Lung cancer is the leading cause of cancer death worldwide.

ORIGINAL ARTICLE A Randomized Phase II Study of a Combination of Docetaxel and S-1 versus Docetaxel Monotherapy in Patients with Non-small Cell Lung Cancer Previously Treated with Platinum-Based Chemotherapy Results of Okayama Lung Cancer Study Group (OLCSG) Trial 0503 Yoshihiko Segawa, MD, PhD,* Katsuyuki Kiura, MD, PhD, Katsuyuki Hotta, MD, PhD, Nagio Takigawa, MD, PhD, Masahiro Tabata, MD, PhD, Keitaro Matsuo, MD, PhD, Hiroshige Yoshioka, MD, Hidetoshi Hayashi, MD, Haruyuki Kawai, MD, PhD, Keisuke Aoe, MD, PhD,* Tadashi Maeda, MD, PhD,* Hiroshi Ueoka, MD, PhD,* and Mitsune Tanimoto, MD, PhD Background: The survival impact of single-agent treatment with docetaxel, the standard regimen for relapsed patients with non-small cell lung cancer (NSCLC), remains modest. We conducted a randomized phase II study to evaluate the efficacy and safety of the combination of docetaxel and S-1 in the second-line setting. Methods: Patients with relapse of NSCLC after first-line platinumbased chemotherapy were randomly assigned to docetaxel alone (60 mg/m 2, day 1, q3 weeks; arm A) or a combination of docetaxel (40 mg/m 2, day 1, q3 weeks) and S-1 (80 mg/m 2, days 1 15; arm B). The primary end point was response rate, whereas secondary endpoints included overall survival, progression-free survival, and toxicity. Results: Between 2005 and 2008, a total of 60 patients were enrolled in the study. The objective response rates were 20.7% and 16.1% in arms A and B, respectively (p 0.81). Progression-free survival was comparable in the two arms (median: 3.7 versus 3.4 months, p 0.27), whereas overall survival time was longer in arm A (22.9 versus 8.7 months, p 0.02). The major toxicity was *Department of Respiratory Medicine, National Hospital Organization Yamaguchi-Ube Medical Center, Yamaguchi; Department of Medicine and Thoracic Oncology, National Hospital Organization Shikoku Cancer Center, Ehime; Department of Hematology, Oncology, and Respiratory Medicine, Okayama University Graduate School of Medicine, Dentistry, and Pharmaceutical Sciences and Okayama University Hospital, Okayama; Department of Epidemiology and Prevention, Aichi Cancer Center Research Institute, Aichi; Department of Respiratory Medicine, Kurashiki Central Hospital, Okayama; and Department of Medicine, Okayama Saiseikai General Hospital, Okayama, Japan. Disclosure: Drs. Kiura and Takigawa were given honoraria from Sanofi Aventis for their lectures. The other authors declare no conflicts of interest. Address for correspondence: Yoshihiko Segawa, MD, PhD, Department of Respiratory Medicine, National Hospital Organization Yamaguchi-Ube Medical Center, 685 Higashi-Kiwa, Ube, Yamaguchi 755-0241, Japan. E-mail: ysegawa1174@aol.com Copyright 2010 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/10/0509-1430 myelosuppression with grade 3 neutropenia in 89.7% of patients versus 64.5% in arms A and B, respectively. Conclusions: This study suggests that docetaxel monotherapy should continue to be considered the standard for second-line chemotherapy against NSCLC. Key Words: Docetaxel, Non-small cell lung cancer, Second-line chemotherapy, S-1. (J Thorac Oncol. 2010;5: 1430 1434) Lung cancer is the leading cause of cancer death worldwide. Non-small cell lung cancer (NSCLC) accounts for approximately 80% of all lung cancers, and most patients with NSCLC present with advanced disease at the time of diagnosis. Platinum-based chemotherapy has become the mainstay of treatment for patients with advanced NSCLC, because it is associated with a modest prolongation of life span and improvement of quality of life. 1 3 However, these patients will ultimately experience disease progression and might require second-line chemotherapy. Thus far, therapeutic efficacy has been demonstrated for several agents in second-line chemotherapy against advanced NSCLC. 4 7 Compared with best supportive care, docetaxel produced superior overall survival and significant improvement of quality of life. 4 Recent noninferiority studies also demonstrated that therapeutic results with some novel agents were equivalent to those with docetaxel in the second-line setting. 6,7 However, the therapeutic benefits of these agents remain modest, with median survival times of approximately 8 months. Thus, it is necessary to improve the efficacy of second-line treatment of advanced NSCLC. S-1 is an oral anticancer agent comprised of tegafur, 5-chloro-2,4-dihydroxypyridine, and potassium oxonate, in a molar ratio of 1:0.4:1, based on biochemical modulation of 5-fluorouracil. 8 In a phase II trial of S-1 for advanced NSCLC conducted in Japan, the response rate was 22%, and the 1430

Randomized Phase II Study in Relapsed NSCLC median survival time was 10.2 months. 9 There were no severe or unexpected nonhematologic toxicities. In addition, there were less severe hematologic adverse events, with incidences of grade 3 and 4 toxicities of 7% for neutropenia, 2% for anemia, and 2% for thrombocytopenia. One theoretical strategy for improving the efficacy of second-line treatment is to combine agents with different mechanisms of action and toxicity. We speculated that a combination of docetaxel and S-1 might have stronger effects against NSCLC than single-agent docetaxel for the following reasons: (1) docetaxel and S-1 have exhibited synergistic effects in human gastric and breast cancer xenograft models 10 12 ; (2) the toxicity profiles, in particular those of hematologic adverse events, of these two agents do not overlap; (3) promising results for a combination of docetaxel and oral fluoropyrimidine capecitabine have been reported in the treatment of advanced NSCLC 13,14 ; and (4) superior survival with a combination of docetaxel and capecitabine versus single-agent docetaxel was obtained in a randomized trial conducted for anthracycline-pretreated patients with advanced breast cancer. 15 With this background, the Okayama Lung Cancer Study Group (OLCSG) conducted a randomized phase II study to compare the efficacy and safety of a combination of docetaxel and S-1 with those of single-agent docetaxel in patients with NSCLC previously treated with platinum-based chemotherapy. PATIENTS AND METHODS Eligibility Criteria Patients aged 75 years or younger with a life expectancy of at least 3 months and histologically and/or cytologically confirmed advanced NSCLC were eligible for the study. In addition, all patients met the following inclusion criteria: stage IIIB or IV disease, at least one target lesion, one prior platinum-based chemotherapy regimen with the last dose administered at least 4 weeks before entry, an Eastern Cooperative Oncology Group performance status (PS) of 0 to 1, adequate bone marrow function (neutrophils 2000/mm 3, platelets 100,000/mm 3, and hemoglobin 10.0 g/dl), hepatic function (total bilirubin 1.5 mg/dl, aspartate aminotransferase and alanine aminotransferase within 2.5 times the upper normal limit), renal function (serum creatinine 1.5 mg/dl and calculated creatinine clearance level 60 ml/min [Cockroft and Gault formula]), and pulmonary function (partial pressure of arterial oxygen 60 Torr). Patients were excluded from the study of they had radiographic signs of interstitial pneumonia, serious or uncontrolled concomitant systemic disorders, a concomitant malignancy, symptomatic brain metastases, pleural and/or pericardial effusion requiring treatment, previous treatment with fluoropyrimidines including S-1, concomitant use of 5-fluorocytosine, phenytoin, or warfarin potassium, which might pharmacologically interfere with S-1, or were pregnant or breast feeding. Patients who had undergone first-line docetaxel-containing regimens were allowed to enter the study if they had achieved objective responses to the first-line chemotherapy. This study was conducted in compliance with the principles of the Declaration of Helsinki, and the protocol was approved by the local Institutional Review Boards. All patients gave written informed consent before study entry. Treatment Randomization was conducted by a registration office (Aichi Cancer Center Research Institute) and was dynamically balanced for PS, previous chemotherapy with docetaxel, and institution. After balancing with respect to the treatment arms in each stratum for each prognostic factor using a minimization method, patients were randomly allocated to receive docetaxel 60 mg/m 2 administered as a 1-hour intravenous infusion every 3 weeks (arm A) or docetaxel 40 mg/m 2 on the same schedule combined with S-1 administered orally at 40 mg/m 2 twice a day after meals between the evening of day 1 and the morning of day 15 (arm B) for a minimum of four courses if possible. The doses of docetaxel and S-1 in this study were determined based on the results of a phase I study conducted in Japanese patients with advanced or recurrent gastric cancer. 16 The actual dose of S-1 was selected as follows: in patients with body surface area (BSA) 1.25 m 2, 40 mg twice a day; BSA 1.25 m 2 but 1.5 m 2, 50 mg twice a day; and BSA 1.5 m 2, 60 mg twice a day. Dexamethasone and/or 5-HT 3 receptor antagonists were administered according to the treatment policy of each individual institution. Reductions of dose of docetaxel (only in arm A) and S-1 were specified per protocol, with a maximum of one dose reduction allowed. In addition, in arm B, one dose increment of docetaxel (50 mg/m 2 ) was specified in the protocol. Assessments of Toxicity and Response Toxicity was assessed and scored during treatment and at monthly visits thereafter according to the Common Terminology Criteria for Adverse Events version 3.0 (Japan Clinical Oncology Group/Japan Society of Clinical Oncology version). 17 Tumor response was measured and recorded using the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines version 1.0 18 at every treatment course and at monthly visits thereafter until disease progression. Confirmation of objective responses was required after at least 4 weeks. End Points and Statistical Considerations The primary end point of this randomized phase II study was objective response rate (ORR). Secondary end points included progression-free survival (PFS), overall survival (OS), and characterization of the toxicities of the two regimens. A total of 56 patients were required with a power of 80% and a 5% two-sided significance level for detection of an absolute 7% increase in response rate in the combined treatment arm, compared with the threshold value of 7% determined on the basis of a previous report. 4 Relative risk (RR) of ORRs and their 95% confidence intervals (CIs) were estimated using a stratified Cox regression model. Assuming that some dropouts would occur in each arm, 60 patients were considered needed in this study. PFS was defined as the period from the date of randomization to the first date of disease progression or of death from any cause. OS was defined as the period from the date of randomization to the date of death from any cause. Comparisons of PFS and OS between the arms were stratified by PS and previous docetaxel chemotherapy. Time-to-event distributions were calculated using the method of Kaplan and Meier, and differences between Copyright 2010 by the International Association for the Study of Lung Cancer 1431

Segawa et al. these distributions were assessed using the stratified log-rank test. Hazard ratios of PFS and OS between the arms were also estimated using a stratified Cox regression model. The 2 test or Fisher s exact probability test was used to compare proportions of categorical variables. P values less than 0.05 on two-tailed analyses were considered significant. Safety data for all patients receiving protocol treatment were tabulated and summarized descriptively. Compliance with S-1 treatment was assessed by comparing the amounts of prescribed and actually taken medication. Statistical analyses were performed using STATA version 10 (College Station, TX). RESULTS Baseline Patient Characteristics Between October 2005 and October 2008, a total of 60 Japanese patients with advanced NSCLC were enrolled at six institutions. Of those, 29 and 31 patients were randomly assigned to arms A and B, respectively. All the patients were evaluable for efficacy and safety, and the baseline characteristics were well balanced across the two arms (Table 1). Median off-treatment time since the last dose of first-line chemotherapy was 4.2 months (range, 1 45 months) in arm A and 3.7 months (range, 1 12 months) in arm B. Nine patients who had received first-line docetaxel-containing chemotherapy were also included in this study (six in arm A, three in arm B). These patients each had a TABLE 1. Baseline Patient Characteristics by Treatment Arm DOC (n 29) DOC S-1 (n 31) Median age (range) 63 (40 75) 63 (44 75) Gender Male 23 26 Female 6 5 ECOG performance status 0 12 10 1 17 21 Smoking history Never-smoker 6 8 Ever-smoker 23 23 Previous treatment with DOC Yes 6 3 No 23 28 Response to previous chemotherapy Partial response 15 14 Stable disease 13 16 Progressive disease 1 1 Time since previous chemotherapy 3 mo 11 11 3 mo 18 20 Disease stage III 6 10 IV 23 21 Histology Adenocarcinoma 21 19 Squamous 4 9 Others 4 3 DOC, docetaxel; ECOG, Eastern Cooperative Oncology Group. TABLE 2. Objective Tumor Response DOC S-1 DOC (n 29) (n 31) Partial response 6 5 Stable disease 13 14 Progressive disease 9 11 Nonevaluable 1 1 Response rate (95% confidence interval) 20.7% (8.0 39.7%) a 16.1% (5.5 33.7%) a a p 0.81 as estimated using a stratified Cox regression model. DOC, docetaxel. partial response (PR) to the first-line chemotherapy, which included a doublet of cisplatin and docetaxel (n 8) or a triplet of cisplatin, docetaxel, and irinotecan (n 1). In addition, activating mutations for epidermal growth factor receptor (EGFR) in tumor were detected in 6 of 21 patients examined (2 of 10 in arm A, 4 of 11 in arm B), although this information was obtained in the post hoc analysis. Treatment Delivered Thirty patients received at least four courses of assigned treatment (17 [58.6%] in arm A, 13 [41.9%] in arm B). The median number of courses of chemotherapy administered was 5 (range, 1 6) in arm A and 3 (range, 1 6) in arm B. Six patients (two in arm A, four in arm B) were withdrawn from protocol treatment because of adverse events, and two in arm B refused further treatment before four courses had been administered. In addition, nine patients (31.0%) had a reduction of dose of docetaxel in arm A and eight patients (25.8%) had reduction of dose of S-1 in arm B. One patient had an increase in dose of docetaxel in arm B. Compliance with S-1 was 92.7% over all treatment courses. Efficacy Evaluation Of the 60 patients entered in this randomized phase II study, 11 had a confirmed PR as best response (Table 2). Six of these patients received docetaxel alone (ORR, 20.7%; 95% CI, 8.0 39.7%), whereas five patients received a combination of docetaxel and S-1 (ORR, 16.1%; 95% CI, 5.5 33.7%). No difference in response rate was found between the two arms (RR, 1.18; 95% CI, 0.31 4.41; p 0.81). In addition, of nine patients who had previously received docetaxel, one patient each in the two arms achieved PR. At the time of analysis, the median follow-up time for all patients was 16.9 months (range, 2.4 39.5 months). Thirty-four (57%) patients had died during the follow-up period, mostly as a result of disease progression (n 32). Median PFS was 3.7 months (95% CI, 1.5 6.7 months) for patients assigned to arm A and 3.4 months (95% CI, 1.9 4.7 months) for those assigned to arm B (p 0.27, Figure 1). Median OS and 1-year survival rate were 22.9 months and 73.8% versus 8.7 months and 33.5% for arms A and B, respectively (p 0.02 for survival, Figure 2). In addition, hazard ratios between the two arms were 0.73 (95% CI, 0.40 1.31; p 0.29) for PFS and 0.42 (95% CI, 0.19 0.89; p 0.02) for OS, respectively. Safety Evaluation Of the 60 patients entered in this study, 59 (98.3%) experienced at least one adverse event. The Common Termi- 1432 Copyright 2010 by the International Association for the Study of Lung Cancer

Randomized Phase II Study in Relapsed NSCLC nology Criteria for Adverse Events grade 3 and 4 toxicities exceeding 5% of patients are listed in Table 3. Leukocytopenia (62.1% versus 54.8%), neutropenia (89.7% versus 64.5%), and febrile neutropenia (13.8% versus 3.2%) were more frequently observed in arm A, whereas anemia (0% versus 6.5%) and liver dysfunction (0% versus 6.5%) were more frequent in arm B. Low-grade oral mucositis and diarrhea were more common in arm B (data not shown). However, differences between the two arms in the incidence of none of the toxicities were significant. In addition, three patients (two in arm A, one in arm B) experienced pulmonary embolism; one of these events resulted in death in arm A. However, events were judged by the Data Safety Monitoring Committee unlikely to be related to study treatment. Poststudy Treatment Thirty-nine patients received at least one form of chemotherapy as poststudy treatment (18 [62.1%] in arm A, 21 [67.7%] in arm B). Twenty-six patients received treatment with EGFR tyrosine kinase inhibitor (EGFR TKI; 14 [48.3%] in arm A, 12 [38.7%] in arm B). PR and stable disease (SD) were observed in six and five patients versus two patients each in arms A and B, respectively. In addition, nine patients (31.0%) in arm A received subsequent single-agent S-1 or S-1 containing chemotherapy; PR was observed in one patient and SD in four. No patients in arm B received subsequent treatment with S-1. FIGURE 1. FIGURE 2. Progression-free survival by treatment arm. Overall survival by treatment arm. DISCUSSION In terms of ORR, the primary end point of this study, we failed to demonstrate additional effect of S-1 to docetaxel in second-line chemotherapy for patients with advanced NSCLC (20.7% versus 16.1%, p 0.81). We initially assumed the ORR of docetaxel to be approximately 7% based on a previous report published in 2000. 4 In contrast, and similar to our findings, docetaxel monotherapy produced higher ORR in a recent Japanese phase III study 19 than in a previous Western study. 4 This difference in ORR might be due to ethnic differences between the two studies and differences in criteria for assessment of response. In contrast, the low ORR in the combination arm compared with the single-agent docetaxel arm might have been due to the reduced dose of docetaxel used in this study. Indeed, two independent Japanese phase I studies of the combination of docetaxel and S-1 reported that the approved dose of docetaxel (60 mg/m 2 ) in Japan 20 could not be tolerated with the standard dose of S-1 (80 mg/m 2 ). 16,21 This seems to be one difficulty with concurrent administration of these agents. Although ORR and PFS were comparable in the two arms, OS in the single-agent arm was significantly longer than that in the combination arm (22.9 versus 8.7 months, p 0.02). The median OS demonstrated in the single-agent arm seems to TABLE 3. CTCAE Grade 3 Adverse Events Exceeding 5% of Patients DOC (n 29) DOC S-I (n 31) No. of Patients No. of Patients Grade 3 Grade 4 Total % Grade 3 Grade 4 Total % Leukocytopenia 15 3 18 62.1 16 1 17 54.8 Neutropenia 8 18 26 89.7 11 9 20 64.5 Anemia 0 0 0 0 2 0 2 6.5 Febrile neutropenia 4 0 4 13.8 1 0 1 3.2 Pulmonary embolism 0 2 a 2 6.9 0 1 1 3.2 Liver dysfunction 0 0 0 0 1 1 2 6.5 a One patient developed fatal pulmonary embolism (grade 5). CTCAE, Common Terminology Criteria for Adverse Events version 3.0; DOC, docetaxel. Copyright 2010 by the International Association for the Study of Lung Cancer 1433

Segawa et al. be extraordinarily prolonged compared with the results of previous studies (7.0 14.0 months). 4,6,7,19 This difference between OS and PFS is considered attributable to interactions resulting from poststudy treatment. Patients assigned to the single-agent docetaxel arm responded well to poststudy treatment with EGFR TKI (42.9% [6 of 14] versus 16.7% [2 of 12] in the single-agent and combination arms, respectively). In addition, it is quite interesting that, in the single-agent arm, almost all patients who obtained disease control (i.e., PR and SD) with poststudy treatment with S-1 did not respond to poststudy treatment with EGFR TKI (data not shown). Although it is difficult to understand this curious result, patient population benefited from treatment with S-1 might be different from those with EGFR TKI, because activating EGFR mutation in tumor had been reported to negatively influence the effect of adjuvant chemotherapy with uracil-tegafur. 22 These effects of poststudy treatment might have biased postrecurrent survival in favor of the single-agent docetaxel arm. A recent meta-analysis demonstrated that combination chemotherapy was more toxic than single-agent chemotherapy in the second-line setting. 23 However, the severity of toxicity in the combination arm is considered tolerable and comparable to that in the single-agent arm. In contrast, toxicity profiles differed somewhat between the two arms, although not to statistically significant extents. In addition, the toxicity profile of the combination arm was quite similar to those observed in a recent single-arm Japanese study. 21,24 The difference in toxicity profiles between our study and the meta-analysis seem to be attributable to the differences in agents used in these studies. Although we could not detect a clinically significant advantage of the combination of docetaxel and S-1 in secondline chemotherapy against NSCLC, it should be noted that this randomized phase II study was conducted in exploratory fashion in a small sample of patients. In addition, determination of mutations activating EGFR in tumor was not specified in the protocol. It is therefore important to interpret the results of this study with caution. In conclusion, we did not find docetaxel in combination with S-1 to be superior in efficacy for relapsed patients with NSCLC, suggesting that docetaxel monotherapy should continue to be considered standard therapy in this setting. This study also corroborates other study results indicating that combination chemotherapy in the second-line and later setting may be detrimental compared with single-agent therapy, potentially because of increased toxicity limiting the amount of treatment being able to be administered feasibly. 23,25 ACKNOWLEDGMENTS The authors thank all members of the OLCSG for participating in this study. REFERENCES 1. 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. 2. Hotta K, Matsuo K, Ueoka H, et al. Meta-analysis of randomized clinical trials comparing cisplatin to carboplatin in patients with advanced non-small-cell lung cancer. J Clin Oncol 2004;22:3852 3859. 3. Hotta K, Matsuo K. Long-standing debate on cisplatin- versus carboplatin-based chemotherapy in the treatment of advanced non-small-cell lung cancer. J Thorac Oncol 2007;2:96. 4. 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;18:2095 2103. 5. Shepherd FA, Pereira J, Ciuleanu TE, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353:123 132. 6. Hanna N, Shepherd FA, Fossella FV, et al. Randomized phase III trial of pemetrexed versus docetacel in patients with non-small-cell lung cancer previously treated with chemotherapy. J Clin Oncol 2004;22:1589 1597. 7. Kim ES, Hirsh V, Mok T, et al. Gefitinib versus docetaxel in previously treated non-small-cell lung cancer (INTEREST): a randomised phase III trial. Lancet 2008;372:1809 1818. 8. Shirasaka T, Shimamoto Y, Fukushima M. Inhibition by oxonic acid of gastrointestinal toxicity of 5-fluorouracil without loss of its antitumor activity in rats. Cancer Res 1993;53:4004 4009. 9. Kawahara M, Furuse K, Segawa Y, et al. Phase II study of S-1, a novel oral fluorouracil, in advanced non-small-cell lung cancer. Br J Cancer 2001;85:939 943. 10. Takeda Y, Yoshizaki Y, Nonaka Y, et al. Docetaxel alone or orally combined with 5-fluorouracil and its derivatives: effects on mouse mammary tumor cell line MM2 in vitro and in vivo. Anticancer Drugs 2001;12:691 698. 11. Wada Y, Yoshida K, Suzuki T, et al. Synergistic effects of docetaxel and S-1 by modulating the expression of metabolic enzymes of 5-fluorouracil in human gastric cancer cell lines. Int J Cancer 2006;119:783 791. 12. Suto A, Kubota T, Fukushima M, et al. Antitumor effect of combination of S-1 and docetaxel on the human breast cancer xenograft transplanted into SCID mice. Oncol Rep 2006;15:1517 1522. 13. Han JY, Lee DH, Kim HY, et al. A phase II study of a weekly docetaxel plus capecitabine for patients with advanced nonsmall cell lung carcinoma. Cancer 2003;98:1918 1924. 14. Kindwall-Keller T, Otterson GA, Young G, et al. Phase II evaluation of docetaxel-modulated capecitabine in previously treated patients with non-small cell lung cancer. Clin Cancer Res 2005;11:1870 1876. 15. O Shaughnessy A, Miles D, Vukelja S, et al. Superior survival with capecitabine plus docetaxel combination therapy in anthracycline-pretreated patients with advanced breast cancer: phase III trial results. J Clin Oncol 2002;20:2812 2823. 16. Yoshida K, Hirabayashi N, Takiyama W, et al. Phase I study of combination therapy with S-1 and docetaxel (TXT) for advanced or recurrent gastric cancer. Anticancer Res 2004;24:1843 1851. 17. Common Terminal Criteria for Adverse Events v3.0 [Japan Clinical Oncology Group Web site]. Available at: http://www.jcog.jp/doctor/ tool/ctcae.html. Accessed February 13, 2010. 18. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000;92:205 216. 19. Maruyama R, Nishiwaki Y, Tamura T, et al. Phase III study, V-15-32, of gefitinib versus docetaxel in previously treated Japanese patients with non-small-cell lung cancer. J Clin Oncol 2008;26:4244 4252. 20. Kunitoh H, Watanabe K, Ohnoshi T, et al. Phase II trial of docetaxel in previously untreated advanced non-small-cell lung cancer: a Japanese cooperative study. J Clin Oncol 1996;14:1649 1655. 21. Atagi S, Kawahara M, Kusunoki Y, et al. Phase I/II study of docetaxel and S-1 in patients with previously treated non-small cell lung cancer. J Thorac Oncol 2008;3:1012 1017. 22. Suehisa H, Toyooka S, Hotta K, et al. Epidermal growth factor receptor mutation status and adjuvant chemotherapy with uracil-tegafur for adenocarcinoma of the lung. J Clin Oncol 2007;25:3952 3957. 23. Di Maio, Chiodini P, Georgoulias V, et al. Meta-analysis of single-agent chemotherapy compared with combination chemotherapy as second-line treatment of advanced non-small-cell lung cancer. J Clin Oncol 2009; 27:1836 1843. 24. Yanagihara K, Yoshimura K, Niimi M, et al. Phase II study of S-1 and docetaxel for previously treated patients with locally advanced or metastatic non-small cell lung cancer [published online ahead of print January 13, 2010]. Cancer Chemother Pharmacol doi: 10.1007/s00280-009-1239-7. 25. Takeda K, Negoro S, Tamura T, et al. Phase III trial of docetaxel plus gemcitabine versus docetaxel in second-line treatment for non-small-cell lung cancer: results of a Japan Clinical Oncology Group trial (JCOG0104). Ann Oncol 2009;20:835 841. 1434 Copyright 2010 by the International Association for the Study of Lung Cancer