J Clin Oncol 22: by American Society of Clinical Oncology INTRODUCTION

Transcription

1 VOLUME 22 NUMBER 6 MARCH JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T From the Thoracic Oncology Service, Departments of Medicine, Biostatistics, Pathology, Surgery, and Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY. Submitted August 22, 2003; accepted January 7, Supported in part by AstraZeneca, Wilmington, DE. Presented in part at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31- June 3, 2003, and the 10th World Conference on Lung Cancer, Vancouver, British Columbia, Canada, Aug 10-14, Authors disclosures of potential conflicts of interest are found at the end of this article. Address reprint requests to Vincent A. Miller, MD, Thoracic Oncology Service, Department of Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; millerv@mskcc.org by American Society of Clinical Oncology X/04/ /$20.00 DOI: /JCO Bronchioloalveolar Pathologic Subtype and Smoking History Predict Sensitivity to Gefitinib in Advanced Non Small-Cell Lung Cancer Vincent A. Miller, Mark G. Kris, Neelam Shah, Jyoti Patel, Christopher Azzoli, Jorge Gomez, Lee M. Krug, William Pao, Naiyer Rizvi, Barbara Pizzo, Leslie Tyson, Ennapadam Venkatraman, Leah Ben-Porat, Natalie Memoli, Maureen Zakowski, Valerie Rusch, and Robert T. Heelan A B S T R A C T Purpose Gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, induces radiographic regressions and symptomatic improvement in patients with non small-cell lung cancer (NSCLC). Phase II trials suggested female sex and adenocarcinoma were associated with response. We undertook this analysis to identify additional clinical and pathologic features associated with sensitivity to gefitinib. Patients and Methods We reviewed medical records, pathologic material, and imaging studies of all 139 NSCLC patients treated on one of three consecutive studies of gefitinib monotherapy performed at our institution. We identified patients experiencing a major objective response and compared their clinical and pathologic features with the others. Univariate and multivariable analyses were performed on potential predictive features associated with sensitivity to gefitinib. Results Of 139 patients, 21 (15%; 95% CI, 9% to 21%), experienced a partial radiographic response. Variables identified as significant in univariate analysis included adenocarcinoma versus other NSCLC (19% v 0%; P.004), adenocarcinoma with bronchioloalveolar features versus other adenocarcinomas (38% v 14%; P.001), never smoker status versus former/current (36% v 8%; P.001), and Karnofsky performance status 80% versus 70% (22% v 8%; P.03). Multivariable analysis revealed the presence of adenocarcinoma with any bronchioloalveolar features (P.004) and being a never smoker (P.006) were independent predictors of response. Conclusion Our data suggest that individuals in whom gefitinib is efficacious are more likely to have adenocarcinomas of the bronchioloalveolar subtype and to be never smokers. These observations may provide clues to mechanisms determining sensitivity to this agent and suggest that NSCLC has a different biology in patients who never smoked and those with bronchioloalveolar carcinoma. J Clin Oncol 22: by American Society of Clinical Oncology INTRODUCTION The epidermal growth factor receptor (EGFR) and its family members, ligands, and downstream effectors have emerged as leading targets in treating non smallcell lung cancer (NSCLC). Activation of the EGFR by ligand binding or mutation stimulates cellular growth, proliferation, invasion, and metastasis and inhibits apoptosis [1]. Therefore, blockade of EGFRmediated effects should theoretically arrest the growth of NSCLCs driven by EGFR signaling. Gefitinib is a synthetic anilinoquinazoline that inhibits the EGFR tyrosine kinase in vitro with an concentration that inhibits 50% of to 0.33 mol/l, which is 100 times lower than the concentration required to block the kinases of HER-2, KDR, MEK-1, and MEK-2 [2]. Three phase I trials have been reported [3-5], and in two of these major objective responses were observed in patients with NSCLC refractory to multiple 1103

2 Miller et al prior chemotherapy regimens. Overall, 10% of NSCLC patients treated on these trials (10 of 100 patients) experienced partial responses that were often accompanied by prompt symptomatic improvement. The phase I data prompted two phase II monotherapy trials designed to determine the objective response rate and measure the frequency and magnitude of symptomatic benefit in patients with NSCLC previously treated with chemotherapy [6,7]. In these trials, radiographic regressions were observed in 14% of patients (61 of 426 patients; 95% CI, 11% to 18%) and symptomatic improvement in 39% of patients (165 of 426 patients; 95% CI, 34% to 43%). These results confirmed those suggested by the phase I trials of rapid and durable responses with clinically meaningful benefit for some patients. Although response rates were modest, these results compare favorably with those of cytotoxic agents commonly used in this setting [8,9]. As EGFR is present in essentially all NSCLC tumors and drug concentrations sufficient to inhibit EGFR tyrosine kinase activation are achieved in most patients, the phase II investigators sought to identify pretreatment clinical characteristics predictive of radiographic regressions. In terms of radiographic response, which in both phase II trials was a surrogate for symptomatic benefit, the most striking differences reported were noted by sex and histologic subtype of NSCLC. Responses were seen in 25% (95% CI, 18% to 32%) of women (39 of 154 patients) as compared with 8% (95% CI, 5% to 11%) of men (22 of 270 patients) and in 19% (95% CI, 15% to 24%) of patients with adenocarcinoma (53 of 274 patients) as compared with eight of 150 patients or 5% (95% CI, 2% to 9%) in other NSCLC histologies. Patients enrolled in Japan had higher radiographic response rates in the Iressa Dose Evaluation in Advanced Lung Cancer (IDEAL) 1 trial [6]. Thus to see if we could identify other pretreatment variables that predicted sensitivity to this agent, we undertook this retrospective review of Memorial Sloan-Kettering Cancer Center patients. We paid particular attention to cell type and smoking history, variables not addressed in the multicenter trials. PATIENTS AND METHODS Patients Between September 1998 and October 2002, 139 patients with NSCLC were treated with gefitinib monotherapy at our center. Eleven patients were treated on our initial phase I trial [3], six patients were treated on the IDEAL 2 study (comparing daily doses of 250 and 500 mg of gefitinib) [7], and 122 patients were treated on the Expanded Access Program, a compassionate use program that provided gefitinib preapproval for advanced NSCLC patients. All patients were cared for by a member of the Thoracic Oncology Service and had a diagnosis of NSCLC confirmed at our center. We retrospectively reviewed the records, pathology slides and reports, and imaging studies of these patients and then compared the characteristics of responders with those of the other patients. These groups of patients were analyzed for significance of stage, sex, age, Karnofsky performance status, cell type, bone metastasis, number of prior chemotherapy regimens, use of prior cisplatin or carboplatin, use of prior docetaxel, and smoking history [10]. Radiologic Evaluation For patients with major responses, we reviewed the imaging studies performed at the initiation of gefitinib, at the time response was first assessed, and thereafter while on gefitinib using standard criteria [11]. In all cases, follow-up imaging was performed not less than 1 month after initiation of the drug, and a confirmatory imaging study was obtained 1 month or more after the initial study that documented a partial response or improvement. All imaging was reviewed by one thoracic radiologist (R.T.H.). Nonresponders were identified by physician s progress notes or review of reports of hospital radiologic studies. Resource limitations did not permit centralized review of nonresponding patients. Pathologic Evaluation We reviewed pathology specimens on all available patients. These were examined by a thoracic cytopathologist (M.Z.) who was unaware of patient outcome and were reported as one of the following: adenocarcinoma, adenocarcinoma with bronchioloalveolar features, bronchioloalveolar carcinoma with focal invasion, pure bronchioloalveolar cancer, squamous cell cancer, large-cell carcinoma with or without neuroendocrine features, or NSCLC not specified [12]. For purposes of analysis and based on our prior work, bronchioloalveolar cell carcinoma and its variants were considered as one entity [12]. Statistical Methods We investigated the associations between sensitivity to gefitinib and patient characteristics. Known and potential prognostic factors explored included age, sex, Karnofsky performance status, presence of bone metastases, number of prior chemotherapy regimens, prior cisplatin, carboplatin or docetaxel use, and smoking history as reported in the patient s initial consultation note. The primary objective was to identify the factors associated with response to gefitinib. The associations between response and the covariates were assessed using Fisher s exact test or t test. Significant variables in the univariate analysis (P.05) were then included in an exact logistic regression model with gefitinib response as the dependent variable. Those variables that were significant in the univariate analysis (P.05) were included in a multivariable Cox regression model. Landmark survival analysis as described by Anderson et al [13] was performed at a 2-month time point when response assessment would commonly be undertaken. RESULTS One hundred thirty-nine patients with NSCLC received gefitinib on one of the three clinical trials. Ninety percent of patients received 250 mg daily; only two patients received doses less than 225 mg daily. The overall observed response rate was 15% (95% CI, 9% to 21%). Table 1 lists the patient demographics. Sixty-five percent of patients were women and 78% had adenocarcinoma. One quarter of patients had never smoked. Six (13%) of 48 males were never smokers, as opposed to 30 (33%) of 91 females (P.009). Ten patients had tumor specimens consistent with pure bronchioloal JOURNAL OF CLINICAL ONCOLOGY

3 Predictors of Sensitivity to Gefitinib in NSCLC Table 1. Demographics of Patients Treated With Gefitinib Variable Patients % Partial response Yes No Stage III 12 9 IV Sex Male Female Age, years Mean 64 Median 64 Range A Other A, BAC A, non-bac Other Bone metastases Yes No prior chemotherapy regimens Prior cisplatin or carboplatin Yes No Prior docetaxel Yes No KPS Smoking history Never Former/current Time from diagnosis to initiation of gefitinib therapy, months Median Range Patients All patients All metastatic/recurrent patients Adenocarcinoma with BAC features Adenocarcinoma with no BAC Other NSCLC Abbreviations: A, adenocarcinoma; BAC, bronchioloalveolar cell carcinoma; KPS, Karnofsky performance status; NSCLC, non-small-cell lung cancer. Numbers do not add up to 100% because of rounding to nearest integer. veolar cell carcinoma, and three of these individuals had radiographic regressions. Table 2 lists results of univariate analysis of the significance of potential prognostic factors Table 2. Univariate Analysis of Features Associated With Sensitivity to Gefitinib Variable Response (n 21) Patients % No Response (n 118) patients % Sex Male Female Age at diagnosis, years Mean Median Range A Other A, BAC A, non-bac Other Bone metastases Yes No prior chemotherapy regimens Prior cisplatin or carboplatin Yes No Prior docetaxel Yes No KPS Smoking history Never Former/current Abbreviations: A, adenocarcinoma; BAC, bronchioloalveolar cell carcinoma; KPS, Karnofsky performance status. for gefitinib sensitivity using Fisher s exact test and a t test for age. None of 38 patients with a NSCLC other than an adenocarcinoma experienced a radiographic regression (0% observed rate; 95% CI, 0% to 9%; P.001). When the adenocarcinoma group was divided into those tumors with and without elements of bronchioloalveolar cell carcinoma, the presence of bronchioloalveolar features was also found to be significant (38% v 14%; P.001). Thirteen of 36 never smokers experienced objective regressions as compared with eight of 104 current or former smokers (36% v 8%; P.001). The small number of current smokers precluded a separate analysis of that subgroup. Patient with Karnofsky performance status 80% were more likely to P

4 Miller et al Table 3. Multivariable Analysis of Features Associated With Sensitivity to Gefitinib Variable Odds Ratio 95% CI P A, BAC to not reached.01 A, non-bac to not reached Other 1.0 Smoking history Never to Former/current 1.0 KPS to Abbreviations: A, adenocarcinoma; BAC, bronchioloalveolar cell carcinoma; KPS, Karnofsky performance status. be sensitive to gefitinib than were those with Karnofsky performance status 70% (22% v 8%; P.03). When these three factors that were significant in univariate analysis were then tested in a multivariable analysis, the presence of bronchioloalveolar features in an adenocarcinoma (or pure bronchioloalveolar cell carcinoma) and a history of never having smoked remained significant (Table 3). Figure 1 displays survival as a function of gefitinib sensitivity or resistance in a landmark analysis of all patients alive at 2 months after commencing treatment. There is a Fig 1. Overall survival as a function of sensitivity to gefitinib. significant difference in survival between sensitive (median, 15 months; 95% CI, 12 months to not reached) and resistant patients (median, 6 months; 95% CI, 5 to 10 months; P.001). DISCUSSION The identification of the signal transduction inhibitor gefitinib as an active agent in heavily pretreated NSCLC patients was an unprecedented observation in the treatment of this disease. The phase II trials showed that use of this drug results in partial radiographic regressions in 14% of cases and that this is associated with symptomatic improvement in 39% of such individuals. Unlike the situation for other targeted therapies such as trastuzumab (Genentech, South San Francisco, CA) in breast cancer where a molecular screen to test for HER-2 gene amplification is standard, no such test has been established for gefitinib (or any EGFR tyrosine kinase inhibitor) in NSCLC [14]. Thus at present, this targeted therapy is given to patients with advanced NSCLC, regardless of whether EGFR signaling is a critical driver of their tumor s growth. We and others have sought to identify features that would permit us to select patients in whom this drug would be beneficial. Efforts on this front have lagged for a number of reasons, including but not limited to the following: lack of availability of adequate tissue samples to perform even the simplest immunohistochemical studies (eg, staining for the presence/absence of EGFR protein) in a majority of patients; frequent lack of readily accessible sites of tumor for rebiopsy of lesions to assess changes in putative molecular markers (eg, a decrease in phosphorylated EGFR, the basally activated form of this receptor in responding patients); the failure to standardize and validate a system for grading expression or overexpression of EGFR in tumor tissue; the complexity of the EGFR signaling pathway; and the molecular complexity of NSCLC as a result of the multiplicity of oncogenes and tumor suppressor genes affected by tobacco. For example, in the IDEAL 2 study, performed by a number of academic institutions in the United States, pathology slides deemed adequate for performing an EGFR immunohistochemical stain were available on only 70 (32%)of 216 patients. In the recently reported Iressa NSCLC Trial Assessing Combination Treatment studies [15,16], phase III trials of cytotoxic chemotherapy with or without gefitinib, approximately 400 samples from 1,100 patients were available from each study. Although conceivably important information could emerge from these correlative studies, skepticism is raised given the challenges discussed above. Given the lack of a ready approach to identify a molecular fingerprint associated with response, we revisited clinical data in the hopes of developing a profile of such pa JOURNAL OF CLINICAL ONCOLOGY

5 Predictors of Sensitivity to Gefitinib in NSCLC tients. A potential advantage of our approach is the relative homogeneity of the clinical, radiographic, and pathologic assessment. All patients were treated by one of several medical oncologists, scans were reviewed by one reference radiologist, and pathology specimens were centrally reviewed by one reference cytopathologist. As a single study site, we could meticulously review the electronic medical records in a manner not feasible in a multicenter trial. Any retrospective review raises several questions. Was our population similar to that of the prospective trials? In comparison with patients treated on IDEAL 2, two exceptions are notable. The first is that approximately 65% of our patients were women, compared with only 36% in the phase II studies. The other difference is that some of our patients had received no or one prior chemotherapy regimen, whereas two or more prior regimens were a prerequisite for entry to IDEAL 2. No study to date has suggested that the number of prior chemotherapy regimens influences likelihood of response. Both IDEAL 1 and 2 identified female sex as a baseline characteristic possibly associated with response. We believe the large proportion of women treated at our center results from both a preferential practice pattern favoring treatment of female patients with this agent and the fact that our patient population at large currently has more women than men. A second potential cause of bias could be in the pathologic review. However, no responding patient was reclassified after a blinded review by the reference pathologist (M.Z.). Among the 118 nonresponders, 12 patients had no diagnostic material retained at our institution. Of the remaining 106 patients, 20 patients underwent reclassification. Twelve patients previously classified as having NSCLC not specified, were categorized as adenocarcinoma (10 patients), squamous cell (one patient), or large-cell neuroendocrine carcinomas (one patient). Two patients previously classified as having adenocarcinoma were considered to have bronchioloalveolar features, whereas two patients who were previously considered to have bronchioloalveolar cell carcinoma or variants thereof were classified as having pure adenocarcinomas. One patient each was recategorized from adenocarcinoma to large-cell neuroendocrine carcinoma, squamous cell to NSCLC not specified, adenocarcinoma to large-cell carcinoma, and adenocarcinoma to NSCLC not specified. Furthermore, the smoking histories obtained may have been inconsistently defined. Never smokers are defined as individuals having a lifetime exposure of fewer than 100 cigarettes [17]. However, the social histories reported by members of our group may have been inconsistent in use of that definition, because no structured smoking questionnaire was used. Efforts to confirm documentation of never smoking status were not always possible. This, as well as the predilection of adenocarcinomas treated, might explain the disproportionately high percentage of never smokers reported herein. On the basis of our observations, future studies of EGFR tyrosine kinase inhibitors should include a validated instrument to record the smoking history. However, a recent review of subgroups of patients treated on IDEAL 2 revealed strikingly similar results to those reported here [18]. In that analysis, 29% (95% CI, 15% to 46%) of 34 nonsmokers responded to gefitinib, as opposed to 4.6% (95% CI, 2% to 11%) of 108 smokers. With these caveats in mind, two new findings emerge from this study. First, the previously reported association of response with adenocarcinoma cell type is likely caused by those cases containing features of bronchioloalveolar carcinoma. Second, smoking status (never v current or former) emerges as a powerful and independent predictor of response. The previously described association of response and female sex was again observed but was not significant in our retrospective analysis, perhaps as a result of comigration, with smoking as a variable not previously reported in IDEAL 1 and 2. These observations raise several questions. Why should never smokers and individuals whose tumors contain bronchioloalveolar features be more likely sensitive to gefitinib? First, lung carcinomas in smokers and never smokers may arise via distinct pathogenetic mechanisms. Genetic analysis of lung carcinomas has revealed that widespread chromosomal abnormalities are frequent in lung adenocarcinoma from smokers but infrequent in such tumors arising in never smokers [19-21]. The differences in smokers are most likely due to direct repeated exposure to tobaccorelated carcinogens such as benzo[a]pyrene and N-nitrosamines, which are known to induce widespread genetic damage. For example, amplification of PI3-K, an antiapoptotic protein downstream of EGFR, occurs in association with smoking in squamous cell carcinomas of the head and neck and is associated with cisplatin resistance in that disease [22]. Conceivably, this aberration downstream of EGFR could also explain the somewhat paradoxical resistance of squamous cell carcinomas that characteristically express more EGFR than adenocarcinomas [23]. A prospective assessment of the frequency of this amplification in newly resected snap-frozen squamous and nonsquamous cell carcinomas of the lung is ongoing. Second, the exact molecular lesions found in never smokers may be different from those found in smokers. For example, one of the most well-characterized cigarette smoke associated genetic mutations involves G to T transversions of the proto-oncogene K-RAS, a downstream signaling component of the EGFR activation cascade [24]. However, activating K-RAS mutations are rarely found in never smokers with primary adenocarcinomas [25]. Moreover, in a study of patients with bronchioloalveolar cell carcinomas, only two of 20 tumors were found to have K-ras mutations; both tumors were derived from patients with significant smoking histories [26]. Thus one would

6 Miller et al speculate that never smokers in our study who responded to gefitinib lack K-ras mutations. Such K-ras lesions might actually confer resistance to gefitinib by activating signaling pathways downstream of EGFR. Because tumors in never smokers seem to be less genetically complex than their counterparts in smokers, they may be more dependent on signaling through one or only a few critical signaling pathways for tumor maintenance and survival. This could explain why gefitinib seems to induce tumor regression so rapidly in those patients that do respond, a mechanism that could be seen as analogous to inhibition by imatinib of specific activated tyrosine kinases in patients with gastrointestinal stromal tumors [27,28]. In this regard, further molecular characterization of gefitinibresponding tumors may uncover possible targets of the drug other than EGFR. Preclinically, Moasser et al [29] characterized the expression of HER family members in a broad panel of cell lines simultaneously tested for sensitivity to gefitinib and found that expression of HER-2 was the best predictor of growth inhibition, whereas EGFR expression was not correlated with the efficacy of gefitinib. Concentrations of gefitinib able to effectively inhibit the HER-2 tyrosine kinase are commonly achieved in the clinic. Intriguingly, adenocarcinomas of the lung are more likely to have 2 or 3 immunohistochemical staining for HER-2 than squamous cell or other NSCLC histologies [30]. More recently, Hirsch et al have looked at expression of EGFR and HER-2 in bronchioloalveolar cell carcinomas and reported that approximately one third have substantial simultaneous EGFR and HER-2 expression (F. Hirsch, personal communication, August, 2003). Thus one might hypothesize that this subset of tumors could be more effectively targeted by gefitinib than other NSCLCs. In conclusion, sensitivity to the EGFR tyrosine kinase inhibitor gefitinib is associated with tumors with bronchioloalveolar features and with never smoking patients. Based on these observations, we plan to construct a tissue array of tumor from sensitive and resistant patients to study the EGFR pathway and genes that are frequently affected by tobacco abuse. In this way, we hope to identify for prospective testing a molecular phenotype of a patient who is likely to respond to gefitinib. Acknowledgment We thank Samir Hattar in facilitating the central pathology review. Authors Disclosures of Potential Conflicts of Interest The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Acted as a consultant within the last 2 years: Vincent A. Miller, AstraZeneca, Glaxo Smith Kline; Mark G. Kris, Glaxo Smith Kline, OSI Pharm; Lee M. Krug, AstraZeneca. Received more than $2,000 per year from a company for either of the last 2 years: Vincent A. Miller, AstraZeneca; Mark G. Kris, Glaxo Smith Kline, OSI Pharm; Lee M. Krug, AstraZeneca; Barbara Pizzo, AstraZeneca. REFERENCES 1. Mendelsohn J: Blockade of receptors for growth factors: An anticancer therapy The fourth annual Joseph H. Burchenal American Association for Cancer Research Clinical Research Award Lecture. Clin Cancer Res 6: , Wakeling AE, Guy SP, Woodburn JR, et al: ZD1839 (Iressa): An orally active inhibitor of epidermal growth factor signaling with potential for cancer therapy. Cancer Res 62: , Ranson M, Hammond LA, Ferry D, et al: ZD1839, a selective oral epidermal growth factor receptor-tyrosine kinase inhibitor, is well tolerated and active in patients with solid, malignant tumors: Results of a phase I trial. J Clin Oncol 20: , Baselga J, Rischin D, Ranson M, et al: Phase I safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types. J Clin Oncol 20: , Herbst RS, Maddox A-M, Rothenberg ML, et al: Selective oral epidermal growth factor receptor tyrosine kinase inhibitor ZD1839 is generally well-tolerated and has activity in non small-cell lung cancer and other solid tumors: Results of a phase I trial. J Clin Oncol 20: , Fukuoka M, Yano S, Giaccone G, et al: Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non small-cell lung cancer. J Clin Oncol 21: , Kris MG, Natale RB, Herbst RS, et al: Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: A randomized trial. JAMA 290: , Fossella FV, DeVore R, Kerr RN, et al: Randomized phase III trial of docetaxel versus vinorelbine or ifosfamide in patients with advanced non small lung cancer previously treated with platinum-containing chemotherapy regimens. J Clin Oncol 18: , Shepherd FA, Dancey J, Ramlau R, et al: Prospective randomized trial of docetaxel versus best supportive care in patients with non smallcell lung cancer previously treated with platinumbased chemotherapy. J Clin Oncol 18: , O Connell J, Kris MG, Gralla RJ, et al: Frequency and prognostic importance of pretreatment clinical characteristics in patients with advanced non small cell lung cancer treated with combination chemotherapy. J Clin Oncol 4: , Green S, Weiss GR: Southwest Oncology Group standard response criteria, endpoint definitions and toxicity criteria. Invest New Drugs 10: , Ebright M, Zakowski M, Martin J, et al: Clinical pattern and pathological stage but not histological features predict outcomes for bronchioloalveolar carcinoma (BAC). Ann Thorac Surg 74: , Anderson JR, Cain KC, Gelber RD: Analysis of survival by tumor response. J Clin Oncol 1: , Slamon DJ, Leyland-Jones B, Shak S, et al: Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344: , Giaccone G, Johnson D, Manegold C, et al: A phase III clinical trial of ZD1839 ( Iressa ) in combination with gemcitabine and cisplatin in chemotherapy-naive patients with advanced non-small cell lung cancer (INTACT 1). Ann Oncol 13:2-3, Johnson DH, Herbst R, Giaccone G, et al: ZD1839 ( Iressa ) in combination with paclitaxel and carboplatin in chemotherapy-naive patients with advanced non-small cell lung cancer 1108 JOURNAL OF CLINICAL ONCOLOGY

7 Predictors of Sensitivity to Gefitinib in NSCLC (NSCLC): Results from a phase III clinical trial (INTACT 2). Ann Oncol 13: , Giovino GA: Epidemiology of tobacco use in the United States. Oncogene 21: , Cohen MH, Williams GA, Sridhara R, et al: FDA drug approval summary: Gefitinib (ZD1839) (Iressa) tablets. Oncologist 8: , Sanchez-Cespedes M, Ahrendt SA, Piantadosi S, et al: Chromosomal alterations in lung adenocarcinoma from smokers and nonsmokers. Cancer Res 61: , Wong MP, Fung LF, Wang E, et al: Chromosomal aberrations of primary lung adenocarcinomas in nonsmokers. Cancer 97: , Wistuba II, Lam S, Behrens C, et al: Molecular damage in the bronchial epithelium of current and former smokers. J Natl Cancer Inst 89: , Singh B, Reddy PG, Goberdhan A, et al: p53 regulates cell survival by inhibiting PIK3CA in squamous cell carcinomas. Genes Dev 16: , Rusch V, Klimstra D, Venkatraman E, et al: Overexpression of the epidermal growth factor receptor and its ligand transforming growth factor alpha is frequent in resectable non-small cell lung cancer but does not predict tumor progression. Clin Cancer Res 3: , Arteaga CL: The epidermal growth factor receptor: From mutant oncogene in nonhuman cancers to therapeutic target in human neoplasia. J Clin Oncol 19:32S-40S, 2001 (suppl) 25. Ahrendt SA, Decker PA, Alawi EA, et al: Cigarette smoking is strongly associated with mutation of the K-ras gene in patients with primary adenocarcinoma of the lung. Cancer 92: , Rusch VW, Reuter VE, Kris MG, et al: Ras oncogene point mutation: An infrequent event in bronchioloalveolar cancer. J Thorac Cardiovasc Surg 104: , Heinrich MC, Corless CL, Duensing A, et al: PDGFRA activating mutations in gastrointestinal stromal tumors. Science 299: , Demetri GD: Targeting c-kit mutations in solid tumors: Scientific rationale and novel therapeutic options. Semin Oncol 28:19-26, Moasser M, Basso A, Averbuch S, et al: The tyrosine kinase inhibitor ZD1839 ( Iressa ) inhibits HER2-driven signaling and suppresses the growth of HER2-overexpressing tumor cells. Cancer Res 61: , Krug LM, Miller VA, Crapanzano J, et al: Randomized phase II trial of trastuzumab (tras) plus either weekly docetaxel (doc) or paclitaxel (pac) in previously untreated advanced non-small cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 20:333a, 2001 (abstr 1328)