Cisplatin-based chemotherapy is considered standard of

STATE OF THE ART: CONCISE REVIEW Second-Line Treatment of Advanced Non-small Cell Lung Cancer Cesare Gridelli, MD,* Andrea Ardizzoni, MD, Fortunato Ciardiello, MD, PhD, Nasser Hanna, MD, John V. Heymach, MD, Francesco Perrone, MD, PhD, Rafael Rosell, MD,# Frances A. Shepherd, MD, FRCPC,** Nick Thatcher, MD, PhD, FRCP, MB, Bchir, Johan Vansteenkiste, MD, PhD, Luigi De Petris, MD, Massimo Di Maio, MD, and Filippo De Marinis, MD Abstract: After failure of first-line chemotherapy for advanced non-small cell lung cancer, many patients remain candidates to receive further antitumor treatment. To guide clinical management of these patients and to suggest priorities for clinical research, an *Medical Oncology, S.G. Moscati Hospital, Avellino, Italy; Medical Oncology, University Hospital, Parma, Italy; Medical Oncology, Second University, Naples, Italy; Indiana University, Indianapolis; Departments of Thoracic/Head and Neck Medical Oncology and Cancer Biology, University of Texas M. D. Anderson Cancer Center, Houston; Clinical Trials Unit, National Cancer Institute, Naples, Italy; #Catalan Institute of Oncology, Barcelona, Spain; **Princess Margaret Hospital, University Health Network, Toronto, Canada; Medical Oncology, Christie Hospital, Manchester, England; Respiratory Oncology Unit, University Hospital, Leuven, Belgium; Karolinska Institutet, Stockholm, Sweden; Medical Oncology, N. Giannettasio Hospital, Rossano (CS), Italy; and Pulmonary Oncological Unit 5 th, Forlanini Hospital, Rome, Italy. Authors contributions: Conception and design: Cesare Gridelli, Filippo De Marinis. Review of the literature, data analysis and interpretation, manuscript writing, final approval of manuscript: Cesare Gridelli, Andrea Ardizzoni, Fortunato Ciardiello, Nasser Hanna, John V Heymach, Francesco Perrone, Rafael Rosell, Frances A. Shepherd, Nick Thatcher, Johan Vansteenkiste, Luigi De Petris, Massimo Di Maio, Filippo De Marinis. Disclosure: Authors disclosure of potential conflicts of interest Consultant or advisory relationships: C. G. (Eli Lilly and Roche), F. C. (Roche, Merck Darmstadt), N. T. (Roche, Astra Zeneca, Sanofi Aventis, Eli Lilly), N. H. (Bayer), F. S. (Eli Lilly, Roche, Sanofi Aventis, Osip, Astra Zeneca), J. V. (Lilly, Amgen, Astra Zeneca, Roche, Glaxo Smith Kline), J. V. H. (Astra Zeneca, Pfizer), M. D. M. (Glaxo Smith Kline). Honoraria: C. G. (Eli Lilly and Roche), A. A. (Eli Lilly, Roche, Glaxo Smith Kline), F. C. (Roche, Merck Darmstadt), F. P. (Roche), N. T. (Roche, Astra Zeneca, Sanofi Aventis, Eli Lilly), N. H. (Eli Lilly), F. S. (Eli Lilly, Roche, Sanofi Aventis, Osip, Astra Zeneca), J. V. (Eli Lilly, Amgen, Astra Zeneca, Roche, Glaxo Smith Kline, Bristol Myers Squibb), F. D. M. (Eli Lilly, Roche, Sanofi Aventis), J. V. H. (Astra Zeneca, Pfizer), M. D. M. (Glaxo Smith Kline, Roche). Stock ownership: F. S. (Eli Lilly, Astra Zeneca). Research funding: F. P. (Roche), J. V. (Eli Lilly, Amgen, Astra Zeneca), N. T. (Roche, Astra Zeneca, Sanofi Aventis, Eli Lilly), N. H. (Sanofi Aventis, Genentech, Onyx, Bristol Myers Squibb, Eli Lilly), J. V. H. (AstraZeneca, Pfizer). Address for correspondence: Dr. Cesare Gridelli, Division of Medical Oncology, S.G. Moscati Hospital, Contrada Amoretta - 83100 Avellino, Italy. E-mail: cgridelli@libero.it Copyright 2008 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/08/0304-0430 430 International Panel of Experts met in Naples (Italy) in April 2007. Results and evidence-based conclusions are presented in this article. Single-agent chemotherapy with docetaxel or pemetrexed is the recommended option for unselected patients with performance status 0 to 2 who are candidates for second-line chemotherapy for advanced non-small cell lung cancer. Docetaxel has demonstrated superiority compared with best supportive care. Pemetrexed has been shown to be noninferior to docetaxel, with a more favorable toxicity profile. Erlotinib is effective in pretreated patients, and can be given second-line in patients not suitable or intolerant to chemotherapy, and in all patients as third-line treatment after failure of second-line chemotherapy. Gefitinib failed to show superiority to placebo as second- or third-line treatment, but it has been shown to be noninferior to docetaxel. In selected patients such as lifetime nonsmokers or those of East-Asian ethnicity, erlotinib, or gefitinib (where licensed) may be considered as second-line treatment even if they are fit for chemotherapy. Best supportive care in addition to active treatment remains important for all patients, but may be the exclusive option for patients unsuitable for more aggressive therapy. Further research is mandatory, to find better treatments, and to identify clinical and molecular predictive markers of efficacy, both for chemotherapy and for novel biologic agents. Key Words: Non-small cell lung cancer, Second-line chemotherapy, Docetaxel, Pemetrexed, Erlotinib. (J Thorac Oncol. 2008;3: 430 440) Cisplatin-based chemotherapy is considered standard of care worldwide for patients with advanced non-small cell lung cancer (NSCLC). 1 Approximately one-third of patients obtain an objective response with first-line chemotherapy, and another 20 to 30% achieve temporary disease stabilization. Unfortunately, all patients ultimately suffer progression. At the time of disease progression, many patients still have a good performance status (PS) and can be considered for further active treatment. Until 2000, there was no evidence supporting the efficacy of second-line treatment, although chemotherapy was often offered to patients in good clinical condition, based on the results of phase II trials of several new drugs showing modest activity in pretreated NSCLC. 2,3 The 1997 guideline of the American Society of Clinical Journal of Thoracic Oncology Volume 3, Number 4, April 2008

Journal of Thoracic Oncology Volume 3, Number 4, April 2008 Second-Line Treatment of Advanced NSCLC Oncology (ASCO) stated that there is no current evidence that either confirms or refutes that second-line chemotherapy improves survival in patients with advanced NSCLC. 4 In recent years, the efficacy of several drugs in the second-line setting has been demonstrated in phase III trials, and secondline treatment can now be considered a standard of care. 5,6 Two chemotherapeutic agents, docetaxel and pemetrexed, and the biologic drug erlotinib are currently approved for clinical use in this setting. METHODS Search Strategy and Selection Criteria With the aims of reviewing the available evidence, obtaining a consensus on second-line treatment of advanced NSCLC in clinical practice, and suggesting the priorities for clinical research in this field, an International Panel of Experts Meeting and Conference took place in Naples, Italy, in April 2007. The oncologists who formed the scientific panel of the meeting were selected by the two chairmen (C.G. and F.D.M.). Principal investigators of all the relevant published trials and other opinion leaders in the field were invited. The Meeting was supported by Roche. The sponsor was not involved in panel discussion, nor in the preparation of manuscript. Published data for this review were identified by a PUBMED search, performed by each of the panellists on the topic assigned. Only articles published in English were considered. Important references from relevant articles were also included. Abstracts presented at the ASCO meetings and at the International Association for Study of Lung Cancer, European Society of Medical Oncology, and European Cancer Conference meetings between 1997 and 2006 were searched. For manuscript preparation, abstracts presented at the 2007 ASCO meeting, International Association for Study of Lung Cancer meeting, and European Cancer Conference meeting were also searched. RESULTS AND RECOMMENDATIONS Second-Line Therapy: Prognosis and Aims of Treatment Patients who experience disease progression during or after first-line treatment for advanced NSCLC have a limited life expectancy. Quality of life is often compromised by diseaserelated symptoms, by residual toxicity of previous chemotherapy, and by comorbid diseases. The aims of second-line treatment should be palliation of symptoms, benefit in quality of life, and prolongation of survival. The benefits of treatment should outweigh toxicity and inconvenience to the patient. Nevertheless, the impact of treatment on the natural history of the disease is modest. As shown in a recent review of 19 phase III trials, 7 in the second-line or later setting, the median objective response rate was 6.8%, and median overall survival was 6.6 months. Life expectancy of these patients is largely dependent on their clinical condition at the start of second-line treatment. Patients with PS 2, who are considered fit for further treatment may have a relative benefit from treatment similar to patients with better PS. Nevertheless, the absolute benefit is small, with most studies reporting a median survival of 3 to 4 months. Overall survival of patients in an individual patient data meta-analysis of phase III randomized trials comparing weekly with every 3 weeks docetaxel 8 was 32.7 weeks in patients with PS 0, 29 weeks in patients with PS 1 and 14 weeks in patients with PS 2. In a series of patients enrolled in a randomized trial of second-line chemotherapy, other variables including gender, stage, and best response to initial therapy as well as PS were independently associated with survival. 9 survival was better in women, in patients with stage III compared with IV, and in patients who had responded to first-line treatment. The length of time spent by patients receiving antitumor treatment has been increasing recently, due to the availability of new drugs. Nevertheless, most patients with advanced NSCLC who receive second-line treatment are near the end of life. In a retrospective review of patients treated for advanced NSCLC in a community oncology setting, the mean line of chemotherapy in patients who were receiving active treatment at the time of death was second-line. 10 Chemotherapy was given within 1 month and 2 weeks of death to 43 and 20% of patients, respectively. This can be explained in part by the increased demand for additional treatment by patients and their relatives who are unable to accept the futility of further therapy and the inevitability of death from progressive NSCLC. It may also be due to physicians inability to correctly predict life expectancy of these patients. This emphasizes the importance of identifying predictive markers of response, survival benefit, and toxicity in pretreated NSCLC patients. Role of Single-Agent Chemotherapy Docetaxel Docetaxel, 75 mg/m 2 every 3 weeks, was the first drug to be approved for second-line treatment of advanced NSCLC, based on the results of 2 phase III trials (Table 1). 11,12 In the TAX317 study, 11 docetaxel 100 mg/m 2 every 3 weeks was compared with best supportive care. When interim safety data monitoring identified a significantly higher toxic death rate in the chemotherapy arm, the protocol was amended and the dose was reduced to 75 mg/m 2. overall survival was significantly longer for patients in the chemotherapy arm (7.0 versus 4.6 months). Interestingly, all quality of life parameters favored docetaxel, although the statistically significant results favoring docetaxel were more evident with the higher dosage. 13 Febrile neutropenia was the most relevant toxicity reported with docetaxel, but was less frequent with the lower dose. In the TAX320 trial, 12 patients were randomized to receive docetaxel (100 or 75 mg/m 2, every 3 weeks) or to a control arm of vinorelbine or ifosfamide at the investigator s discretion. Although overall response rates were significantly higher and time-to-progression was significantly longer in the docetaxel arms, median overall survival was not significantly different among the three groups. Nevertheless, 1-year survival was significantly greater with docetaxel 75 mg/m 2 than with the control treatment. Patients assigned to docetaxel had more neutropenia and febrile neutropenia compared with the control arm, but the lower dose of docetaxel generally was well tolerated. Copyright 2008 by the International Association for the Study of Lung Cancer 431

Gridelli et al. Journal of Thoracic Oncology Volume 3, Number 4, April 2008 TABLE 1. Study (First Author, Year of Publication) Shepherd, 2000 11 Fossella, 2000 12 Hanna, 2004 24 Ramlau, 2006 26 Bonomi, 2005 27 Krzakowski, 2007 29 Randomized Trials of Single Agent Chemotherapy in Second-Line Treatment of Advanced NSCLC Main Eligibility Criteria One or more prior platinum-based One or more prior platinum-based One prior chemotherapy for advanced disease, One prior chemotherapy, One prior chemotherapy, One prior platinum-based TXT, docetaxel; w, weeks; PS, performance status; mo, months. Treatment Arm Patients (n) Response Rate (%) Progression-Free Overall 1-yr (%) Best supportive care 100 1.5 4.6 19 TXT 100 or 75 mg/m 2 every 3 w 104 5.8 2.4 7.0 29 TXT 100 mg/m 2 every 3 w 49 6.3 n.a. 5.9 19 TXT 75 mg/m 2 every 3 w 55 5.5 n.a. 7.5 37 Vinorelbine or ifosfamide 123 0.8 1.8 5.6 19 TXT 100 mg/m 2 every 3 w 125 10.8 1.9 5.5 21 TXT 75 mg/m 2 every 3 w 125 6.7 2.0 5.7 32 TXT 75 mg/m 2 every 3 w 288 8.8 2.9 7.9 29.7 Pemetrexed 500 mg/m 2 every 3 w 283 9.1 2.9 8.3 29.7 TXT 75 mg/m 2 every 3 w 415 5 3.0 7.1 28.7 Topotecan 2.3 mg/m 2 orally day 414 5 2.6 6.4 25.1 1 5 every 3 w TXT 75 mg/m 2 every 3 w 422 n.a. n.a. 6.9 29 Paclitaxel poliglumex 210 mg/m 2 427 n.a. n.a. 6.9 25 (PS 0 1) 175 mg/m 2 (PS 2) every 3 w TXT 75 mg/m 2 every 3 w 277 5.5 2.3 7.2 n.a. Vinflunine 320 mg/m 2 every 3 w 274 4.4 2.3 6.7 n.a. Retrospective pharmacoeconomic studies have shown that the cost of second-line docetaxel in NSCLC is in keeping with the cost of second-line treatment in other tumors, and is within an acceptable range for health care interventions. 14,15 Weekly scheduling of docetaxel may reduce toxicity, particularly neutropenia and febrile neutropenia, in pretreated NSCLC patients 16,17 without decreasing efficacy. Several randomized trials have been conducted, comparing weekly TABLE 2. Randomized Trials of Every 3 wk vs. Weekly Schedule of Docetaxel in Second-Line Treatment of Advanced NSCLC Study (First Author, Year of Publication) Phase Main Eligibility Criteria Docetaxel Schedule Gridelli, 2004 18 III One prior chemotherapy, Gervais, 2005 19 II One prior platinum-based Schuette, 2005 20 III One or more prior Lai, 2005 22 II One prior chemotherapy, Camps, 2006 21 III One prior platinum-based Chen, 2006 23 II One or more prior platinum-based mo, months; w, weeks; PS, performance status. Patients (n) Response Rate (%) Overall 1-yr (%) Febrile Neutropenia (%) 75 mg/m 2 every 3 w 110 2.7 6.7 21 5 33.3 mg/m 2 weekly for 6 w, then2wofrest 110 5.5 5.8 31 0 75 mg/m 2 every 3 w 62 4.8 5.8 18 6.5 40 mg/m 2 weekly for 6 w, 63 3.2 5.5 6 0 then2wofrest 75 mg/m 2 every 3 w 103 12.6 6.3 27 2 35 mg/m 2 weekly for 3 w, 105 10.5 9.2 39 1 then1wofrest 66 mg/m 2 every 3 w 25 12 7.8 36 4 33 mg/m 2 weekly for 2 w, 25 24 7.3 36 1 then1wofrest 75 mg/m 2 every 3 w 129 9.3 6.6 27 8 36 mg/m 2 weekly for 6 w, 125 4.8 5.4 22 1 then2wofrest 75 mg/m 2 every 3 w 33 6.1 9.5 29 12 40 mg/m 2 weekly for 2 w, 64 10.9 7.2 32 5 then1wofrest 35 mg/m 2 weekly for 3 w, 64 17.2 8.4 33 2 then1wofrest 432 Copyright 2008 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 3, Number 4, April 2008 Second-Line Treatment of Advanced NSCLC with the standard schedule of docetaxel in the second-line treatment of advanced NSCLC and are summarized in Table 2. 18 23 No trial had sufficient power to detect small, but potentially relevant differences in overall survival. An individual patient data meta-analysis, including five of these randomized trials, has been published recently. 8 overall survival was similar, without significant heterogeneity among the trials, and without relevant differential effect in subgroup analyses. Significantly less severe neutropenia and febrile neutropenia were confirmed with the weekly schedule of docetaxel. Pemetrexed Pemetrexed is a multitargeted antifolate agent. Pemetrexed 500 mg/m 2 was compared with docetaxel 75 mg/m 2 in a phase III trial (Table 1). 24 The study was originally designed to test for superiority of pemetrexed, but in an amendment to the protocol a noninferiority design was adopted. Results were not sufficient to demonstrate the noninferiority of pemetrexed according to the fixed margin method. Nevertheless, using the percent retention method, results were adequate to define noninferiority. Patients receiving docetaxel experienced significantly higher rates of severe neutropenia, febrile neutropenia, neutropenia with infections and hospitalization due to neutropenic events. Although neither superiority nor noninferiority for overall survival could be strictly demonstrated according to primary efficacy analysis, comparable response rates and progressionfree survival times supported the conclusion that an effect of pemetrexed on survival was reasonably likely. The trial results led to drug approval. 25 Other Cytotoxic Agents Tested in Phase III Trials Oral topotecan has been compared with standard docetaxel in a phase III study that was designed to test the noninferiority of oral topotecan (Table 1). 26 One-year survival rates were 25.1% with topotecan and 28.7% with docetaxel, and this difference met the predefined criteria for noninferiority of topotecan. Nevertheless, median overall survival was 27.9 weeks with topotecan and 30.7 weeks with docetaxel and the higher survival rate with docetaxel was maintained across the entire treatment period, although the difference did not reach statistical significance. Grade 3/4 neutropenia occurred more frequently with docetaxel, while grade 3/4 anemia and thrombocytopenia occurred more frequently with topotecan. Patients receiving docetaxel had significant advantages in quality of life compared with patients receiving oral topotecan. At the moment, topotecan is not approved for the second-line treatment of advanced NSCLC. Paclitaxel poliglumex (PP) is a macromolecular drug conjugate that links paclitaxel with a biodegradable polymer (poly-l-glutamic acid). The STELLAR II trial compared PP with docetaxel as second-line therapy in patients with advanced NSCLC (Table 1). 27 Patients assigned to PP experienced significantly less hematological toxicity and less febrile neutropenia. As for nonhematological toxicity, PP was associated with a lower incidence of diarrhea, stomatitis, respiratory symptoms, and fatigue, but worse neuropathy. The trial did not demonstrate any survival benefit for PP therapy when compared with docetaxel. Although PP showed comparable efficacy to standard docetaxel, with a favorable tolerability profile, the study was not designed as a noninferiority trial, and these results cannot be considered sufficient to recommend PP as a standard approach. PP is not approved for clinical use in the second-line treatment of advanced NSCLC. Vinflunine is a new microtubule inhibitor of the vinca alkaloid class with clinical activity in NSCLC. 28 Vinflunine was compared with docetaxel in a phase III trial in pretreated NSCLC patients (Table 1). 29 The primary end-point was progression free survival, with a noninferiority analysis. All efficacy endpoints including median progression free survival, response rate, and median overall survival were similar. Although the toxicity profile of the experimental arm was judged manageable, vinflunine was characterized by higher incidence of grade 3/4 anemia, abdominal pain, constipation, and fatigue. Vinflunine is not yet approved by the Food and Drug Administration (FDA, United States of America) or by the European Medicines Agency (European Union) and its use remains experimental. Role of Combination Chemotherapy In first-line treatment of good PS patients, doublet chemotherapy has been shown to be more effective than single-agent, both in terms of response and survival. 30 In the second-line setting, the clinical trials showing efficacy of chemotherapy all tested single-agent treatment. With the aim of obtaining better results, several trials compared a doublet with single agent chemotherapy. Docetaxel was used in the control arms in three of the trials 31 33 whereas other single agents were used in two studies enrolling patients who had already received docetaxel as first-line treatment. 34,35 These trials are summarized in Table 3. Only one Japanese trial that compared single-agent docetaxel with the combination of docetaxel and gemcitabine had a phase III design. 31 It was terminated early after the report of an unexpectedly high incidence of interstitial lung disease and three treatmentrelated deaths (5%) due to interstitial lung disease, seen only in combination arm. Considering all the trials, combination chemotherapy is characterized by higher toxicity, with some advantage in terms of objective response and progression-free survival, but without a significant difference in overall survival. Nevertheless, with one exception, 31 the trials were not designed as comparative phase III trials, but were randomized phase II trials, with a small sample size and inadequate statistical power to exclude potentially relevant differences. Role of Epidermal Growth Factor Receptor Inhibitors Epidermal Growth Factor receptor (EGFR) expression is common in NSCLC, and the EGFR pathway has been the target of development of new drugs against this disease. 36 Erlotinib and gefitinib are small molecules that inhibit the tyrosine kinase activity of EGFR and have been studied extensively in advanced NSCLC. Copyright 2008 by the International Association for the Study of Lung Cancer 433

Gridelli et al. Journal of Thoracic Oncology Volume 3, Number 4, April 2008 TABLE 3. Randomized Trials of Combination vs. Single-Agent Chemotherapy in Second-Line Treatment of Advanced NSCLC Study (First Author, Year of Publication) Phase Main Eligibility Criteria Treatment Arm Takeda, 2004 31 III One prior platinum-based chemotherapy, PS 0 1 Patients (n) Response Rate (%) Progression-Free Overall 1-yr (%) Docetaxel 60 mg/m 2 day 1 q3w 65 6.7 2.2 10.1 42 Docetaxel 60 mg/m 2 day 8 65 7 3.0 10.2 46 Gemcitabine 800 mg/m 2 day 1, 8 q3w Wachters, 2005 32 II One prior chemotherapy Docetaxel 75 mg/m 2 day 1 q3w 56 16 4.2 7.4 n.a. Docetaxel 60 mg/m 2 day 1 52 10 3.5 6.3 n.a. Irinotecan 200 mg/m 2 day 1 q3w Pectasides, 2005 33 II Platinum refractory, Docetaxel 75 mg/m 2 day 1 q3w 65 14 4.8 6.4 34 Docetaxel 30 mg/m 2 day 1, 8 65 20 5.6 6.5 37 Irinotecan 60 mg/m 2 day 1, 8 q3w Georgoulias, 2004 34 II Patients progressing after Irinotecan 300 mg/m 2 day 1 q3w 71 4 5.0 7 29 cisplatin-docetaxel, Gemcitabine 1000 mg/m 2 day 1, 8 76 18 7.5 9 25 Irinotecan 300 mg/m 2 day 8 q3w Georgoulias, 2005 35 II Patients pretreated with taxane gemcitabine, w, weeks; mo, months; PS, performance status. Cisplatin 80 mg/m 2 day 1 q3w 73 7 2.1 8.8 32 Cisplatin 80 mg/m 2 day 8 74 22 2.6 7.8 34 Irinotecan 110 mg/m 2 day 1, 100 mg/m 2 day 8, q3w Erlotinib Erlotinib was compared with placebo in the BR.21 phase III study (Table 4). 37 Patients were eligible if they had received one or two regimens of chemotherapy. Accrual started in 2001, after the demonstration of the efficacy of docetaxel in the second-line setting, but the placebo control arm was considered ethical by the investigators, because second-line patients were only included if they were not eligible for further chemotherapy. In other words, these patients are not comparable to the patients enrolled in the clinical trials evaluating the efficacy of docetaxel and pemetrexed. Patients were randomized in a 2:1 ratio to receive either erlotinib 150 mg orally daily or placebo. Statistically significant and clinically relevant differences were observed for progression-free survival and overall survival. This was the first randomized trial to demonstrate that an EGFR tyrosine kinase inhibitor is able to prolong survival after chemotherapy for advanced NSCLC. Treatment with erlotinib also showed significant benefits in quality of life and lung cancer-related symptoms of cough, dyspnea, and pain. 38 Based on the results of the BR.21 study, erlotinib was approved by the FDA in 2004 and is now approved worldwide for the treatment of advanced NSCLC patients failing after previous chemotherapy. 39 TABLE 4. Randomized Phase III Trials With EGFR Tyrosine Kinase Inhibitors in Second-Line Treatment of Advanced NSCLC Study (First Author, Year of Publication) Main Eligibility Criteria Treatment Arm Shepherd, 2005 37 Thatcher, 2005 41 Niho, 2007 43 Douillard, 2007 44 One or 2 prior chemotherapy regimens, unfit for further chemotherapy, PS 0 3 One or 2 prior chemotherapy regimens, PS 0 3 One or 2 chemotherapy regimens, at least one containing platinum, One or 2 chemotherapy regimens, at least one containing platinum, a Time to treatment failure. mo, months; w, weeks; PS, performance status. Patients (n) Response Rate (%) Progression-Free Overall 1-yr (%) Placebo 243 1 1.8 4.7 22 Erlotinib 150 mg daily 488 8.9 2.2 6.7 31 Placebo 563 1 2.6 a 5.1 21 Gefitinib, 250 mg daily 1129 8 3.0 a 5.6 27 Docetaxel 60 mg/m 2 every 244 12.8 2.0 14.0 54 3w Gefitinib, 250 mg daily 245 22.5 2.0 11.5 48 Docetaxel 75 mg/m 2 every 733 7.6 2.7 8.0 34 3w Gefitinib, 250 mg daily 733 9.1 2.2 7.6 32 434 Copyright 2008 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 3, Number 4, April 2008 Second-Line Treatment of Advanced NSCLC Gefitinib In 2003, on the basis of encouraging phase II data, gefitinib received accelerated approval by the FDA for patients with locally advanced or metastatic NSCLC, after failure of both platinum-based and docetaxel chemotherapies. 40 Nevertheless, no controlled phase III trial had demonstrated gefitinib efficacy, and accelerated approval regulations required the sponsor to conduct additional studies to confirm the activity and efficacy of gefitinib. Subsequently, gefitinib was compared with placebo in the ISEL phase III trial, for patients with advanced NSCLC who had received one or two regimens of chemotherapy and who were refractory to (defined as recurrent or progressive disease within 90 days of the last chemotherapy dose) or intolerant of their latest chemotherapy regimen (Table 4). 41 In this study, 1692 patients were randomly assigned in a 2:1 ratio to receive either gefitinib (orally, 250 mg daily) or placebo, plus best supportive care. The primary end-point of the trial was survival in the overall population and in the subgroup of patients with adenocarcinoma. Overall survival did not differ significantly between the groups, neither in the overall population nor among the 812 patients with adenocarcinoma. Because of these negative results, the conditions of gefitinib approval have been restricted, and the drug was relabeled by the FDA for use in patients already receiving it and obtaining a clinical benefit. 42 Gefitinib is not presently available in the United States, Canada, or most European countries, but it remains approved in Asian countries including India, Japan, and China. Results of two phase III trials comparing gefitinib to docetaxel in patients with NSCLC after one or two chemotherapy regimens have been presented recently (Table 4). 43,44 These studies have not yet been published in a peer-reviewed journal, but their results have been considered for this review, because they add important evidence to the topic of choice between chemotherapy and EGFR inhibitors. Both trials were designed as noninferiority trials, with overall survival as the primary end-point. In the smaller study, conducted in Japanese patients by Niho et al., 43 according to predefined criteria, noninferiority for gefitinib was not proven. Nevertheless, the study design allowed cross-over, and the impact of postprogression treatment on overall survival is difficult to assess. Secondary end-points, such as objective response and progression-free survival, largely unaffected by subsequent therapy, showed similar or superior efficacy for gefitinib. In the INTEREST trial, 1466 patients in progression after one or two chemotherapy regimens were randomized to docetaxel or gefitinib. 44 Results of the trial are summarized in Table 4. The study met the primary end point of demonstrating noninferiority of gefitinib related to docetaxel in terms of overall survival, in a population of patients not selected for molecular characteristics. Predictive Factors of Benefit from Erlotinib or Gefitinib Clinical benefit from the administration of erlotinib and gefitinib appears greater in certain subgroups of patients, but currently it is not possible to identify patients who will not derive benefit from treatment. Efforts have been made to identify predictive factors of the activity and efficacy of treatment with erlotinib and gefitinib. Potential predictive factors for antitumor efficacy of EGFR inhibitors in patients with advanced NSCLC can be divided into several categories: clinical characteristics, pathologic characteristics, and molecular characteristics. In BR.21, objective response to erlotinib was more frequent in women, in patients with adenocarcinoma, in patients with East-Asian ethnicity and in patients who had never smoked. 37 Nevertheless, the objective response rate obtained with EGFR tyrosine-kinase inhibitors in unselected patients is similar, if not higher, to the results obtained with cytotoxic agents. 11,12,24 Furthermore, relatively higher activity in selected subgroups of patients should not be interpreted as a proof that the efficacy is limited to those patients. The correlation between a surrogate end-point (objective response) and overall survival in advanced NSCLC is even debated for cytotoxic drugs. 45 With targeted agents, the concept that clinical benefit is limited to the occurrence of an objective response is even weaker. 46 Subgroup analyses of the BR.21 trial show that erlotinib had a beneficial effect on survival in almost all subgroups tested. 37 A significant interaction was only found between smoking history and treatment, with erlotinib showing superiority compared with placebo in never smokers, but not in smokers. 37,47 These results are similar to those obtained with gefitinib in the ISEL trial, 41 where significant benefit was shown for gefitinib over placebo in never-smokers, but not in smokers. Preplanned subgroup analyses showed significantly longer survival in gefitinibtreated patients of East-Asian ethnicity and for never-smokers. Nevertheless, these data should not be seen as a proof of the absence of efficacy of EGFR inhibitors in smokers. In an unplanned subset analysis of BR.21, despite a very low tumor response rate, male ever-smokers with squamous cell NSCLC derived a significant survival benefit from erlotinib. 47 Analyses of the predictive role of molecular markers in the BR.21 and ISEL trials 48 50 are summarized in Table 5. According to the results of molecular analyses, selection or exclusion of patients for treatment with EGFR inhibitors should not be based on EGFR protein expression, as determined by immunohistochemistry. 51 In recent years, it has been repeatedly shown that a substantial percentage of tumors with objective response to gefitinib or erlotinib harbor somatic mutations in the EGFR gene. 52 55 Nevertheless, at the present, there are no definitive data to consider mutational analysis necessary to identify patients for treatment with EGFR inhibitors, since patients with wild-type EGFR clearly derive considerable survival benefit from treatment. Similarly, the analysis of KRAS mutational status has no established role in the management of patients who are candidates to receive erlotinib. Fluorescence in situ hybridization can determine the number of gene copies of EGFR in tumor cell DNA. 49 Further prospective studies are needed before a conclusion can be made about the role of EGFR gene amplification as a predictive test for efficacy of an EGFR inhibitor, although at this time, it appears that this may be the most sensitive test to predict for a differential survival benefit from EGFR tyrosine kinase inhibitor therapy. In the INTEREST Copyright 2008 by the International Association for the Study of Lung Cancer 435

Gridelli et al. Journal of Thoracic Oncology Volume 3, Number 4, April 2008 TABLE 5. Analysis of Molecular Predictive Factors in the BR.21 and ISEL Trials Number of Samples HR For BR.21 49,50 p Test For Interaction Multivariate Analysis Number of Samples HR For ISEL 48 p Test For Interaction Multivariate Analysis EGFR expression 325 Not significant Not 379 0.049 Not feasible Negative 141 (43%) 0.68 (0.49 0.95) 0.02 (0.25) significant 115 (30%) 1.57 (0.86 2.87) 0.140 Positive 184 (57%) 0.93 (0.63 1.36) 0.70 264 (70%) 0.77 (0.56 1.08) 0.126 EGFR copy number 159 Not significant (0.12) 0.009 370 0.045 Not feasible 98 (62%) 0.80 (0.49 1.29) 0.35 256 (69%) 1.16 (0.81 1.64) 0.417 61 (38%) 0.43 (0.23 0.78) 0.004 114 (31%) 0.61 (0.36 1.04) 0.067 EGFR mutational status 204 Not significant (0.47) Not significant 215 The limited amount of data prevents meaningful evaluation of clinical outcomes in relation to Wild-type 170 (83%) 0.74 (0.52 1.05) 0.09 189 (88%) these mutations Mutated 34 (17%) 0.55 (0.25 1.19) 0.12 26 (12%) KRAS mutational status 206 Not significant (0.09) Not significant 152 The limited amount of data prevents meaningful evaluation of clinical outcomes in relation to Wild-type 176 (85%) 0.69 (0.49 0.97) 0.03 140 (92%) these mutations Mutated 30 (15%) 1.67 (0.62 4.50) 0.31 12 (8%) trial, there was no evidence that high EGFR gene copy number, measured by fluorescence in situ hybridization, predicted superior survival for gefitinib over docetaxel. 44 In the subset of 174 patients with high EGFR copy number, median overall survival was 7.5 months with docetaxel and 8.4 months with gefitinib, and this difference was not statistically significant. CONSENSUS FOR CLINICAL PRACTICE Who should be Treated? Only patients with clinical and/or radiologic evidence of disease progression after or during first-line chemotherapy should be considered for second-line treatment. In patients achieving objective response or disease stabilization with first-line therapy, maintenance treatment with the same or different agents has not definitively been shown to improve overall survival, and should not be considered a standard approach. 56 59 In a recent trial, 59 307 stable and responding patients were randomized to receive docetaxel either immediately after first-line chemotherapy or on disease progression. overall survival showed a trend favoring immediate treatment, but the difference did not reach statistical significance. Clinical trials showing efficacy of second-line chemotherapy were limited to patients with good or intermediate PS (ECOG 0 2). 8,11,12,24 There is no evidence supporting the effectiveness of second-line chemotherapy in patients with worse PS. In BR.21, patients with PS 2 3 were eligible, and according to subgroup analysis, erlotinib was beneficial in these patients. 37 Nevertheless, the decision to treat a patient in poor clinical condition should be based on careful evaluation of the expected toxicity profile and of the potential benefit in terms of survival and of symptom control. To date, there are no clinical trials of second-line therapy dedicated to elderly patients. The only evidence of efficacy of second-line treatment in the elderly patients comes from retrospective subgroup analyses. 60 Particular caution should be used when treating elderly patients, considering the palliative aim of second-line treatment and the higher risk of toxicity in these patients. Which Treatment? To date, the study conducted in Japanese patients and the INTEREST trial, comparing gefitinib to docetaxel, are the only available comparison between chemotherapy and an EGFR inhibitor in patients who are candidates for second-line treatment. 43,44 These trials have not yet been published in a peer-reviewed journal. In the Japanese trial, noninferiority of gefitinib was not proven, but the study has several limitations, because the primary end-point could have been affected by cross-over. In the INTEREST trial, characterized by a larger sample size, primary end point of noninferiority of gefitinib was met. According to this study, gefitinib represents a reasonable option for second-line treatment of patients with advanced NSCLC. Nevertheless, definitive publication of these results, and results from other trials currently ongoing should be awaited. Which Chemotherapy? According to the results of phase III trials, chemotherapy with one approved drug (docetaxel or pemetrexed) should be the first choice in patients with good PS, experiencing disease progression during or after first-line treatment. Although a retreatment strategy is sometimes considered in patients with long progression-free intervals after first-line treatment, there are no randomized data supporting this strategy. With the evidence currently available, there is no established role for other single agents, or for combination chemotherapy. These and other undefined issues for the second-line chemotherapy of advanced NSCLC in clinical practice are listed in Table 6. 436 Copyright 2008 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 3, Number 4, April 2008 Second-Line Treatment of Advanced NSCLC TABLE 6. Undefined Issues in the Administration of Second-Line Chemotherapy of Advanced NSCLC in Clinical Practice Maintenance treatment after first-line Optimal duration of treatment Rechallenge with drugs used in first-line Role of other cytotoxic drugs as single-agent Costs 3rd line Maintenance treatment after first-line vs. second-line treatment after progressive disease? Currently, there is no definitive evidence from randomized trials pro or contra this strategy Maintenance studies with erlotinib, gefitinib, gemcitabine, and pemetrexed are currently on-going Limited number of cycles? Continuing treatment until progressive disease or unacceptable toxicity? Re-induction with drugs obtaining objective response in first-line in patients with a long progression-free interval? Currently, there is no evidence from randomized clinical trials pro or contra this strategy Other drugs besides those showing positive results (docetaxel, pemetrexed) or negative results (topotecan, vinorelbine)? Currently, there are no randomized phase III trials supporting a role of other single-agent chemotherapy What role should be given to pharmacoeconomic considerations in the choice of second-line treatment? Is there a role for further administration of chemotherapy in patients failing second-line treatment? Standard docetaxel is administered 75 mg/m 2 every 3 weeks. 11,12 Weekly docetaxel produces significantly less hematological toxicity, with comparable efficacy, and may be considered as an alternative. 8 Nevertheless, some clinicians feel that the inconvenience of the weekly schedule is not justified, given the availability of other drugs administered in more convenient schedules. 61 The clinical trials comparing weekly with every 3-week docetaxel used different doses and schedules in the weekly arms, and there is no evidence of superiority of one of these schedules. EGFR Inhibitors Based on the BR.21 trial, 37 erlotinib should be administered as second-line treatment in patients considered unfit for chemotherapy or as third-line treatment. The negative results of the ISEL trial 41 discouraged the use of gefitinib in unselected patients, but the recently reported results of the INTEREST trial, conducted in patients who were fit for further chemotherapy after the failure of one or two previous chemotherapy regimens, 44 support the consideration of EGFR inhibitors as a reasonable alternative to second-line chemotherapy. At present, there is no evidence to recommend the routine use of molecular marker analysis (EGFR expression, EGFR gene mutations, KRAS or other gene mutations, EGFR gene copy number) to select patients for treatment with EGFR inhibitors. Research Issues in Second-Line Treatment of Advanced NSCLC A number of questions concerning second-line treatment of advanced NSCLC remain to be answered. Questions, that in the opinion of the panellists should drive future clinical research in this field, are reported in Table 7. Chemotherapy seems to have reached a plateau in efficacy, and it is unlikely that significant advances would come from new chemotherapeutic agents. For this reason, clinical research to answer remaining questions regarding chemotherapy is of low priority. Of greater relevance are the questions regarding biologic agents, and in particular EGFR inhibitors. Comparison between standard chemotherapy and an EGFR inhibitor in unselected patients, and in patients positively or negatively selected for the presence of predictive factors is a research issue of high priority. Results of trials comparing gefitinib to docetaxel have recently given the first important evidences regarding this issue. 43,44 Further phase III trials comparing an EGFR inhibitor with chemotherapy are currently ongoing; some of these studies are conducted in unselected population, others in patients selected according to the presence of molecular predictive factors. Both types of trials will give important answers for clinical practice. Another interesting issue is the optimal sequence of chemotherapy and EGFR inhibitors. The TORCH phase III trial, conducted in Italy and Canada, is currently assessing in unselected patients whether an experimental strategy including first-line erlotinib followed at progression by cisplatin-based chemotherapy is not inferior in terms of survival to the standard treatment of first-line chemotherapy, followed at progression by erlotinib. Of course, of great importance is the conduct of clinical trials testing the activity of other new drugs. The low objective response rate usually obtained in second-line treatment can result in premature rejection of potentially useful drugs, if end-points alternative to objective response are not properly identified and applied. 62 Among drugs under clinical development in second-line treatment of advanced NSCLC are new molecules targeting the EGFR pathway 63 and several drugs targeting the vascular endothelial growth factor (VEGF) pathway. 64 In particular, there is strong biologic rationale for therapeutic approaches targeting both pathways. 65 Bevacizumab, a monoclonal antibody against VEGF, has shown promising results in advanced, nonsquamous NSCLC, in a randomized phase II trial, both in combination with chemotherapy and in combination with erlotinib. 66 In particular, the bevacizumab plus erlotinib combination, targeting both VEGF and EGFR pathways, may represent an attractive alternative to chemotherapy in relapsed NSCLC, if the results are confirmed in a fully powered phase III trial. Vandetanib (ZD6474) is a multitarget tyrosine kinase inhibitor, inhibiting two key pathways in tumor growth. Vandetanib showed promising results in randomized phase II trials, both as a single agent 67 and in combination with chemotherapy. 68 Among EGFR inhibitors, a number of newer molecules are currently in clinical development. These include monoclonal antibodies, such as panitumumab and matu- Copyright 2008 by the International Association for the Study of Lung Cancer 437

Gridelli et al. Journal of Thoracic Oncology Volume 3, Number 4, April 2008 TABLE 7. Research Strategies in Second-Line Treatment of Advanced NSCLC Experimental Issue General issue Prospective pharmacoeconomic analysis of second-line treatment, including quality of life evaluation Chemotherapy Weekly docetaxel vs. pemetrexed Re-induction with chemotherapy used as first-line vs. standard second-line treatment in sensitive patients Combination chemotherapy vs. single agent chemotherapy (ongoing: pemetrexed vs. pemetrexed carboplatin; docetaxel vs. docetaxel carboplatin) Duration of second-line treatment: limited number of cycles vs. until progression Maintenance chemotherapy after first-line treatment vs. second-line treatment after progressive disease Maintenance studies with gemcitabine and pemetrexed currently on-going Special populations Study dedicated to PS3 patients Study dedicated to elderly patients EGFR inhibitors EGFR TKIs vs. chemotherapy in molecularly-selected patients (ongoing: erlotinib vs. docetaxel or pemetrexed, gefitinib vs. pemetrexed) EGFR TKIs vs. chemotherapy in unselected patients (ongoing: erlotinib vs. pemetrexed with stratification based on EGFR copy number) Optimal sequence: EGFR inhibitor as first-line followed by chemotherapy as second-line vs. chemotherapy as first-line followed by EGFR inhibitor as second-line (ongoing: erlotinib followed by cisplatin/gemcitabine vs. cisplatin/gemcitabine followed by erlotinib) EGFR TKIs vs. chemotherapy in negatively selected patients (smokers, squamous histology, EGFR negative tumors) Chemotherapy vs. chemotherapy EGFR inhibitor (ongoing: pemetrexed vs. pemetrexed cetuximab) Maintenance treatment after first-line treatment vs. treatment after progressive disease Maintenance studies with erlotinib and gefitinib are currently on-going Other targeted agents Chemotherapy vs. chemotherapy VEGF pathway inhibitor (ongoing: chemotherapy vs. chemotherapy ZD6474) Erlotinib vs. erlotinib bevacizumab (ongoing) Erlotinib vs. erlotinib multitarget inhibitor (ongoing erlotinib vs. erlotinib sunitinib) Chemotherapy vs. chemotherapy bevacizumab in patients progressing after first-line chemotherapy plus bevacizumab Molecular predictive markers Discovery and validation of predictive markers on tumor tissue and/or serum of the efficacy of target-based agents Discovery and validation of predictive markers on tumor tissue and/or serum of the efficacy of chemotherapy Priority Medium Medium Medium Medium zumab; dual inhibitors of EGFR and VEGF receptor, such as vandetanib and AEE788, inhibitors of multiple EGFR family members, and irreversible inhibitors, such as canertinib and HKI272. 63 Preclinical studies suggest that some of these agents may have activity in tumors refractory to erlotinib or gefitinib. Further research is needed to elucidate the role of these agents in patients with EGFR inhibitor-naive and EGFR inhibitor-refractory disease, to define the molecular characteristics that predict response, and to determine whether these drugs should be used in combination with other targeted agents or chemotherapy. The discovery and validation of predictive markers of the efficacy, both for chemotherapy and for new drugs using biologic samples, is a research strategy of high priority. The availability of clinical or molecular factors that are able to predict sensitivity, resistance and even toxicity will be critical for patient selection. There has been little progress in the use of molecular testing to identify patients for standard chemotherapy. Nevertheless, this will likely not be the case for the molecularly targeted agents, and recent studies also suggest that it may be possible to predict benefit using molecular tests for chemotherapy as well. Available pharmacoeconomic studies on second-line treatment are retrospective models, with important limitations. Prospective pharmacoeconomic analysis of second-line treatment, in the opinion of the panelists would be useful to inform clinical practice in this setting. ACKNOWLEDGMENTS The International Panel of Experts Meeting and Conference was supported by Roche. The sponsor was not involved in panel discussion, nor in the preparation of manuscript. REFERENCES 1. Non-small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. BMJ 1995;311: 899 909. 2. Fossella FV, Lee JS, Hong WK. Management strategies for recurrent non-small cell lung cancer. Semin Oncol 1997;24:455 462. 3. Huisman C, Smit EF, Giaccone G, Postmus PE. Second-line chemotherapy in relapsing or refractory non-small-cell lung cancer: a review. J Clin Oncol 2000;18:3722 3730. 4. American Society of Clinical Oncology. Clinical practice guidelines for the treatment of unresectable non-small-cell lung cancer. J Clin Oncol 1997;15:2996 3018. 5. Pfister DG, Johnson DH, Azzoli CG, et al. American Society of Clinical Oncology treatment of unresectable non-small-cell lung cancer guideline: update 2003. J Clin Oncol 2004;22:330 353. 438 Copyright 2008 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 3, Number 4, April 2008 Second-Line Treatment of Advanced NSCLC 6. Ettinger DS, Bepler G, Bueno R, et al. Non-small cell lung cancer clinical practice guidelines in oncology. J Natl Compr Canc Netw 2006;4:548 582. 7. Hotta K, Fujiwara Y, Kiura K, et al. Relationship between response and survival in more than 50,000 patients with advanced non-small cell lung cancer treated with systemic chemotherapy in 143 phase III trials. J Thorac Oncol 2007;2:402 407. 8. Di Maio M, Perrone F, Chiodini P, et al. Individual patient data meta-analysis of docetaxel administered once every 3 weeks compared with once every week second-line treatment of advanced non-small-cell lung cancer. J Clin Oncol 2007;25:1377 1382. 9. Weiss G, Rosell R, Fossella F, et al. The impact of induction chemotherapy on the outcome of second-line therapy with pemetrexed or docetaxel in patients with advanced non-small-cell lung cancer. Ann Oncol 2007;18:453 460. 10. Murillo JR, Koeller J. Chemotherapy given near the end of life by community oncologists for advanced non-small-cell lung cancer. Oncologist 2006;11:1095 1099. 11. Shepherd FA, Dancey J, Ramlau R, et al. Prospective randomized trial of docetaxel versus best supportive care in patients with Non-Small-Cell Lung Cancer previously treated with platinum-based chemotherapy. J Clin Oncol 2000;18:2085 2103. 12. Fossella FV, DeVore R, Kerr RN, et al. for the TAX 320 Non-Small- Cell Lung Cancer Study Group. Randomized phase III trial of docetaxel versus vinorelbine or ifosfamide in patients with Non-Small Cell Lung Cancer previously treated with platinum - containing chemotherapy regimens. J Clin Oncol 2000;18:2354 2362. 13. Dancey J, Shepherd FA, Gralla RJ, Kim YS. Quality of life assessment of second-line docetaxel versus best supportive care in patients with non-small-cell lung cancer previously based with platinum-based chemotherapy: results of a prospective, randomized phase III trial. Lung Cancer 2004;43:183 194. 14. Holmes J, Dunlop D, Hemmett L, et al. A cost-effectiveness of docetaxel in the second-line treatment of non-small-cell lung cancer. Pharmacoeconomics 2004;22:581 589. 15. Leighl NB, Shepherd FA, Kwong R, et al. Economic analysis of the TAX 317 trial: docetaxel versus best supportive care as second-line therapy of advanced non-small-cell lung cancer. J Clin Oncol 2002;20: 1344 1352. 16. Hainsworth JD, Burris HA 3rd, Erland JB, et al. Phase I trial of docetaxel administered by weekly infusion in patients with advanced refractory cancer. J Clin Oncol 1998;16:2164 2168. 17. Lilenbaum RC, Schwartz MA, Seigel L, et al. Phase II trial of weekly docetaxel in second-line therapy for non-small cell lung carcinoma. Cancer 2001;92:2158 2163. 18. Gridelli C, Gallo C, Di Maio M, et al. A randomised clinical trial of two docetaxel regimens (weekly vs 3 week) in the second-line treatment of non-small-cell lung cancer. The DISTAL 01 study. Br J Cancer 2004; 91:1996 2004. 19. Gervais R, Ducolone A, Breton JL, et al. Phase II randomised trial comparing docetaxel given every 3 weeks with weekly schedule as second-line therapy in patients with advanced non-small-cell lung cancer (NSCLC). Ann Oncol 2005;16:90 96. 20. Schuette W, Nagel S, Blankenburg T, et al. Phase III study of secondline chemotherapy for advanced non-small-cell lung cancer with weekly compared with 3-weekly docetaxel. J Clin Oncol 2005;23:8389 8395. 21. Camps C, Massuti B, Jimenez A, et al. Randomized phase III study of 3-weekly versus weekly docetaxel in pretreated advanced non-small-cell lung cancer: a Spanish Lung Cancer Group trial. Ann Oncol 2006;17: 467 472. 22. Lai CL, Tsai CM, Chiu CH, et al. Phase II randomized trial of tri-weekly versus day 1 and 8 weekly docetaxel as a second-line treatment of advanced non-small cell lung cancer. Jpn J Clin Oncol 2005;35:700 706. 23. Chen YM, Shih JF, Perng RP, et al. A randomized trial of different docetaxel schedules in non-small cell lung cancer patients who failed previous platinum-based chemotherapy. Chest 2006;129:840 842. 24. Hanna N, Shepherd FA, Fossella FV, et al. Randomized phase III trial of pemetrexed versus docetaxel in patients with non-small-cell lung cancer previously treated with chemotherapy. J Clin Oncol 2004;22:1589 1597. 25. Cohen MH, Johnson JR, Wang JC, et al. FDA drug approval summary: pemetrexed for injection (Alimta) for the treatment of non-small cell lung cancer. Oncologist 2005;10:363 368. 26. Ramlau R, Gervais R, Krzakowski M, et al. Phase III study comparing oral topotecan to intravenous docetaxel in patients with pretreated advanced non-small-cell lung cancer. J Clin Oncol 2006;24:2800 2807. 27. Bonomi P, Paz-Ares L, Langer C, et al. Xyotax versus docetaxel for the second-line treatment of non small cell lung cancer: the STELLAR 2 phase III study. Lung Cancer 2005;49:S35. 28. Bennouna J, Breton JL, Tourani JM, et al. Vinflunine - an active chemotherapy for treatment of advanced non-small-cell lung cancer previously treated with a platinum-based regimen: results of a phase II study. Br J Cancer 2006;94:1383 1388. 29. Krzakowski M, Douillard J, Ramlau R, et al. Phase III study of vinflunine versus docetaxel in patients with advanced non-small cell lung cancer previously treated with a platinum-containing regimen. J Clin Oncol, ASCO Annual Meeting Proceedings Part I. Vol 25, No. 18S (June 20 Supplement), 2007:7511. 30. Delbaldo C, Michiels S, Syz N, et al. Benefits of adding a drug to a single-agent or a 2-agent chemotherapy regimen in advanced non-smallcell lung cancer: a meta-analysis. JAMA 2004;292:470 484. 31. Takeda K, Negoro S, Tamura T, et al. Docetaxel versus docetaxel plus gemcitabine for second-line treatment of non-small cell lung cancer: results of a JCOG randomized trial (JCOG0104). J Clin Oncol, 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition). Vol 22, No 14S (July 15 Supplement), 2004:7034. 32. Wachters FM, Groen HJ, Biesma B, et al. A randomised phase II trial of docetaxel vs docetaxel and irinotecan in patients with stage IIIb-IV non-small-cell lung cancer who failed first-line treatment. Br J Cancer 2005;92:15 20. 33. Pectasides D, Pectasides M, Farmakis D, et al. Comparison of docetaxel and docetaxel-irinotecan combination as second-line chemotherapy in advanced non-small-cell lung cancer: a randomized phase II trial. Ann Oncol 2005;16:294 299. 34. Georgoulias V, Kouroussis C, Agelidou A, et al. Irinotecan plus gemcitabine vs irinotecan for the second-line treatment of patients with advanced non-small-cell lung cancer pretreated with docetaxel and cisplatin: a multicentre, randomised, phase II study. Br J Cancer 2004; 91:482 488. 35. Georgoulias V, Agelidou A, Syrigos K, et al. Second-line treatment with irinotecan plus cisplatin vs cisplatin of patients with advanced nonsmall-cell lung cancer pretreated with taxanes and gemcitabine: a multicenter randomised phase II study. Br J Cancer 2005;93:763 769. 36. Ciardiello F, De Vita F, Orditura M, Tortora G. The role of EGFR inhibitors in nonsmall cell lung cancer. Curr Opin Oncol 2004;16:130 135. 37. Shepherd FA, Rodrigues PJ, Ciuleanu T, et al. Erlotinib in previously treated non-small cell lung cancer. N Engl J Med 2005;353:123 132. 38. Bezjak A, Tu D, Seymour L, et al. Symptom improvement in lung cancer patients treated with erlotinib: quality of life analysis of the National Cancer Institute of Canada Clinical Trials Group Study BR.21. J Clin Oncol 2007;24:3831 3837. 39. Johnson JR, Cohen M, Sridhara R, et al. Approval summary for erlotinib for treatment of patients with locally advanced or metastatic non-small cell lung cancer after failure of at least one prior chemotherapy regimen. Clin Cancer Res 2005;11:6414 6421. 40. Cohen MH, Williams GA, Sridhara R, et al. United States Food and Drug Administration drug approval summary: gefitinib (ZD1839, Iressa) tablets. Clin Cancer Res 2004;10:1212 1218. 41. Thatcher N, Chang A, Parikh P, et al. Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small-cell lung cancer: results from a randomised, placebo-controlled, multicentre study (Iressa Evaluation in Lung cancer). Lancet 2005;366: 1527 1537. 42. Blackhall F, Ranson M, Thatcher N. Where next for gefitinib in patients with lung cancer? Lancet Oncol 2006;7:499 507. 43. Niho S, Ichinose Y, Tamura T, et al. Results of a randomized Phase III study to compare the overall survival of gefitinib (IRESSA) versus docetaxel in Japanese patients with non-small-cell lung cancer who failed one or two chemotherapy regimens. J Clin Oncol, 2007 ASCO Annual Meeting Proceedings Part I. Vol 25, No. 18S (June 20 Supplement), 2007:LBA7509. 44. Douillard JY, Kim ES, Hirsh V, et al. Phase III, randomized, open-label, Copyright 2008 by the International Association for the Study of Lung Cancer 439