyrosine kinase trials were initiated based on preclinical observations suggesting that chemotherapy plus EGFR tyrosine kinase inhibitors was at least additive, and possibly synergistic. Other considerations for combining chemotherapy with EGFR TKIs included the single-agent activity of the EGFR tyrosine kinase inhibitors and their nonoverlapping toxicity with chemotherapy. Two large phase III chemotherapy trials were conducted for gefitinib (Iressa) and two for erlotinib (Tarceva). Each of these trials showed no significant difference in survival. In fact, although the difference was not significant, the initial rate of death was slightly more rapid in the patients who received chemotherapy and the EGFR tyrosine kinase inhibitor. The second strategy for deciding to proceed with a phase III trial was based on the results of a single-arm phase II trial in which chemotherapy was combined with the rexinoid bexarotene (Targretin). Two phase III trials using different chemotherapy doublets with or without bexarotene were conducted. Each of these trials failed to show a significant survival difference despite the fact that the survival results were relatively encouraging in a small nonrandomized phase II trial. The third approach to make a go/no go decision for a phase III trial was based upon a relatively small randomized phase II study. Patients were randomly assigned to paclitaxel/ carboplatin alone vs different doses of bevacizumab (Avastin), 7.5 and 15 mg/kg, combined with the same chemotherapy regimen. There were trends for higher response rates and longer survival in patients treated with chemotherapy and the higher dose of bevacizumab compared to chemotherapy alone; there was a statistically significantly longer time to treatment progression (7.4 vs 4.2 months, P = .023) for patients treated with chemotherapy and the higher dose of bevacizumab. These observations resulted in a phase III trial that showed a significantly higher response rate, significantly longer progression-free survival, and significantly longer survival for patients who received chemotherapy and bevacizumab (15 mg/ kg) compared to chemotherapy alone. Trying to decide which regimens should be considered for testing in a phase III trial is extremely difficult, but extremely important, because considerable resources are utilized in conducting a phase III trial that has a relatively low chance of showing a positive result. Some critics are strongly opposed to randomized phase II trials, believing that they are too small to provide a go/no go signal. However, this methodology has several advantages. First, the inclusion of a reference treatment arm minimizes the risk of testing a new regimen in a group of patients with an extremely poor or extremely good prognosis. In the randomized phase II trial of chemotherapy with or without bevacizumab, each of the indicators of efficacy showed a trend for superior results; in the case of time to progression the difference was statistically significant. In most randomized phase II trials statistically significant differences will not be observed, but nevertheless, there will be some indication of whether the new regimen appears to have a positive effect. Having some statistical evaluation will certainly be worthwhile, and perhaps deciding that a P value of .1 or.2 in a randomized phase II study might be sufficient to go forward with a stage III trial might be a way to make a decision. Chemotherapy Plus EGFR Tyrosine Kinase Inhibitors
Drs. Laskin and Sandler, as well as Drs. Buter and Giaccone elsewhere in this issue, have discussed the disappointing results with chemotherapy and simultaneous EGFR TKIs. EGFR tyrosine kinase inhibitors reduce the percentage of tumor cells engaged in the cell cycle through an inhibitory effect on wild-type EGFR. They discuss the possibility that chemotherapy and EGFR tyrosine kinase inhibitors might be used intermittently or sequentially. Another strategy for combining chemotherapy and EGFR TKIs might involve selecting patients whose tumors might have greater sensitivity to EGFR TKIs. This group includes never-smokers, women, patients with adenocarcinoma and bronchoalveolar features, and Asians. Trials of chemotherapy combined with EGFR TKIs might produce different results in these groups. Other investigators have suggested that patients be selected for treatment with chemotherapy and EGFR TKIs based on the molecular profiles of their tumors. It has been shown that patients who have had good responses to EGFR tyrosine kinases have activating mutations in the ATP binding site of the EGFR protein. Also, increased gene copy number and downstream markers like phospho-AKT appear to identify tumors that are more sensitive to EGFR TKIs. It would be reasonable to conduct trials of chemotherapy combined with EGFR TKIs in patients whose tumors contained EGFR mutations, increased EGFR gene copy number, and possibly phosphorylated AKT. As Drs. Laskin and Sandler have indicated, additional studies are needed to determine the best way to integrate biologic targeted therapies with conventional chemotherapy and to combine different types of biologic therapeutic agents. Coordinated effort between basic scientists, clinicians, and molecular pathologists will be required to carry out future research.


Dr. Bonomi has served on the speakers bureaus of Eli Lilly, Genentech, and OSI Pharmaceuticals. He has received research support from Eli Lilly, Genentech, OSI, Bristol-Myers Squibb, Pfizer, AstraZeneca, Human Genome Science, ImClone, EMD Pharmaceuticals, Cell Therapeutics, Inc, and Celgene.