EGFR mutation lung cancerArticlestherapy and more than 1 month for oral tegafururacil therapy. Patients were not eligible if theyMedian(95%CI)progression-free survivalreceived previous drug therapy that had targeted EGFRhad a history of interstitial lung disease, severe dnyGefitinib=86)92 months(80-130)Cisplatin and (n=86) 6.3 months(5-87-8docetaxelallergy, aclive infection or other serious disease condilion,symptomatic brain metastases, poorly controlled pleuraleffusion, pericardial effusion or ascites necessitating860drainage, active double cancer, or severe hypersensitivityto drugs containing polysolvate 80 Patients in pregnancyp<00001or lactation, or whose participation in the trial was judgedto be inappropriate by the attending doctor, were noteligible. All patients provided written informed consentSludy approval was oblained from independent ethicscommittees at everyinstitution. The study was undertakenin accordance with the declaration of helsinkiNumber at riskProceduresGefitinib 86114320Patients were randomly assigned in a 11 ratio to receivedocelaxgefitinib(250 mg/day, administered orally), or docetaxel(60 mg/m2, administered intravenously over a 1 h periodBGefitinib (n=35) 13.7 months(7.2-20.5followed by cisplatin (80 mg/m2, administeredCisplatin and(n=36) 8.1 manths(6.2-10. 2)intravenously over a 90-min period), with adequatoctanehydration, in cycles of once every 21 days for three tosix cycles Treatment continued until progression of thedisease, development of unacceptable toxic effects, arequest by the patient lo discontinue treatment, seriousnon-compliance with the protocol, or completion of threep=0069to six chemotherapy cycles. Further therapy afterprogression of the disease was at the physiciandiscretion. The primary endpoint was progression-freesurvival. Secondary endpoints included overall survivalresponse rate. Tertiary endpoints were diseasecontrol rate, safety, and mutation-type-specific survival0Initially palienls were screened for EGFR mutation in aNumber at riskcentral laboratory at the Department of MolecularGefitinib 3580Diagnostics, Aichi Cancer Centre Hospital, Nagoya, JapanCisplatin and 366docetaxelThe exon 19 deletion mutation was screened by fragmentanalysis and the l858R point mutation was screened byGefi(n=51)8.4 onths(78-99)the cycleave method, as described previously followed byCisplatin and (n=50) 5.3 months(4-4-6-4)confirmation by direct sequencing. On Feb 16, 2008, thedocetaxelprotocolwas amended to allow outsourcing of egFrgenetic testing from each institution to commercial clinicallaboratories, either at SRl in Tokyo(direct sequencing)Mitsubishi Chemical medience in Tokyo (peptide nucleicacid-locked nucleic acid PCR clamp), or BML in Tokyop3nausea, myelosuppression, fatigue, and alopeciaNon-haematological toxicityOther potentially treatment-related toxicities includedRash*allergic reaction(one in gefitinib group, four in cisplatinolus degrALTA61243group, seven in the cisplatin plus docetaxel group). TwoDry skin30patients in the gefitinib group developed interstitial lungDiarrhoea47disease. There was one lrealmenl-relaled death in thegefitinib group due to interstitial lung disease; there wereParonychiano deaths in the cisplatin plus docetaxel group. ThereStomaticwere no other serious adverse eventsConstipationDiISCUSSIOnAlopecia670Our results show that first-line treatment with gefitinibSensory disturbance* 7conferred longer progression-free survival than treatmentHaematological toxicitywith cisplatin plus docetaxel in a molecularly definedleucocytopenia(ie, EGFR mutation positive) group of patients withNSCLCThrombacytopenlaNeutropenia8the Ipass study for patients with lung adenoLoma with no or former light smoking history theAnaemia33progression-free survival of patients treated with gefitinibALT=alanine aminotransterase. AST=aspartate aminotransferase. CTC=Nationalwas significantly longer. However, the curves crossed atCancer Institute Common Terminology Criteria. p<0.001.the 6-month timepoint (initially chemotherapy was better,Table 3: Adverse events occurring in more than 10%of either of thewhile gefitinib was better later). Molecular analysis fortreatment groups listed according to incidence in the gefitinib groupabout a third of the patients suggested that the benefit ofimbalance of mutation types was not likely to affect the gefitinib was limited to patients with EGFR mutationswith an HR of 0. 48(95%CI 0. 36-0 64)and that gefitinibinterpretation of the overall resultstreatment was detrimental for patients without mutationsThe objeclive response rale in the overall populaLion (HR 2.85). 5 This result might seem similar lo ourswith measurable disease (n=ll7) was 62.1%(36 of however, the primary objective of the IPASS study was to58 patients)in the gefitinib group and 32. 2%(19 of assess gefitinib treatment in clinically selected patients126www.thelancet.com/oncologyVol11feBruary2010ArticlesPatient groupMedian progression-free survival(months)Median overall survival (months)Gefitinib Chemotherapy HR(95% CiGefitinib ChemotheraNndomised-CAMPiJapanese, EGFR mutation35(023-052)27725bset analyses of the phase 3 trials for patients selected according to clinical backgroundsPassSEast Asian, light-non-smoker, adenocarcinoma 261 9.50.48(036064)First SigNAl33 Korean, non-smoker adenocarcinoma478.40.61(031-1276.5Phase 3 trials of patients selected according to EGFR mutation statusJapanese, EGFR mutation19410.40357(0.252-0507)280236WToG3405 Japanese, EGFR mutation17292630.489(0336-0710)Table 4: Recent clinical trials assessing EGFR mutations as predictors of efficacy of gefitinib compared with chemotherapyand not in molecularly selected patients, as was the case in subgroup, although a direct comparison between exon 19our trial. In this context, a HR ofo. 36(95%CI0 25-0. 51 )26 deletion and L858R in the gefitinib group was not donefor gefitinib compared with carboplatin plus paclitaxel in However, recent Japanese trials, including I-CAMp andpatients selected by EGFR mutation is highly relevant. this study, did not detect any difference. The reason forFurthermore, our pooled analyses based on individual this discrepancy is not clear, although it might bepatient data from seven Japanese phase 2 studies that attributable to ethnic differences or difference of egfrassessed prospectively the efficacy of gefitinib for patients Tki used between study populationswith EGFR mutations( I-CAMP study) "and the pooled Two patients in the gefitinib group (2.3%)developedanalysis of 1006 patients enrolled in a phase 3 trial of interstitial lung disease, one of whom died. This incidencegefitinibz7also showed similar progression-free survival of was low compared with previous Japanese reports of 4. 0%about 10 months for patients harbouring an EGFr(59 /1482)and 3. 5%(70/1976). Selecting patientsmutation who were Treated with gelilinb, while the median according lo EGFR mulalion slatus is expecled lo reduceprogression-free survival of patients treated with the risk of interstitial lung disease, because risk factors forchemotherapy was 60 months(table 4). These results interstitial lung disease include smoking, male sex, andand not the clinical background of patients, determines the presence of EGFR mutations. y negative predictors ofstrongly suggest that the presence of EGFR mutations, squamous histology, all of which areknowledge should lead to Our study indicates that EGFR genetic testing is feasiblemolecularly based, personalised treatment of lung cancer. and should be done when possible. Although patientsSince the median duration of each treatment was quite without EGFR mutations were not included in our study,diferent(165 days for geliLinib compared with 64 days for polential harm of first-line gefitinib therapy compared withchemotherapy), one interpretation might be that a chemotherapy for patients without EGFR mutation shownmaintenance effect of gefitinib therapy contributed to the in the iPASS2 and the First-sIGnaL33 study indicate thepositive progression-free survival outcome, at least in part. necessity of patient selection by EgFR mutationIndeed, the progression-free survival curves of both groups Clinical background might help identify patients whoin IPASS were initially similar, and then separate at about have a higher chance of carrying EGFR mutationsthe time that chemotherapy stops. However, this was not However, it should be noted that in a previous study, 9the case in our trial, especially in patients with stage eight of 37(22%)patients with lung adenocarcinomaIlIB/IV disease. Furthermore, the SATURN and the with a history of heavy smoking (>50 pack- years)FAST-ACT29 trials that tested maintenance erlotinib after harboured egfr mutationschemotherapy showed that progression-free survival (both In conclusion, gefitinib significantly prolonged thetrials)and overall survival (SaTURn) was prolonged. The progression-free survival of patients with NSClC whobenefit was much greater in patients with an EGFr carry EGFR mutations compared with cisplatin plusmutation than in those without it in the saturn trial. docetaxel. It is not yet known whether the prolongedAccording to analyses of five US and European clinical progression-free survival conferred by gefitinib willtrials that assessed first-line Tki treatment, 2 patients translate into prolonged overall survival; we will continuerith the exon 19 deletion have a significantly longer to carefully follow-up our patients to determine itsprogression-free and overall survival than patients with long-term effects. Considering the efficacy and toxicity ofL858R(30.8 vS 14.8 months; p<00001). A similar trend gefitinib, it is a reasonable option for the first-linewas shown in a recent Spanish study. In IPASS, the hr treatment of patients with activating EGFR mutationsfor progression-free survival for gefitinib versus Contributorchemotherapy was 038(95% CI 0. 25-0. 56) in the TM, SM, SN, TS, MS, NK, and KN were involved in the conception andsubgroup of patients with exon 19 deletions, and design of the study KN and MF supervised the study. TM,IO, TS,M0.55(95% CI 0. 35-0 87) in the L858R mutationKA NK MT HY KS, SK, ES, HS, and ST were involved in he Is. hT,www.thelancet.com/oncologyVol11February2010Articlesprovision of study material, patients, and data acquisition. TM, SM,YY, SN, 15 Takanc T, Ohe Y, Sakamoto H, et al. Epidermal growth factor receptorIO, JT, TH, NK, MT, HY, KS, ES, HS, ST, and Kn were involved in datagene mutations and increased capy numbers predict gefitinibanalysis and interpretation. SM was in charge of the statistical design of thesensitivity in patients with recurrent non-small-cell lung cancer.study. YY was in charge of EgFR gene testing at the central laboratory. AllJ Clin Oncol2005;23:6829-37authors were involved in writing the report and approved the final version. 16 Maruyama R, Nishiwaki Y, Tamura T,et al. Phase Ill study,V-15-32,ofgefitinib versus docetaxel in previously treated Japanese patients withConflicts ofinterestnon-small-cell lung cancer. Clin Oncol 2008: 26: 4244-52TM has received lecture fees from AstraZeneca, Chugai, and17 Thatcher N, Chang A, Parikh P, et al. Gefitinib plus best supportiveBochringer-Ingclhcim. SN has rcccivcd honoraria from AstraZencca andcare in previously treated patients with refractory advancedSanofi Aventis. ms has received honoraria from astraZeneca, ht hasnon-small-cell lung cancer results from a randomisedreceived honoraria from Astra Zeneca and Sanofi-Aventis. St hasplacebo-controlled, multicentre study (Iressa Survival Evaluation inreceived honoraria from AstraZeneca and Chugai kN has receivedung Cancer ). Lancet 2005: 366: 152/-37.lecture fees from Astrazeneca, Chugai, and Boehringer-Ingelheim18 Kim ES, Hirsh V, Mok T, et al. Geftinib versus docetaxel in previouslyMf has received lecture fees from AstraZeneca, Chugai, andtreated non-small-cell lung cancer(INTEREST): a randomised phaseBoehringer-Ingelheim. All other authors declared that they have noIII trial. Lancet 2008: 372: 1800-18conficts of interest.19 Yatabe Y,lida. Ilorio Y Kosaka T, Takahashi, Mitsudomit'A rapid, sensitive assay to detect EGFR mutation in small biopsyspecimens from lung cancer./ Mol Diagn 2006: 8: 335-41This study was supported by the West Japan Oncology Group, a non20 Nagai Y, Miyazawa H, Huqun, et al. Genetic heterogeneity of theprofit organisation supported by unrestricted donations from severalepidermal growth factor receptor in non-small cell lung cancer cellpharmaceutical companies. We thank patients and their families forlines revealed by a rapid and sensitive detection system, the peptideheir support and participation in this trial. We are also grateful to datanucleic acid-locked nucleic acid PCR clamp. Cancer Res 2005managers and other support staff of wjoG, especially ShinichiroNakamura and naomi oizumi21 Mast A, de Arruda M. Invader assay for single-nucleotideRefepolymorphism genotyping and gene copy number1 Jcmal A, Sicgcl R, Ward E, jct al. Canccr statistics, 2009CA Cancer Clin 2009: 59: 225-4922 Pocock SJ, Simon R. Sequential trcatment assignment with balancingSchiller jH, Harrington D, Belani CP,for prognostic factors in the controlled clinical trial. Biometrics 1975;omparison3:103-15chemotherapy regimens for advanced non-small-cell lung cancerN Engl J mec2002;346:92-9823 Eberhard DA, Johnson BE, Amler LC, et al. Mutations in theScagliotti G, Hanna N, Fossella F, et al. The differential efficacy ofepidermal growth factor receptor and in KRAS are predictive andpemetrexed according to NSCLC histology: a rcvicw of two phase Illprognostic indicators in patients with non-small-cell lung cancertreated with chemotherapy alone and in combination with erlotinibstudies. Oncologist 2009; 14: 253-63J Clin oncol2005:23:5900-094 Herbst RS, Heymach JV, Lippman SM. Lung cancerN Engl Med2008:359:1367-8024 Mok TS, Wu YL, Thongprasert S, et al. Gefitinib orcarboplatin paclitaxel in pulmonary adenocarcinoma. N EnglJ Med5 Fukuoka m. yano s giaccone g, et al, multi-institut2009;361:94757randomized phase Il trial of gefitinib for previously treated patients 25 Kobayashi K, Inouc A, Macmondo M, ct al. First-line gefitinib versusvith advanced non-small-cell lung canccr. Clin Oncol 200321:223746first-line chemotherapy by carboplatin(CBDCA)plus paclitaxel (TXL)cell lu(NSCLC) patients(pts)with EGFR6 Kris MG, Natale RB, Herbst RS, et al. Efficacy of gefitinib,kimutations: a phase Ill study(002 )by North East Japan Gefitinibinhibitor of the epidermal growth factor receptor tyrosineStudy Group. J Clin Oncol 2009; 27(suppl abstr 8016in synpturnatic patients with nonl- snall cell lung Carcerharmonisea randomized trial. JAMA 2003; 290: 2149-59sed Tripartite Guideline Statistical principles for7 Lynch Tl, Bcll DW, Sordella r, ct al. Activating mutations in thclinical trials. International Conference on Harmonisation E9 ExpertWorking Group. Stat Med 1999: 18: 1905-42epidermal growth factor receptor underlying responsiveness ofon-small-cell lung cancer to gefitinib. N Eng!J Med 200427 Kris M, Mok I, Kim E, Douillard JY, Fukucka M, Thatcher N50:2129-39Response and progression-free survival in 1006 patients with known8 Paez jG Janne PA, Lee jC, et al. EGFR mutations in lung Cancerindividuals with non-small cell lung cancer. Eur Cancer doo ib inEGFR mutation status in phase Ill randomized trials of gefitinicorrelation with clinical response to gefitinib therapy. Science 20047(suppl): abstr O-9003304:1497-50028 Cappuzzo F, Coudert BP, WierzbickiR, et al. Efficacy and safety of9 Kosaka T Yatabe y Endoh H. Kuwano H. Takahashi T. Mitsudomi Terlotinib as first-line maintenance in NSCLC followingmutations of the epidermal growth factor receptor gene in lungnon-progression with chemotherapy: results from the phase IIIcancer: biological and clinical inp ications. Cuncer Res 200464:891923SATURN Study. Thorac Oncol 2009; 4(suppl): abstr A2110 Tamura K, Okamoto L, Kashii T, et al. Multicentre prospective phase Il 29 Mok TS, Wu YT, Yu CJ, et al. Randomized, placebo-controlled, phaseII study of sequential erlotinib and chemotherapy as first-lintrial of gefitinib for advanced non-small cell lung cancer withtreatment for advanced non-small-cell lung cancer. J Clin Oncol 2009epidermal growth factor receptor mutations: results of the West Japan27:508087Thoracic Oncology Group trial(WTOGo403). BrJ Cancer 200898:907-14.30 Mok TS, To KF, Srimunimimit V, et al. Clinical outcomes of patientswith epidermal growth factor receptor(EGFR)mutations(Mut)in11 Morita S, Okamoto I, Kobayashi K, et al. Combined survival analysisIPASS IRESSATM Pan Asia Study. Thorac Oncol 2009of prospective clinical trials of gefitinib for non-small cell lung cancer(suppl): abstr D9.5with EgFR mutations. Clin Cancer Res 2009: 15: 4493-9812 Jackman DM, Miller VA. Cioffredi LA, et al. Impact of epidermal31 Kudoh S, Kato Il, Nishiwaki Y, et al. Interstitial lung disease inJapanese patients with lung cancer: a cohort and nested case-controlgrowth factor receptor and KRAS mutations on clinical outcomes indy. Am Respir Crit Care med 2008: 177: 1348-5previously untreated non-small cell lung cancer patients: results of an 32 Ando m okamoto l. Yamamoto N. et al. Predicitive factors foronline tumor registry of clinical trials. Clin Cancer res 200915:5267-73interstitial lung disease, anti-tumor response and survival innon-small cell lung cancer patients treated with gefitinib13 Roser, Moran T, Queralt C, et al. Screening for epidermal growthJ Clin Oncol2006;24:2549-56factor receptor mutations in lung cancer. N Engl Med 2009361:9586733 Lee JS, Park K, Kim S-W, et al. A randomized phase Ill study offitinib versus standard chemotherapy (gemcitabine plus cisplatin14 Mitsudomi T, Kosaka T, Endoh H, et al. Mutations of the epidermalas a first-line treatment for never smokers with advanced ornemctastatic adenocarcinoma of the lung. Thorac Oncol 20002812www.thelancet.com/oncologyVol11February2010