Epidermal Growth Factor Receptor Gene Mutations and Increased Copy Numbers Predict Gefitinib Sensitivity in Patients With Recurrent Non–Smal
http://www.100md.com
《临床肿瘤学》
the Divisions of Internal Medicine and Diagnostic Radiology and Clinical Laboratory Division, National Cancer Center Hospital
Genetics and Pathology Divisions, National Cancer Center Research Institute
Statistics and Cancer Control Division, Research Center for Cancer Prevention and Screening, National Cancer Center, Tokyo, Japan
ABSTRACT
PURPOSE: To evaluate epidermal growth factor receptor (EGFR) mutations and copy number as predictors of clinical outcome in patients with non–small-cell lung cancer (NSCLC) receiving gefitinib.
PATIENTS AND METHODS: Sixty-six patients with NSCLC who experienced relapse after surgery and received gefitinib were included. Direct sequencing of exons 18 to 24 of EGFR and exons 18 to 24 of ERBB2 was performed using DNA extracted from surgical specimens. Pyrosequencing and quantitative real-time polymerase chain reaction were performed to analyze the allelic pattern and copy number of EGFR.
RESULTS: Thirty-nine patients (59%) had EGFR mutations; 20 patients had deletional mutations in exon 19, 17 patients had missense mutations (L858R) in exon 21, and two patients had missense mutations (G719S or G719C) in exon 18. No mutations were identified in ERBB2. Response rate (82% [32 of 39 patients] v 11% [three of 27 patients]; P < .0001), time to progression (TTP; median, 12.6 v 1.7 months; P < .0001), and overall survival (median, 20.4 v 6.9 months; P = .0001) were significantly better in patients with EGFR mutations than in patients with wild-type EGFR. Increased EGFR copy numbers ( 3/cell) were observed in 29 patients (44%) and were significantly associated with a higher response rate (72% [21 of 29 patients] v 38% [14 of 37 patients]; P = .005) and a longer TTP (median, 9.4 v 2.6 months; P = .038). High EGFR copy numbers ( 6/cell) were caused by selective amplification of mutant alleles.
CONCLUSION: EGFR mutations and increased copy numbers were significantly associated with better clinical outcome in gefitinib-treated NSCLC patients.
INTRODUCTION
The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase of the ErbB family that has been implicated in cell proliferation and survival and is frequently overexpressed in many solid tumors, including non–small-cell lung cancer (NSCLC). Gefitinib (Iressa; AstraZeneca, Osaka, Japan) is an orally active, selective EGFR tyrosine kinase inhibitor that binds to the adenosine triphosphate–binding pocket of the EGFR kinase domain and blocks downstream signaling pathways. Two phase II studies, IRESSA Dose Evaluation in Advanced Lung Cancer 1 and 2 (IDEAL 1 and 2), have demonstrated that gefitinib monotherapy exerts an antitumor activity in patients with advanced NSCLC who had previously received platinum-based chemotherapy.1,2 Gefitinib was approved in Japan for the treatment of inoperable or recurrent NSCLC in July 2002.
The IDEAL trials and retrospective studies have revealed that women, never smokers, patients with adenocarcinoma, and Japanese patients have higher response rates to gefitinib.1-4 Among patients with adenocarcinoma, histologic subtypes have been studied; one study showed that responses were more frequent in patients with bronchioloalveolar carcinoma (BAC) features (38% v 14%; P < .001),3 whereas another study showed that the response rate was higher in patients with a papillary-dominant subtype (76% v 21%; P = .002).5
Although no predictive molecular markers had been identified at the time of approval, somatic mutations in the kinase domain of EGFR have been subsequently linked to gefitinib sensitivity. According to three initial reports, 20 of 24 gefitinib-responsive tumors contained EGFR mutations, whereas 19 nonresponsive tumors did not contain any mutations.6-8 The mutations were detected in exons 18 to 21 of EGFR, close to the region coding the adenosine triphosphate–binding pocket of the kinase domain, and most of them were observed in two hotspots: in-frame deletions including amino acids at codons 747 to 749 in exon 19 and an amino acid substitution at codon 858 (L858R) in exon 21. Analyses of surgically resected NSCLC tumors revealed that such mutations were more frequent among women, never smokers, patients with adenocarcinoma, and Japanese or East Asian patients,7-13 consistent with the known clinical predictors of gefitinib sensitivity.
To evaluate the exact predictive value, we studied consecutive patients with recurrent NSCLC who received gefitinib therapy. To insure high-quality genetic analyses of the archived tissues, we used methanol-fixed, paraffin-embedded surgical specimens, which are known to preserve DNA better than formalin-fixed tissues,14 and performed laser capture microdissection (LCM).
Recently, some other biomarkers of NSCLC have been studied. The EGFR and chromosome 7 copy numbers in NSCLC were assessed using fluorescence in situ hybridization (FISH), and more than 3.0 EGFR copies per cell (balanced polysomy or gene amplification) were detected in 39 (22%) of 183 patients.15 A correlation between an increased EGFR copy number and gefitinib sensitivity was also proposed in another study.16 In yet other studies, mutations in the kinase domain of ERBB2 (HER2), a gene coding another receptor tyrosine kinase of the ErbB family, were detected in 16 (3.6%) of 445 patients with lung adenocarcinoma.17,18 In the current study, we also analyzed the EGFR copy number and the presence of ERBB2 mutations to assess their impact on clinical outcome.
The expression of EGFR and related proteins has been more widely studied using immunohistochemistry. Some studies suggested that high expression of phosphorylated Akt19,20 or low expression of phosphorylated mitogen-activated protein kinase20,21 was associated with better outcome in gefitinib-treated patients, but in general, methods, criteria, and results were inconsistent among studies. We thought that protein expression should be analyzed in another exploratory study, and in the current study, we focused on the genetic analyses.
PATIENTS AND METHODS
Patients
After searching the pharmaceutical records of the National Cancer Center Hospital, 279 patients with NSCLC who had begun receiving gefitinib monotherapy (250 mg/d) between July 2002 and May 2004 were identified. Seventy-three of these patients had undergone surgical resection of primary NSCLC at the hospital and subsequently relapsed. Recurrences were not necessarily confirmed pathologically but were diagnosed clinically. Seven patients were ineligible for inclusion in this study because methanol-fixed tissues were not available (n = 5) or their informed consent to the genetic analysis was not obtained (n = 2); consequently, 66 patients were included.
Genetic Analyses of EGFR and ERBB2
On a protocol approved by the institutional review board of the National Cancer Center, we performed mutational analyses of exons 18 to 24 of EGFR and exons 18 to 24 of ERBB2 and analyzed the EGFR copy number. Methanol-fixed, paraffin-embedded surgical specimens of primary NSCLC were collected retrospectively, and DNA was extracted from bulk tumor tissue, laser capture microdissected tumor tissue, and normal lung tissue from each patient. LCM was performed using a PixCell II LCM system (Arcturus Engineering Inc, Mountain View, CA) according to a previously described method.22 If appropriate, tumor cells were captured separately from two areas with different histologic subtypes, such as an area with a BAC subtype and another area with stromal invasion. Nested polymerase chain reaction (PCR) was performed to amplify exons 18 through 24 of EGFR using previously described primers,6 and standard PCR was used to amplify exons 18 through 24 of ERBB2. Direct sequencing of the PCR products was performed using ABI PRISM 3700 and 3100 DNA Sequencers (Applied Biosystems, Foster City, CA). All sequencing reactions were performed in both forward and reverse directions, and single nucleotide substitutions, insertions, and deletions were detected using an application program named NAMIHEI.23 Pyrosequencing was performed to verify the sequencing data of the hotspots of EGFR and to assess the proportion of mutant alleles in the laser-captured tumor cells using a Pyrosequencing PSQ 96MA (Pyrosequencing, Uppsala, Sweden).24 On the basis of the proportion of mutant alleles, EGFR mutations were divided into two patterns: balanced heterozygous (BH) pattern (< 60%) and mutant-allele-dominant (MD) pattern ( 60%). The cutoff level of 60% was decided because if more than 60%, the superiority of the mutant over the wild-type sequences was obvious on the direct sequencing chromatograms. Quantitative, real-time, TaqMan duplex PCR was performed to analyze the EGFR copy number using an ABI PRISM 7000 Sequence Detection System (Applied Biosystems). The EGFR primers were 5'-GGAGGACCGTCGCTTGGT-3' and 5'-AACACCGCAGCATGTCAAGA-3'; the probe (5'-CACCGCGACCTGGCAGCCA-3') was labeled with the reporter dye 6-carboxyfluorescein (FAM). RNaseP was coamplified in the same reaction mixture as the endogenous reference gene using TaqMan RNaseP Control Reagents (6-carboxyrhodamine [VIC] dye; Applied Biosystems). The average EGFR copy number per cell was calculated from the differences in the threshold amplification cycles between EGFR and RNaseP. Peripheral-blood samples obtained from healthy volunteers were analyzed as normal controls. Decreased, normal, moderately increased, and highly increased EGFR copy numbers were defined as less than 1.5, 1.5 to 3.0, 3.0 to 6.0, and 6.0 copies per cell, respectively.
Pathologic Evaluation
We reviewed the histologic features of the 66 patients using hematoxylin and eosin–stained slides of tumor samples. Two board-certified pathologists (K.T. and Y.M.) who were unaware of the patients' outcome and mutational status examined all the specimens independently; in case of discrepancy, final diagnoses were established by consensus. Adenocarcinoma was categorized in two ways. The first categorization was based on the WHO's classification of lung tumors,25 which includes four major subtypes of adenocarcinoma: papillary, acinar, BAC, and solid; the dominant subtype in the total tumor mass of each case was documented. The second categorization was based on a report from the Memorial Sloan-Kettering Cancer Center,26 in which adenocarcinomas were classified into adenocarcinoma without BAC features (Ad), adenocarcinoma with BAC features (AwBF), BAC with focal invasion (BwFI), and pure BAC (PBAC). If two or more tumors were present in one patient, the diagnosis of the most invasive tumor in each case was documented.
Radiologic Evaluation
In patients who had measurable lesions, imaging studies were performed at baseline, approximately 4 weeks after the initiation of gefitinib treatment, and periodically thereafter throughout the treatment. One board-certified radiologist (U.T.) who was unaware of the patients' mutational status reviewed the baseline, first follow-up, and confirmatory imaging studies and classified the tumor responses into complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) using standard bidimensional measurements.27 Responders were defined as patients with CR or PR. In this study, SD was subdivided into minor response (MR) and no response. MR was defined as a 25% decrease in the sum of the products of the perpendicular diameters of all measurable lesions at any point during gefitinib treatment. Time to progression (TTP) was defined as the time from the start of gefitinib administration to confirmed disease progression or death.
Statistical Analysis
The associations among mutational status, EGFR copy number, patient characteristics, and tumor response to gefitinib were assessed using a 2 test. The differences in TTP and overall survival (OS) according to the patient subgroups were compared using Kaplan-Meier curves and log-rank tests. Multivariate analyses using logistic regression models and Cox proportional hazard models were performed to assess the association between the biomarkers and clinical outcome while adjusting for the baseline patient characteristics. All analyses were performed using the SPSS statistical package (SPSS version 11.0 for Windows; SPSS Inc, Chicago, IL).
RESULTS
Patient Characteristics
The patient characteristics are listed in Table 1. All of the patients were Japanese. The proportions of women (39%), never smokers (47%), and patients with adenocarcinoma (94%) in this study were higher than those in a database of more than 1,000 patients with advanced or recurrent NSCLC treated at our hospital during the four most recent years (27%, 27%, and 73%, respectively). Twenty-two patients (33%) had been included in our phase II trial for first-line gefitinib therapy for patients with recurrent NSCLC, and the others had been treated with gefitinib in clinical practice settings. The operations for primary NSCLC were performed between February 1994 and August 2003, and the median time from the operations to the start of gefitinib was 2.3 years (range, 0.6 to 9.1 years).
Clinical Outcome
Sixty-four patients had measurable lesions at the start of gefitinib administration. CR and PR were observed in two and 32 patients, respectively. MR was observed in three of nine patients with SD. Twenty-one patients had PD, including six patients who died before the first follow-up imaging studies. Two patients had only unmeasurable bone lesions at baseline; one patient showed rapid symptom improvement and continued to receive gefitinib therapy without progression for 13.8+ months, whereas the other patient developed new lesions and died on day 71. These patients were included in the analysis as a responder and a nonresponder, respectively. The overall response rate was 53%. Forty-one patients died, and the median follow-up time for the 25 survivors was 14.6 months (range, 10.3 to 32.3 months). Eleven patients were still receiving gefitinib without progression at the time of the analysis. The median TTP and the median survival time (MST) for all patients were 5.2 and 16.3 months, respectively.
EGFR and ERBB2 Mutations
Forty-three mutations in the EGFR tyrosine kinase domain were detected in 39 (59%) of the 66 patients. All the mutations detected in this study are shown in Table 2. Twenty patients had deletional mutations in exon 19, and 17 patients had missense mutations (L858R) in exon 21. In exons 18 and 20, five types of missense mutations were detected. Two of them (G719S and G719C) occurred at a codon considered to be a third hotspot.6,7,9-12 The others (L703V, E709K, and S768I) were detected in patients who also had mutations at the hotspots. Because these mutations were not detected in the normal lung tissues from the same patients, they were considered to be somatic mutations. No somatic mutations were detected in exons 22 to 24. Silent single nucleotide polymorphisms were identified at nucleotides 2361 (G/A; Q787Q), 2370 (G/A; T790T), and 2457 (G/A; V819V) in exon 20, and at nucleotide 2709 (C/T; T903T) in exon 23, but the association between these polymorphisms and the somatic mutations was not observed. In this study, no mutations and no polymorphisms were detected in exons 18 to 24 of ERBB2.
All 43 mutations were detected in LCM samples, but 11 (26%) of these mutations were not detected in the bulk tumor samples. In 13 patients, LCM was performed at separate areas with different histologic subtypes, but no heterogeneity was identified; the same mutations were detected in nine patients, and no mutations were detected in four patients. Mutational analyses of synchronous double lung cancers were performed in two patients; one patient had a tumor with wild-type EGFR and a more invasive tumor with L858R + S768I, and the other patient had a tumor with a 9-bp deletion (del L747-E749) and a more invasive tumor with a 15-bp deletion (del E746-T751insA) + L703V.
Among the 39 patients with EGFR mutations, the proportion of mutant alleles ranged from 29% to 94%. Nineteen patients showed a BH pattern and 20 patients showed an MD pattern.
EGFR Copy Number
The EGFR copy number in the laser-captured tumor cells ranged from 1.27 to 31.2 per cell, and increased EGFR copy numbers ( 3.0 per cell) were observed in 29 patients (44%). The relation between the copy number and the proportion of mutant alleles is shown in Figure 1. Increased copy numbers were observed more frequently in patients with EGFR mutations than in patients with wild-type EGFR (56% [22 of 39 patients] v 26% [seven of 27 patients]; P = .014). High copy numbers ( 6.0 per cell) were observed only in patients with an MD pattern of mutations. The copy number and the proportion of mutant alleles among patients with EGFR mutations was positively correlated (Spearman correlation coefficient = 0.643; P < .001), implying that the mutant alleles were selectively amplified in patients with an MD pattern. One patient with an MD pattern had a tumor with only approximately one copy per cell, indicating a hemizygous mutation with a loss of wild-type allele. No alterations in the gene copy number were observed in normal lung tissues.
EGFR Mutations, EGFR Copy Number, and Clinical Outcome
The tumor responses to gefitinib according to the mutational status of EGFR are shown in Table 3. The response rates of patients with mutant and wild-type EGFR were 82% and 11%, respectively (P < 10–7). Seven patients with EGFR mutations were nonresponders; three patients had PD at 0.3 (early death), 2.3, and 2.3 months, and four patients had SD. Three of the four patients with SD had MR (TTP, 2.5, 5.2, and 6.9 months), and the other patient continued to receive gefitinib therapy without progression for 24.2 months, whereas all SD tumors with wild-type EGFR progressed within 5 months without MR. Meanwhile, three patients with wild-type EGFR exhibited PR, and two of these patients were still receiving gefitinib therapy without progression at 10.9+ and 21.1+ months. The Kaplan-Meier plots of TTP and OS according to the presence of the EGFR mutations are shown in Figures 2 and 3, respectively. Patients with EGFR mutations had a significantly longer TTP and OS compared with those with wild-type EGFR.
Univariate analyses were performed to assess the correlations among patient characteristics, EGFR mutations, EGFR copy number, and clinical outcome (Tables 4 and 5). The response rates were significantly higher in women, never/former smokers, and patients with BAC features and were marginally higher in patients with a papillary-dominant subtype. The response rates among these subgroups were approximately consistent with the rates of EGFR mutations. An increased EGFR copy number was also significantly associated with a higher response rate and a longer TTP.
The results of multivariate analyses among 62 patients with adenocarcinoma are shown in Table 6. The presence of EGFR mutations was strongly associated with a higher response rate, a longer TTP, and a longer OS. An increased EGFR copy number was also a significant or marginally significant predictor of a higher response rate and a longer TTP. These results did not change substantially if any combinations of variables were included in the models.
Among patients with wild-type EGFR, TTP was significantly longer in patients with increased EGFR copy numbers (median, 3.0 v 1.4 months; log-rank P = .021), and both of the two long-term responders had tumors with moderately increased EGFR copy numbers (3.20 and 3.45/cell). Among patients with EGFR mutations, TTP and OS were not significantly different according to the types of mutations, the presence of additional mutations, the proportion of mutant alleles, or the EGFR copy number (data not shown).
DISCUSSION
This study strongly implies that the mutational status of EGFR is a major determinant of gefitinib sensitivity in patients with NSCLC. The response rate was 82%, the median TTP was 12.6 months, and the MST was 20.4 months in gefitinib-treated patients with EGFR-mutant NSCLC. EGFR mutations might be a good prognostic factor independent of treatment, but these remarkable results suggest a survival benefit from gefitinib therapy in patients with EGFR mutations. Four of seven nonresponders with EGFR mutations also seemed to experience some clinical benefits because they had MR or a long SD ( 6 months). Among nine patients with SD, MR, or a long SD was observed only in patients with EGFR mutations. Although the sample size was too small to draw a firm conclusion, this finding suggests that EGFR mutations are also associated with clinical benefits in SD.
The EGFR mutations detected in this study were concentrated in three hotspots, deletions around codons 747 to 749, L858R, and G719S (or G719C), similar to the results of previous reports.6-13 Some genetic variations existed among these mutations. Together with one of the hotspot mutations, additional missense mutations in exons 18 or 20 were detected in four patients. Among the 39 patients with EGFR mutations, an MD pattern was observed in 20 patients. Because the EGFR copy number in their tumor cells increased as the proportion of mutant alleles increased, this pattern was assumed to be caused not by homozygous mutations but by the selective amplification of the mutant alleles. Because one patient had a hemizygous mutation without amplification, the loss of wild-type alleles was also thought to be responsible for the pattern. The moderately increased copy number in patients with a BH pattern or wild-type EGFR can be explained by EGFR amplification and/or polysomy of chromosome 7.
Among the patients with EGFR mutations, three patients had PD and eight of the other 36 patients had tumor regrowth within 6 months. This suggests the presence of other factors associated with intrinsic or acquired resistance to gefitinib. Although any genetic alterations of EGFR-mutant tumors at the time of primary surgery were not significantly associated with clinical outcome, that might be because further alterations occurred after the primary surgery or after gefitinib administration. Recently, a secondary mutation (C T at nucleotide 2369; T790M) in exon 20 was detected in patients with EGFR-mutant NSCLC who had tumor regrowth during gefitinib therapy after exhibiting an initial response to the agent; this mutation was thought to be associated with acquired resistance.28,29 To elucidate the determinants and the mechanism of resistance to gefitinib, genetic analyses of tumor samples obtained after gefitinib treatment are needed.
In this study, three (11%) of the 27 patients with wild-type EGFR responded to gefitinib. Various explanations for this result are possible: (1) the mutational analyses of the responders were false-negative, (2) the EGFR mutations occurred in their tumors after the primary surgery, (3) the recurrent tumors originated from a source other than the analyzed tumor cells, or (4) other determinants of gefitinib sensitivity were present.
The results of multivariate analyses suggest that the EGFR copy number is another independent predictor of gefitinib sensitivity. It is noteworthy that an increased EGFR copy number was observed in two of the three responders with wild-type EGFR, and was significantly associated with a longer TTP among patients with wild-type EGFR. Because patients with EGFR mutations had favorable clinical outcome regardless of EGFR copy numbers, the impact of increased copy numbers on EGFR-mutant NSCLC was unclear. In the overall population, an increased EGFR copy number was significantly associated with a higher response rate and a longer TTP, but not with a longer OS, which might be because an increased copy number had an unfavorable impact on prognosis, as suggested by another study.15 In chronic myeloid leukemia, as well as BCR-ABL mutations that were structurally corresponding to T790M in EGFR, an increased BCR-ABL gene copy number was reported as a determinant of resistance to imatinib, a BCR-ABL tyrosine kinase inhibitor.30 Therefore, we should consider the possibility that an increased EGFR copy number is associated with not only sensitivity but also resistance to gefitinib.
Among adenocarcinomas, the presence of BAC features was significantly associated with gefitinib sensitivity and EGFR mutations, but the BAC component was relatively small in most of the responders. The dominant subtype associated with a higher response rate was not BAC but papillary; both of the two patients with BwFI had PD, and all three patients with pure papillary adenocarcinoma without BAC features had PR. The association between pathologic features and gefitinib sensitivity or EGFR mutations is also the subject of further investigation.
In never/former smokers, both the EGFR mutation rate and the response rate were significantly higher than in current smokers. We speculate that EGFR mutations occur equally throughout the entire population, regardless of smoking history, and account for smoking-unrelated carcinogenesis. Because many other genetic alterations, like KRAS mutations, occur and induce lung adenocarcinoma more frequently in smokers, the EGFR mutation rate seems to be relatively lower in smokers with lung adenocarcinoma.
The response rate of 53% and the EGFR mutation rate of 59% observed in this study were higher than previously reported rates. These results can partially be attributed to the fact that the physicians tended to select patients with characteristics known to be predictive for gefitinib sensitivity: women, never-smokers, and patients with adenocarcinoma. Consequently, this cohort was not necessarily representative of unselected NSCLC populations in Japan. However, other recent studies have also shown relatively high frequencies (32% to 55%) of EGFR mutations in Japanese or East Asian patients with lung adenocarcinoma who underwent surgical resection.7,9-11,13 The reason why such somatic mutations occur selectively in East Asian people remains unknown. Environmental or genetic factors common among East Asian populations should be investigated to answer this question.
Recently, no significant survival benefit of gefitinib was reportedly observed in the initial analysis of the IRESSA Survival Evaluation in Lung Cancer (ISEL) trial, a phase III trial comparing gefitinib monotherapy to a placebo as a second- or third-line treatment for patients with advanced NSCLC.31 Because subgroup analyses of the trial suggested survival benefits in never smokers or Asian patients, the selection of patients is thought to be crucial when considering gefitinib treatment. Because the present study showed that the EGFR mutation status is a major determinant of gefitinib sensitivity, mutational analyses in patients with advanced NSCLC should be considered before deciding on a course of treatment.
In this study, we performed LCM and direct sequencing using methanol-fixed surgical specimens to obtain high-quality data. If we had analyzed only bulk tumor samples without LCM, nine of the 39 patients with EGFR mutations would have been misjudged as having wild-type EGFR. Thus such procedures with LCM are presently recommended for the detection of EGFR mutations. However, obtaining appropriate tumor samples is often difficult in patients with advanced NSCLC, and performing LCM and direct sequencing in all patients is not practical. Thus more practical methods for detecting the major EGFR mutations using small tumor samples contaminated with normal tissue should be developed and validated.
Other than EGFR mutations, some candidate predictive biomarkers have been studied. The EGFR copy number is the leading candidate, and it can also be detected by FISH. Practicality and accuracy should be assessed comparing FISH and quantitative real-time PCR. The impact of ERBB2 mutations on clinical outcome remains to be investigated because we could not detect any mutations in ERBB2 in the present study. Protein expression analyses by IHC are easier to perform than the genetic analyses, but their significance is still controversial. Further studies are required to evaluate the predictive values of these biomarkers and to determine whether they are independent predictors of gefitinib sensitivity or surrogate markers of EGFR mutations.
In conclusion, this study indicates that EGFR mutations and increased copy numbers predict better clinical outcome in patients with NSCLC treated with gefitinib. Further research and clinical trials are needed to incorporate these markers into clinical practice appropriately.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Yukihiro Yoshida, MD; Shunichi Watanabe, MD; Kenji Suzuki, MD; Hisao Asamura, MD; and Ryosuke Tsuchiya, MD, for providing surgical specimens and helpful advice, and Chizu Kina, Chie Hirama, Sanae Kobayashi, Yasuko Kuwahara, Go Maeno, Sachiyo Mimaki, Yoko Odaka, Shizuka Shinohara, Takahiro Taniguchi, and Mineko Ushiama for LCM and DNA analysis. We also thank Setsuo Hirohashi, MD, for his invaluable direction and support of the study.
NOTES
Supported by a program for the promotion of Fundamental Studies in Health Sciences of the Pharmaceuticals and Medical Devices Agency and by Health and Labour Science Research Grants from the Ministry of Health, Labour and Welfare.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Fukuoka M, Yano S, Giaccone G, et al: A multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small cell lung cancer (The IDEAL 1 Trial). J Clin Oncol 21:2237-2246, 2003
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 290:2149-2158, 2003
Miller VA, Kris MG, Shah N, et al: Bronchioloalveolar pathologic subtype and smoking history predict sensitivity to gefitinib in advanced non–small-cell lung cancer. J Clin Oncol 22:1103-1109, 2004
Takano T, Ohe Y, Kusumoto M, et al: Risk factors for interstitial lung disease and predictive factors for tumor response in patients with advanced non-small cell lung cancer treated with gefitinib. Lung Cancer 45:93-104, 2004
Kim YH, Ishii G, Goto K, et al: Dominant papillary subtype is a significant predictor of the response to gefitinib in adenocarcinoma of the lung. Clin Cancer Res 10:7311-7317, 2004
Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004
Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 304:1497-1500, 2004
Pao W, Miller V, Zakowski M, et al: EGF receptor gene mutations are common in lung cancers from "never smokers" and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci U S A 101:13306-13311, 2004
Kosaka T, Yatabe Y, Endoh H, et al: Mutations of the epidermal growth factor receptor gene in lung cancer: Biological and clinical implications. Cancer Res 64:8919-8923, 2004
Huang SF, Liu HP, Li LH, et al: High frequency of epidermal growth factor receptor mutations with complex patterns in non–small cell lung cancers related to gefitinib responsiveness in Taiwan. Clin Cancer Res 10:8195-8203, 2004
Shigematsu H, Lin L, Takahashi T, et al: Clinical and biological features associated with epidermal growth factor receptor gene mutations in lung cancers. J Natl Cancer Inst 97:339-346, 2005
Marchetti A, Martella C, Felicioni L, et al: EGFR mutations in non–small-cell lung cancer: Analysis of a large series of cases and development of a rapid and sensitive method for diagnostic screening with potential implications on pharmacologic treatment. J Clin Oncol 23:857-865, 2005
Tokumo M, Toyooka S, Kiura K, et al: The relationship between epidermal growth factor receptor mutations and clinicopathologic features in non–small cell lung cancers. Clin Cancer Res 11:1167-1173, 2005
Noguchi M, Furuya S, Takeuchi T, et al: Modification formalin and methanol fixation methods for molecular biological and morphological analyses. Pathol Int 47:685-691, 1997
Hirsch FR, Varella-Garcia M, Bunn PA Jr, et al: Epidermal growth factor receptor in non-small-cell lung carcinomas: Correlation between gene copy number and protein expression and impact on prognosis. J Clin Oncol 21:3798-3807, 2003
Capuzzo F, Hirsch FR, Rossi E, et al: Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non–small-cell lung cancer. J Natl Cancer Inst 97:643-655, 2005
Stephens P, Hunter C, Bignell G, et al: Lung cancer: Intragenic ERBB2 kinase mutations in tumours. Nature 431:525-526, 2004
Shigematsu H, Takahashi T, Nomura M, et al: Somatic mutations of the HER2 kinase domain in lung adenocarcinomas. Cancer Res 65:1642-1646, 2005
Cappuzzo F, Magrini E, Ceresoli GL, et al: Akt phosphorylation and gefitinib efficacy in patients with advanced non-small-cell lung cancer. J Natl Cancer Inst 96:1133-1141, 2004
Han SW, Hwang PG, Chung DH, et al: Epidermal growth factor receptor (EGFR) downstream molecules as response predictive markers for gefitinib (Iressa, ZD1839) in chemotherapy resistant non-small-cell lung cancer. Int J Cancer 113:109-115, 2005
Franklin WA, Chansky K, Gumerlock PH, et al: Association between activation of ErbB pathway genes and survival following gefitinib treatment in advanced BAC (SWOG 0126). J Clin Oncol 22:618, 2004 (suppl; abstr 7015)
Emmert-Buck MR, Bonner RF, Smith PD, et al: Laser capture microdissection. Science 274:998-1001, 1996
Maeno G, Isobe T, Tokoro Y, et al: NAMIHEI: A novel algorithm for genomic polymorphism detection from DNA sequence. Genome Informatics 14:595-596, 2003
Ronaghi M: Pyrosequencing sheds light on DNA sequencing. Genome Res 11:3-11, 2001
Travis WD, Colby TV, Corrin B, et al: Histologic Typing of Tumors of Lung and Pleura: World Health Organization International Classification of Tumors (ed 3). New York, NY, Springer Verlag, 1999
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 74:1640-1646, 2002
Green S, Weiss GR: Southwest Oncology Group standard response criteria, endpoint definitions and toxicity criteria. Invest New Drugs 10:239-253, 1992
Kobayashi S, Boggon TJ, Dayaram T, et al: EGFR mutation and resistance of non-small-cell lung cancer to gefitinib. N Engl J Med 352:786-792, 2005
Pao W, Miller VA, Politi KA, et al: Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain. PLoS Med 2:e73, 2005
Gorre ME, Mohammed M, Ellwood K, et al: Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 293:876-880, 2001
Thatcher N, Chang A, Parikh P, et al: Results of a phase III placebo-controlled study (ISEL) of gefitinib (IRESSA) plus best supportive care (BSC) in patients with advanced non-small-cell lung cancer (NSCLC) who had received 1 or 2 prior chemotherapy regimens. Proc Am Assoc Cancer Res 46, 2005 (abstr LB-6)(Toshimi Takano, Yuichiro )
Genetics and Pathology Divisions, National Cancer Center Research Institute
Statistics and Cancer Control Division, Research Center for Cancer Prevention and Screening, National Cancer Center, Tokyo, Japan
ABSTRACT
PURPOSE: To evaluate epidermal growth factor receptor (EGFR) mutations and copy number as predictors of clinical outcome in patients with non–small-cell lung cancer (NSCLC) receiving gefitinib.
PATIENTS AND METHODS: Sixty-six patients with NSCLC who experienced relapse after surgery and received gefitinib were included. Direct sequencing of exons 18 to 24 of EGFR and exons 18 to 24 of ERBB2 was performed using DNA extracted from surgical specimens. Pyrosequencing and quantitative real-time polymerase chain reaction were performed to analyze the allelic pattern and copy number of EGFR.
RESULTS: Thirty-nine patients (59%) had EGFR mutations; 20 patients had deletional mutations in exon 19, 17 patients had missense mutations (L858R) in exon 21, and two patients had missense mutations (G719S or G719C) in exon 18. No mutations were identified in ERBB2. Response rate (82% [32 of 39 patients] v 11% [three of 27 patients]; P < .0001), time to progression (TTP; median, 12.6 v 1.7 months; P < .0001), and overall survival (median, 20.4 v 6.9 months; P = .0001) were significantly better in patients with EGFR mutations than in patients with wild-type EGFR. Increased EGFR copy numbers ( 3/cell) were observed in 29 patients (44%) and were significantly associated with a higher response rate (72% [21 of 29 patients] v 38% [14 of 37 patients]; P = .005) and a longer TTP (median, 9.4 v 2.6 months; P = .038). High EGFR copy numbers ( 6/cell) were caused by selective amplification of mutant alleles.
CONCLUSION: EGFR mutations and increased copy numbers were significantly associated with better clinical outcome in gefitinib-treated NSCLC patients.
INTRODUCTION
The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase of the ErbB family that has been implicated in cell proliferation and survival and is frequently overexpressed in many solid tumors, including non–small-cell lung cancer (NSCLC). Gefitinib (Iressa; AstraZeneca, Osaka, Japan) is an orally active, selective EGFR tyrosine kinase inhibitor that binds to the adenosine triphosphate–binding pocket of the EGFR kinase domain and blocks downstream signaling pathways. Two phase II studies, IRESSA Dose Evaluation in Advanced Lung Cancer 1 and 2 (IDEAL 1 and 2), have demonstrated that gefitinib monotherapy exerts an antitumor activity in patients with advanced NSCLC who had previously received platinum-based chemotherapy.1,2 Gefitinib was approved in Japan for the treatment of inoperable or recurrent NSCLC in July 2002.
The IDEAL trials and retrospective studies have revealed that women, never smokers, patients with adenocarcinoma, and Japanese patients have higher response rates to gefitinib.1-4 Among patients with adenocarcinoma, histologic subtypes have been studied; one study showed that responses were more frequent in patients with bronchioloalveolar carcinoma (BAC) features (38% v 14%; P < .001),3 whereas another study showed that the response rate was higher in patients with a papillary-dominant subtype (76% v 21%; P = .002).5
Although no predictive molecular markers had been identified at the time of approval, somatic mutations in the kinase domain of EGFR have been subsequently linked to gefitinib sensitivity. According to three initial reports, 20 of 24 gefitinib-responsive tumors contained EGFR mutations, whereas 19 nonresponsive tumors did not contain any mutations.6-8 The mutations were detected in exons 18 to 21 of EGFR, close to the region coding the adenosine triphosphate–binding pocket of the kinase domain, and most of them were observed in two hotspots: in-frame deletions including amino acids at codons 747 to 749 in exon 19 and an amino acid substitution at codon 858 (L858R) in exon 21. Analyses of surgically resected NSCLC tumors revealed that such mutations were more frequent among women, never smokers, patients with adenocarcinoma, and Japanese or East Asian patients,7-13 consistent with the known clinical predictors of gefitinib sensitivity.
To evaluate the exact predictive value, we studied consecutive patients with recurrent NSCLC who received gefitinib therapy. To insure high-quality genetic analyses of the archived tissues, we used methanol-fixed, paraffin-embedded surgical specimens, which are known to preserve DNA better than formalin-fixed tissues,14 and performed laser capture microdissection (LCM).
Recently, some other biomarkers of NSCLC have been studied. The EGFR and chromosome 7 copy numbers in NSCLC were assessed using fluorescence in situ hybridization (FISH), and more than 3.0 EGFR copies per cell (balanced polysomy or gene amplification) were detected in 39 (22%) of 183 patients.15 A correlation between an increased EGFR copy number and gefitinib sensitivity was also proposed in another study.16 In yet other studies, mutations in the kinase domain of ERBB2 (HER2), a gene coding another receptor tyrosine kinase of the ErbB family, were detected in 16 (3.6%) of 445 patients with lung adenocarcinoma.17,18 In the current study, we also analyzed the EGFR copy number and the presence of ERBB2 mutations to assess their impact on clinical outcome.
The expression of EGFR and related proteins has been more widely studied using immunohistochemistry. Some studies suggested that high expression of phosphorylated Akt19,20 or low expression of phosphorylated mitogen-activated protein kinase20,21 was associated with better outcome in gefitinib-treated patients, but in general, methods, criteria, and results were inconsistent among studies. We thought that protein expression should be analyzed in another exploratory study, and in the current study, we focused on the genetic analyses.
PATIENTS AND METHODS
Patients
After searching the pharmaceutical records of the National Cancer Center Hospital, 279 patients with NSCLC who had begun receiving gefitinib monotherapy (250 mg/d) between July 2002 and May 2004 were identified. Seventy-three of these patients had undergone surgical resection of primary NSCLC at the hospital and subsequently relapsed. Recurrences were not necessarily confirmed pathologically but were diagnosed clinically. Seven patients were ineligible for inclusion in this study because methanol-fixed tissues were not available (n = 5) or their informed consent to the genetic analysis was not obtained (n = 2); consequently, 66 patients were included.
Genetic Analyses of EGFR and ERBB2
On a protocol approved by the institutional review board of the National Cancer Center, we performed mutational analyses of exons 18 to 24 of EGFR and exons 18 to 24 of ERBB2 and analyzed the EGFR copy number. Methanol-fixed, paraffin-embedded surgical specimens of primary NSCLC were collected retrospectively, and DNA was extracted from bulk tumor tissue, laser capture microdissected tumor tissue, and normal lung tissue from each patient. LCM was performed using a PixCell II LCM system (Arcturus Engineering Inc, Mountain View, CA) according to a previously described method.22 If appropriate, tumor cells were captured separately from two areas with different histologic subtypes, such as an area with a BAC subtype and another area with stromal invasion. Nested polymerase chain reaction (PCR) was performed to amplify exons 18 through 24 of EGFR using previously described primers,6 and standard PCR was used to amplify exons 18 through 24 of ERBB2. Direct sequencing of the PCR products was performed using ABI PRISM 3700 and 3100 DNA Sequencers (Applied Biosystems, Foster City, CA). All sequencing reactions were performed in both forward and reverse directions, and single nucleotide substitutions, insertions, and deletions were detected using an application program named NAMIHEI.23 Pyrosequencing was performed to verify the sequencing data of the hotspots of EGFR and to assess the proportion of mutant alleles in the laser-captured tumor cells using a Pyrosequencing PSQ 96MA (Pyrosequencing, Uppsala, Sweden).24 On the basis of the proportion of mutant alleles, EGFR mutations were divided into two patterns: balanced heterozygous (BH) pattern (< 60%) and mutant-allele-dominant (MD) pattern ( 60%). The cutoff level of 60% was decided because if more than 60%, the superiority of the mutant over the wild-type sequences was obvious on the direct sequencing chromatograms. Quantitative, real-time, TaqMan duplex PCR was performed to analyze the EGFR copy number using an ABI PRISM 7000 Sequence Detection System (Applied Biosystems). The EGFR primers were 5'-GGAGGACCGTCGCTTGGT-3' and 5'-AACACCGCAGCATGTCAAGA-3'; the probe (5'-CACCGCGACCTGGCAGCCA-3') was labeled with the reporter dye 6-carboxyfluorescein (FAM). RNaseP was coamplified in the same reaction mixture as the endogenous reference gene using TaqMan RNaseP Control Reagents (6-carboxyrhodamine [VIC] dye; Applied Biosystems). The average EGFR copy number per cell was calculated from the differences in the threshold amplification cycles between EGFR and RNaseP. Peripheral-blood samples obtained from healthy volunteers were analyzed as normal controls. Decreased, normal, moderately increased, and highly increased EGFR copy numbers were defined as less than 1.5, 1.5 to 3.0, 3.0 to 6.0, and 6.0 copies per cell, respectively.
Pathologic Evaluation
We reviewed the histologic features of the 66 patients using hematoxylin and eosin–stained slides of tumor samples. Two board-certified pathologists (K.T. and Y.M.) who were unaware of the patients' outcome and mutational status examined all the specimens independently; in case of discrepancy, final diagnoses were established by consensus. Adenocarcinoma was categorized in two ways. The first categorization was based on the WHO's classification of lung tumors,25 which includes four major subtypes of adenocarcinoma: papillary, acinar, BAC, and solid; the dominant subtype in the total tumor mass of each case was documented. The second categorization was based on a report from the Memorial Sloan-Kettering Cancer Center,26 in which adenocarcinomas were classified into adenocarcinoma without BAC features (Ad), adenocarcinoma with BAC features (AwBF), BAC with focal invasion (BwFI), and pure BAC (PBAC). If two or more tumors were present in one patient, the diagnosis of the most invasive tumor in each case was documented.
Radiologic Evaluation
In patients who had measurable lesions, imaging studies were performed at baseline, approximately 4 weeks after the initiation of gefitinib treatment, and periodically thereafter throughout the treatment. One board-certified radiologist (U.T.) who was unaware of the patients' mutational status reviewed the baseline, first follow-up, and confirmatory imaging studies and classified the tumor responses into complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) using standard bidimensional measurements.27 Responders were defined as patients with CR or PR. In this study, SD was subdivided into minor response (MR) and no response. MR was defined as a 25% decrease in the sum of the products of the perpendicular diameters of all measurable lesions at any point during gefitinib treatment. Time to progression (TTP) was defined as the time from the start of gefitinib administration to confirmed disease progression or death.
Statistical Analysis
The associations among mutational status, EGFR copy number, patient characteristics, and tumor response to gefitinib were assessed using a 2 test. The differences in TTP and overall survival (OS) according to the patient subgroups were compared using Kaplan-Meier curves and log-rank tests. Multivariate analyses using logistic regression models and Cox proportional hazard models were performed to assess the association between the biomarkers and clinical outcome while adjusting for the baseline patient characteristics. All analyses were performed using the SPSS statistical package (SPSS version 11.0 for Windows; SPSS Inc, Chicago, IL).
RESULTS
Patient Characteristics
The patient characteristics are listed in Table 1. All of the patients were Japanese. The proportions of women (39%), never smokers (47%), and patients with adenocarcinoma (94%) in this study were higher than those in a database of more than 1,000 patients with advanced or recurrent NSCLC treated at our hospital during the four most recent years (27%, 27%, and 73%, respectively). Twenty-two patients (33%) had been included in our phase II trial for first-line gefitinib therapy for patients with recurrent NSCLC, and the others had been treated with gefitinib in clinical practice settings. The operations for primary NSCLC were performed between February 1994 and August 2003, and the median time from the operations to the start of gefitinib was 2.3 years (range, 0.6 to 9.1 years).
Clinical Outcome
Sixty-four patients had measurable lesions at the start of gefitinib administration. CR and PR were observed in two and 32 patients, respectively. MR was observed in three of nine patients with SD. Twenty-one patients had PD, including six patients who died before the first follow-up imaging studies. Two patients had only unmeasurable bone lesions at baseline; one patient showed rapid symptom improvement and continued to receive gefitinib therapy without progression for 13.8+ months, whereas the other patient developed new lesions and died on day 71. These patients were included in the analysis as a responder and a nonresponder, respectively. The overall response rate was 53%. Forty-one patients died, and the median follow-up time for the 25 survivors was 14.6 months (range, 10.3 to 32.3 months). Eleven patients were still receiving gefitinib without progression at the time of the analysis. The median TTP and the median survival time (MST) for all patients were 5.2 and 16.3 months, respectively.
EGFR and ERBB2 Mutations
Forty-three mutations in the EGFR tyrosine kinase domain were detected in 39 (59%) of the 66 patients. All the mutations detected in this study are shown in Table 2. Twenty patients had deletional mutations in exon 19, and 17 patients had missense mutations (L858R) in exon 21. In exons 18 and 20, five types of missense mutations were detected. Two of them (G719S and G719C) occurred at a codon considered to be a third hotspot.6,7,9-12 The others (L703V, E709K, and S768I) were detected in patients who also had mutations at the hotspots. Because these mutations were not detected in the normal lung tissues from the same patients, they were considered to be somatic mutations. No somatic mutations were detected in exons 22 to 24. Silent single nucleotide polymorphisms were identified at nucleotides 2361 (G/A; Q787Q), 2370 (G/A; T790T), and 2457 (G/A; V819V) in exon 20, and at nucleotide 2709 (C/T; T903T) in exon 23, but the association between these polymorphisms and the somatic mutations was not observed. In this study, no mutations and no polymorphisms were detected in exons 18 to 24 of ERBB2.
All 43 mutations were detected in LCM samples, but 11 (26%) of these mutations were not detected in the bulk tumor samples. In 13 patients, LCM was performed at separate areas with different histologic subtypes, but no heterogeneity was identified; the same mutations were detected in nine patients, and no mutations were detected in four patients. Mutational analyses of synchronous double lung cancers were performed in two patients; one patient had a tumor with wild-type EGFR and a more invasive tumor with L858R + S768I, and the other patient had a tumor with a 9-bp deletion (del L747-E749) and a more invasive tumor with a 15-bp deletion (del E746-T751insA) + L703V.
Among the 39 patients with EGFR mutations, the proportion of mutant alleles ranged from 29% to 94%. Nineteen patients showed a BH pattern and 20 patients showed an MD pattern.
EGFR Copy Number
The EGFR copy number in the laser-captured tumor cells ranged from 1.27 to 31.2 per cell, and increased EGFR copy numbers ( 3.0 per cell) were observed in 29 patients (44%). The relation between the copy number and the proportion of mutant alleles is shown in Figure 1. Increased copy numbers were observed more frequently in patients with EGFR mutations than in patients with wild-type EGFR (56% [22 of 39 patients] v 26% [seven of 27 patients]; P = .014). High copy numbers ( 6.0 per cell) were observed only in patients with an MD pattern of mutations. The copy number and the proportion of mutant alleles among patients with EGFR mutations was positively correlated (Spearman correlation coefficient = 0.643; P < .001), implying that the mutant alleles were selectively amplified in patients with an MD pattern. One patient with an MD pattern had a tumor with only approximately one copy per cell, indicating a hemizygous mutation with a loss of wild-type allele. No alterations in the gene copy number were observed in normal lung tissues.
EGFR Mutations, EGFR Copy Number, and Clinical Outcome
The tumor responses to gefitinib according to the mutational status of EGFR are shown in Table 3. The response rates of patients with mutant and wild-type EGFR were 82% and 11%, respectively (P < 10–7). Seven patients with EGFR mutations were nonresponders; three patients had PD at 0.3 (early death), 2.3, and 2.3 months, and four patients had SD. Three of the four patients with SD had MR (TTP, 2.5, 5.2, and 6.9 months), and the other patient continued to receive gefitinib therapy without progression for 24.2 months, whereas all SD tumors with wild-type EGFR progressed within 5 months without MR. Meanwhile, three patients with wild-type EGFR exhibited PR, and two of these patients were still receiving gefitinib therapy without progression at 10.9+ and 21.1+ months. The Kaplan-Meier plots of TTP and OS according to the presence of the EGFR mutations are shown in Figures 2 and 3, respectively. Patients with EGFR mutations had a significantly longer TTP and OS compared with those with wild-type EGFR.
Univariate analyses were performed to assess the correlations among patient characteristics, EGFR mutations, EGFR copy number, and clinical outcome (Tables 4 and 5). The response rates were significantly higher in women, never/former smokers, and patients with BAC features and were marginally higher in patients with a papillary-dominant subtype. The response rates among these subgroups were approximately consistent with the rates of EGFR mutations. An increased EGFR copy number was also significantly associated with a higher response rate and a longer TTP.
The results of multivariate analyses among 62 patients with adenocarcinoma are shown in Table 6. The presence of EGFR mutations was strongly associated with a higher response rate, a longer TTP, and a longer OS. An increased EGFR copy number was also a significant or marginally significant predictor of a higher response rate and a longer TTP. These results did not change substantially if any combinations of variables were included in the models.
Among patients with wild-type EGFR, TTP was significantly longer in patients with increased EGFR copy numbers (median, 3.0 v 1.4 months; log-rank P = .021), and both of the two long-term responders had tumors with moderately increased EGFR copy numbers (3.20 and 3.45/cell). Among patients with EGFR mutations, TTP and OS were not significantly different according to the types of mutations, the presence of additional mutations, the proportion of mutant alleles, or the EGFR copy number (data not shown).
DISCUSSION
This study strongly implies that the mutational status of EGFR is a major determinant of gefitinib sensitivity in patients with NSCLC. The response rate was 82%, the median TTP was 12.6 months, and the MST was 20.4 months in gefitinib-treated patients with EGFR-mutant NSCLC. EGFR mutations might be a good prognostic factor independent of treatment, but these remarkable results suggest a survival benefit from gefitinib therapy in patients with EGFR mutations. Four of seven nonresponders with EGFR mutations also seemed to experience some clinical benefits because they had MR or a long SD ( 6 months). Among nine patients with SD, MR, or a long SD was observed only in patients with EGFR mutations. Although the sample size was too small to draw a firm conclusion, this finding suggests that EGFR mutations are also associated with clinical benefits in SD.
The EGFR mutations detected in this study were concentrated in three hotspots, deletions around codons 747 to 749, L858R, and G719S (or G719C), similar to the results of previous reports.6-13 Some genetic variations existed among these mutations. Together with one of the hotspot mutations, additional missense mutations in exons 18 or 20 were detected in four patients. Among the 39 patients with EGFR mutations, an MD pattern was observed in 20 patients. Because the EGFR copy number in their tumor cells increased as the proportion of mutant alleles increased, this pattern was assumed to be caused not by homozygous mutations but by the selective amplification of the mutant alleles. Because one patient had a hemizygous mutation without amplification, the loss of wild-type alleles was also thought to be responsible for the pattern. The moderately increased copy number in patients with a BH pattern or wild-type EGFR can be explained by EGFR amplification and/or polysomy of chromosome 7.
Among the patients with EGFR mutations, three patients had PD and eight of the other 36 patients had tumor regrowth within 6 months. This suggests the presence of other factors associated with intrinsic or acquired resistance to gefitinib. Although any genetic alterations of EGFR-mutant tumors at the time of primary surgery were not significantly associated with clinical outcome, that might be because further alterations occurred after the primary surgery or after gefitinib administration. Recently, a secondary mutation (C T at nucleotide 2369; T790M) in exon 20 was detected in patients with EGFR-mutant NSCLC who had tumor regrowth during gefitinib therapy after exhibiting an initial response to the agent; this mutation was thought to be associated with acquired resistance.28,29 To elucidate the determinants and the mechanism of resistance to gefitinib, genetic analyses of tumor samples obtained after gefitinib treatment are needed.
In this study, three (11%) of the 27 patients with wild-type EGFR responded to gefitinib. Various explanations for this result are possible: (1) the mutational analyses of the responders were false-negative, (2) the EGFR mutations occurred in their tumors after the primary surgery, (3) the recurrent tumors originated from a source other than the analyzed tumor cells, or (4) other determinants of gefitinib sensitivity were present.
The results of multivariate analyses suggest that the EGFR copy number is another independent predictor of gefitinib sensitivity. It is noteworthy that an increased EGFR copy number was observed in two of the three responders with wild-type EGFR, and was significantly associated with a longer TTP among patients with wild-type EGFR. Because patients with EGFR mutations had favorable clinical outcome regardless of EGFR copy numbers, the impact of increased copy numbers on EGFR-mutant NSCLC was unclear. In the overall population, an increased EGFR copy number was significantly associated with a higher response rate and a longer TTP, but not with a longer OS, which might be because an increased copy number had an unfavorable impact on prognosis, as suggested by another study.15 In chronic myeloid leukemia, as well as BCR-ABL mutations that were structurally corresponding to T790M in EGFR, an increased BCR-ABL gene copy number was reported as a determinant of resistance to imatinib, a BCR-ABL tyrosine kinase inhibitor.30 Therefore, we should consider the possibility that an increased EGFR copy number is associated with not only sensitivity but also resistance to gefitinib.
Among adenocarcinomas, the presence of BAC features was significantly associated with gefitinib sensitivity and EGFR mutations, but the BAC component was relatively small in most of the responders. The dominant subtype associated with a higher response rate was not BAC but papillary; both of the two patients with BwFI had PD, and all three patients with pure papillary adenocarcinoma without BAC features had PR. The association between pathologic features and gefitinib sensitivity or EGFR mutations is also the subject of further investigation.
In never/former smokers, both the EGFR mutation rate and the response rate were significantly higher than in current smokers. We speculate that EGFR mutations occur equally throughout the entire population, regardless of smoking history, and account for smoking-unrelated carcinogenesis. Because many other genetic alterations, like KRAS mutations, occur and induce lung adenocarcinoma more frequently in smokers, the EGFR mutation rate seems to be relatively lower in smokers with lung adenocarcinoma.
The response rate of 53% and the EGFR mutation rate of 59% observed in this study were higher than previously reported rates. These results can partially be attributed to the fact that the physicians tended to select patients with characteristics known to be predictive for gefitinib sensitivity: women, never-smokers, and patients with adenocarcinoma. Consequently, this cohort was not necessarily representative of unselected NSCLC populations in Japan. However, other recent studies have also shown relatively high frequencies (32% to 55%) of EGFR mutations in Japanese or East Asian patients with lung adenocarcinoma who underwent surgical resection.7,9-11,13 The reason why such somatic mutations occur selectively in East Asian people remains unknown. Environmental or genetic factors common among East Asian populations should be investigated to answer this question.
Recently, no significant survival benefit of gefitinib was reportedly observed in the initial analysis of the IRESSA Survival Evaluation in Lung Cancer (ISEL) trial, a phase III trial comparing gefitinib monotherapy to a placebo as a second- or third-line treatment for patients with advanced NSCLC.31 Because subgroup analyses of the trial suggested survival benefits in never smokers or Asian patients, the selection of patients is thought to be crucial when considering gefitinib treatment. Because the present study showed that the EGFR mutation status is a major determinant of gefitinib sensitivity, mutational analyses in patients with advanced NSCLC should be considered before deciding on a course of treatment.
In this study, we performed LCM and direct sequencing using methanol-fixed surgical specimens to obtain high-quality data. If we had analyzed only bulk tumor samples without LCM, nine of the 39 patients with EGFR mutations would have been misjudged as having wild-type EGFR. Thus such procedures with LCM are presently recommended for the detection of EGFR mutations. However, obtaining appropriate tumor samples is often difficult in patients with advanced NSCLC, and performing LCM and direct sequencing in all patients is not practical. Thus more practical methods for detecting the major EGFR mutations using small tumor samples contaminated with normal tissue should be developed and validated.
Other than EGFR mutations, some candidate predictive biomarkers have been studied. The EGFR copy number is the leading candidate, and it can also be detected by FISH. Practicality and accuracy should be assessed comparing FISH and quantitative real-time PCR. The impact of ERBB2 mutations on clinical outcome remains to be investigated because we could not detect any mutations in ERBB2 in the present study. Protein expression analyses by IHC are easier to perform than the genetic analyses, but their significance is still controversial. Further studies are required to evaluate the predictive values of these biomarkers and to determine whether they are independent predictors of gefitinib sensitivity or surrogate markers of EGFR mutations.
In conclusion, this study indicates that EGFR mutations and increased copy numbers predict better clinical outcome in patients with NSCLC treated with gefitinib. Further research and clinical trials are needed to incorporate these markers into clinical practice appropriately.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Yukihiro Yoshida, MD; Shunichi Watanabe, MD; Kenji Suzuki, MD; Hisao Asamura, MD; and Ryosuke Tsuchiya, MD, for providing surgical specimens and helpful advice, and Chizu Kina, Chie Hirama, Sanae Kobayashi, Yasuko Kuwahara, Go Maeno, Sachiyo Mimaki, Yoko Odaka, Shizuka Shinohara, Takahiro Taniguchi, and Mineko Ushiama for LCM and DNA analysis. We also thank Setsuo Hirohashi, MD, for his invaluable direction and support of the study.
NOTES
Supported by a program for the promotion of Fundamental Studies in Health Sciences of the Pharmaceuticals and Medical Devices Agency and by Health and Labour Science Research Grants from the Ministry of Health, Labour and Welfare.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Fukuoka M, Yano S, Giaccone G, et al: A multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small cell lung cancer (The IDEAL 1 Trial). J Clin Oncol 21:2237-2246, 2003
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 290:2149-2158, 2003
Miller VA, Kris MG, Shah N, et al: Bronchioloalveolar pathologic subtype and smoking history predict sensitivity to gefitinib in advanced non–small-cell lung cancer. J Clin Oncol 22:1103-1109, 2004
Takano T, Ohe Y, Kusumoto M, et al: Risk factors for interstitial lung disease and predictive factors for tumor response in patients with advanced non-small cell lung cancer treated with gefitinib. Lung Cancer 45:93-104, 2004
Kim YH, Ishii G, Goto K, et al: Dominant papillary subtype is a significant predictor of the response to gefitinib in adenocarcinoma of the lung. Clin Cancer Res 10:7311-7317, 2004
Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004
Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 304:1497-1500, 2004
Pao W, Miller V, Zakowski M, et al: EGF receptor gene mutations are common in lung cancers from "never smokers" and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci U S A 101:13306-13311, 2004
Kosaka T, Yatabe Y, Endoh H, et al: Mutations of the epidermal growth factor receptor gene in lung cancer: Biological and clinical implications. Cancer Res 64:8919-8923, 2004
Huang SF, Liu HP, Li LH, et al: High frequency of epidermal growth factor receptor mutations with complex patterns in non–small cell lung cancers related to gefitinib responsiveness in Taiwan. Clin Cancer Res 10:8195-8203, 2004
Shigematsu H, Lin L, Takahashi T, et al: Clinical and biological features associated with epidermal growth factor receptor gene mutations in lung cancers. J Natl Cancer Inst 97:339-346, 2005
Marchetti A, Martella C, Felicioni L, et al: EGFR mutations in non–small-cell lung cancer: Analysis of a large series of cases and development of a rapid and sensitive method for diagnostic screening with potential implications on pharmacologic treatment. J Clin Oncol 23:857-865, 2005
Tokumo M, Toyooka S, Kiura K, et al: The relationship between epidermal growth factor receptor mutations and clinicopathologic features in non–small cell lung cancers. Clin Cancer Res 11:1167-1173, 2005
Noguchi M, Furuya S, Takeuchi T, et al: Modification formalin and methanol fixation methods for molecular biological and morphological analyses. Pathol Int 47:685-691, 1997
Hirsch FR, Varella-Garcia M, Bunn PA Jr, et al: Epidermal growth factor receptor in non-small-cell lung carcinomas: Correlation between gene copy number and protein expression and impact on prognosis. J Clin Oncol 21:3798-3807, 2003
Capuzzo F, Hirsch FR, Rossi E, et al: Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non–small-cell lung cancer. J Natl Cancer Inst 97:643-655, 2005
Stephens P, Hunter C, Bignell G, et al: Lung cancer: Intragenic ERBB2 kinase mutations in tumours. Nature 431:525-526, 2004
Shigematsu H, Takahashi T, Nomura M, et al: Somatic mutations of the HER2 kinase domain in lung adenocarcinomas. Cancer Res 65:1642-1646, 2005
Cappuzzo F, Magrini E, Ceresoli GL, et al: Akt phosphorylation and gefitinib efficacy in patients with advanced non-small-cell lung cancer. J Natl Cancer Inst 96:1133-1141, 2004
Han SW, Hwang PG, Chung DH, et al: Epidermal growth factor receptor (EGFR) downstream molecules as response predictive markers for gefitinib (Iressa, ZD1839) in chemotherapy resistant non-small-cell lung cancer. Int J Cancer 113:109-115, 2005
Franklin WA, Chansky K, Gumerlock PH, et al: Association between activation of ErbB pathway genes and survival following gefitinib treatment in advanced BAC (SWOG 0126). J Clin Oncol 22:618, 2004 (suppl; abstr 7015)
Emmert-Buck MR, Bonner RF, Smith PD, et al: Laser capture microdissection. Science 274:998-1001, 1996
Maeno G, Isobe T, Tokoro Y, et al: NAMIHEI: A novel algorithm for genomic polymorphism detection from DNA sequence. Genome Informatics 14:595-596, 2003
Ronaghi M: Pyrosequencing sheds light on DNA sequencing. Genome Res 11:3-11, 2001
Travis WD, Colby TV, Corrin B, et al: Histologic Typing of Tumors of Lung and Pleura: World Health Organization International Classification of Tumors (ed 3). New York, NY, Springer Verlag, 1999
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 74:1640-1646, 2002
Green S, Weiss GR: Southwest Oncology Group standard response criteria, endpoint definitions and toxicity criteria. Invest New Drugs 10:239-253, 1992
Kobayashi S, Boggon TJ, Dayaram T, et al: EGFR mutation and resistance of non-small-cell lung cancer to gefitinib. N Engl J Med 352:786-792, 2005
Pao W, Miller VA, Politi KA, et al: Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain. PLoS Med 2:e73, 2005
Gorre ME, Mohammed M, Ellwood K, et al: Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 293:876-880, 2001
Thatcher N, Chang A, Parikh P, et al: Results of a phase III placebo-controlled study (ISEL) of gefitinib (IRESSA) plus best supportive care (BSC) in patients with advanced non-small-cell lung cancer (NSCLC) who had received 1 or 2 prior chemotherapy regimens. Proc Am Assoc Cancer Res 46, 2005 (abstr LB-6)(Toshimi Takano, Yuichiro )