The review by Oxnard and Miller provides a thoughtful update on the use of the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) gefitinib(Drug information on gefitinib) (Iressa) and erlotinib (Tarceva) as front-line therapy in patients with non–small-cell lung cancer (NSCLC). It was pure coincidence that EGFR TKIs selected because of their high selective inhibitory activity on wild-type EGFR, turned out to be even more selective for several EGFR-mutated variants that drive tumorigenesis in a small subgroup of patients with NSCLC. The evidence that oral EGFR TKIs result in significantly higher tumor response rates and prolonged progression-free survival (PFS) in EGFR-mutated tumors compared with chemotherapy is rapidly accumulating. A number of confirmatory trials are ongoing. Based on the results of the Iressa Pan-Asia Study (IPASS) and other supporting phase II and phase III trials (see below), I agree with the authors that EGFR TKI therapy is a reasonable initial therapy for patients with advanced EGFR-mutated NSCLC.
When EGFR Status Is Unknown
Many practical questions of interest to the physicians who treat lung cancer patients remain, some of them listed by the authors. One that is not specifically addressed is what is the best treatment for patients with a high probability of having an EGFR-mutated tumor but in whom the test cannot be performed, which still constitute the majority of patients. At the 13th World Conference on Lung Cancer, Lee et al presented the preliminary analysis of the First-line Single-agent Iressa vs Gemcitabine(Drug information on gemcitabine) and cisplatin(Drug information on cisplatin) trial in Never-smokers with Adenocarcinoma of the Lung (First-SIGNAL), a randomized phase III trial of gefitinib vs gemcitabine (Gemzar)/cisplatin as first-line therapy for "never-smokers" with advanced adenocarcinoma of the lung in Korean patients. Because this study was not mentioned by the authors, I will summarize its results briefly.
A total of 309 patients were entered, and EGFR mutation analysis was performed in 96 (31%). The primary endpoint, survival, was not met (hazard ratio [HR] = 1.003), possibly because 80% of patients on the chemotherapy arm received gefitinib after progressing on chemotherapy. PFS favored the gefitinib arm, with an HR of 0.61 (0.74 in IPASS). Response rates were similar in both arms, 53% in the gefitinib arm and 46% in the chemotherapy arm (43% and 32% in the IPASS trial, respectively). In the subgroup of EGFR-mutated tumors, the response rate was twofold higher in the gefitinib arm (84% vs 37%, 71% vs 47% in the IPASS trial), and the PFS also favored the gefitinib arm (HR = 0.61, 0.48 in IPASS), whereas there was no difference in survival. In the subgroup of EGFR wild-type tumors, the response rate was twofold higher in the chemotherapy arm (52% vs 26%, 71% vs 1% in IPASS). The PFS also favored the chemotherapy arm (HR = 1.61, 2.85 in IPASS), and there was no difference in survival.
Although the results of this study confirm the benefit of gefitinib as front-line therapy in patients with EGFR-mutated tumors as shown by the IPASS trial, they are not so clear in demonstrating a detriment with the use of front-line gefitinib in the EGFR wild-type tumors, as the response rate was a respectable 26% and there was no difference in survival compared to patients receiving front-line chemotherapy. As a result, the conclusions of the authors were that (1) gefitinib should be the treatment of choice in patients with known EGFR mutations, (2) chemotherapy should be the treatment of choice in patients with known wildtype EGFR tumors, and (3) gefitinib is a reasonable option in patients with tumors of unknown EGFR status. The first two conclusions are in agreement with the review by Oxnard and Miller and widely accepted. The third conclusion is more controversial but not necessarily unreasonable. My personal position is that initial therapy with an EGFR TKI is reasonable for patients with a high probability of having an EGFR-mutated tumor provided they are closely followed and switched to chemotherapy as soon as progression is documented. This option is particularly attractive in elderly patients and/or those with a poor performance status.
Concurrent Combinations vs Sequential Therapy
A second issue is whether patients with EGFR-mutated tumors should be treated with combinations of chemotherapy and TKIs rather than using a sequential approach. These tumors may have a higher response to chemotherapy alone than wild-type tumors. Although synergism between EGFR TKIs and chemotherapy could not be proved in unselected patients with NSCLC, the combinations were well tolerated. In addition, as noted by the authors, the subgroup analysis of the Tarceva Responses in Conjunction with Carboplatin(Drug information on carboplatin) and Paclitaxel(Drug information on paclitaxel) (TRIBUTE) trial showed a potential benefit of combining erlotinib with carboplatin/paclitaxel in never-smokers. The hypothesis generated by this observation was that erlotinib would synergize with chemotherapy in tumors where its main effect is not cellcycle arrest but apoptosis. However, the observation within the same trial (not mentioned by the authors) that no such benefit was observed in the subgroup of patients with documented EGFR mutations—which is somewhat surprising—does not support the hypothesis. As a result, my position is that such combinations should not be used at the present time until evidence of enhanced efficacy becomes available.
The Cancer and Leukemia Group B (CALGB) 30406 trial was undertaken to compare erlotinib alone with erlotinib plus carboplatin/paclitaxel in never- smokers. Results will be presented at the 2010 meeting of the American Society of Clinical Oncology (ASCO) and will apparently include subset analysis of patients with EGFR mutations. This analysis should help answer the question of the role of these combinations in patients with EGFR-mutated NSCLC. Unfortunately, a chemotherapy-only arm was not included in this trial.
Second-Line Therapy
A third practical issue is how to treat patients with EGFR-mutated tumors upon progression after initial TKI therapy. A widely common practice in the second-line setting has been to continue TKI therapy beyond progression in patients (not necessarily with documented EGFR mutations) who have benefited from these agents, are slow progressors, and do not develop definite new lesions while on therapy. This practice is based on evidence that accelerated growth may be observed in slow progressors after discontinuation of erlotinib.
In the front-line setting, switching to chemotherapy after failure of TKI therapy rather than continuing TKI therapy beyond progression is probably the most appropriate strategy, as these tumors are chemosensitive. However, progression after chemotherapy will be the rule, and therapy upon progression with front-line EGFR TKI therapy followed by second-line chemotherapy should be guided by identification of the signaling pathways that are activated and drive tumorigenesis in resistant tumors. This includes activation of other pathways like c-MET or persistent EGFR pathway activation through second EGFR mutations. Second EGFR mutations that have a markedly enhanced affinity for ATP while conserving similar affinity for the currently used TKIs have been reported in as many as 50% of patients. Whether these are de novo mutations that occur after therapy is initiated or represent selection of preexistent clones with those mutations is unclear, but it would seem that the latter possibility is more likely.
Specific and irreversible inhibitors of these receptors carrying second mutations are in clinical development. If it turns out that tumors contain multiple clones with different EGFR mutations that are being selected out by specific inhibitors, effective treatment will require alternating inhibitors with a different spectrum of activity or the identification of new inhibitors with a broader spectrum of activity against the different mutations. It is, however, more than likely that resistance will be multifactorial and may require combinations of new agents addressing the different mechanisms involved. There may be some basis for restarting EGFR TKI therapy after a period of discontinuation, as reemergence of sensitive EGFR-mutated clones may occur with time once the selective pressure exerted by the EGFR TKIs is removed. As the disease becomes driven by less sensitive mutations, using doses close to the maximum tolerated level may be more effective in postponing progression.
Preliminary pooled data of the efficacy of gefitinib and erlotinib in patients with EGFR-mutated NSCLC suggest a superiority of erlotinib. Only a head-to-head comparison would answer the question of whether these two agents are equivalent in terms of the clinical benefit they provide. An analysis of rash vs survival in these patients might help us determine whether dose intensity is important in these patients as it probably is when these agents are used as second-line therapy in unselected populations.
Closing Remarks
Patients with EGFR-mutated tumors are "lucky" patients, which many oncologists do not hesitate to let them know. However, it is still not clear how much luckier they really are in this new era of EGFR TKIs. In other words, we do not know to what degree TKIs are changing the natural history of EGFR-mutated tumors. In spite of impressive response rates and prolongation of PFS, the great majority of these patients progress within 2 years. Cures have not been reported and may be anecdotal. In addition, EGFR-mutated tumors respond well to chemotherapy and seem to have a less virulent course. Allegedly because of crossover, all trials are showing at best a modest trend toward an improved survival in patients randomized to the TKI arms.
Could EGFR mutations be predictive of response to EGFR TKIs and PFS but not of survival, in other words mostly prognostic? In that case, the only real advantage of initial TKI therapy would be the postponement of more toxic and less convenient therapeutic modalities. We hope that this is not the case, but the fact that clear evidence of an impact on survival has not yet emerged is of some concern.
The high response rate and universal failure of EGFR-mutated tumors to EGFR TKIs makes this disease resemble another disease familiar to thoracic oncologists—small-cell lung cancer, for which no good salvage therapies exist. Fortunately, EGFR-mutated tumors are less aggressive than smallcell lung cancer, and relapses and progressions after TKI therapy are slow or indolent in many cases. But most importantly, the main difference is that we have a better understanding of the biology of EGFR-mutated tumors and we should be able to exploit that knowledge to develop new, effective, and rational therapies for this disease when progression after TKI therapy inevitably occurs.
—Roman Perez-Soler, MD
Financial Disclosure: The author has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
