Analysis of ErbB Receptors in Pulmonary Carcinoid Tumors

Lung cancer is the leading cause of cancer-related death in the United States (1). Pulmonary carcinoid tumors are malignant neoplasms comprising neuroendocrine cells that account for 2% to 5% of all lung cancers (2). Pulmonary typical carcinoid, which is considered a low-grade tumor, has scarce mitotic figures (<2 per 10 high-powered fields) and absence of necrosis in histologic specimens (3). Pulmonary atypical carcinoid, a higher-grade tumor, has 2 to 10 mitotic figures per 10 high-powered fields, and/or areas of focal necrosis (3). Typical carcinoid tumors usually present at an average age of 45 years, whereas atypical carcinoid tumors present a decade later. Pulmonary carcinoid tumors can develop without any known risk factors for lung cancer, and in particular less than half of patients are cigarette smokers. Epidemiologic data show that pulmonary typical carcinoid tumors are about four times more frequent than atypical carcinoid tumors, and that women are at higher risk than men of developing these tumors.

The 5-year survival for pulmonary typical carcinoid tumors averages 88% and ranges from 25% to 56% for atypical carcinoid tumors (4–6). Surgical resection of the tumor in the affected lung is the standard treatment, but complete surgical excision can be difficult or unattainable depending upon the location of the tumor in the chest and whether it has metastasized. Chemotherapy and radiation therapy have limited success in treating pulmonary carcinoid tumors (7–9). Widely metastatic pulmonary carcinoid tumors can result in carcinoid heart disease or the carcinoid syndrome (10). At the time of diagnosis, up to 30% of patients will have the carcinoid syndrome with symptoms due to excessive production of serotonin from the tumor burden (11). The outcome for patients with widely metastatic disease is poor. A better understanding of the biology of these tumors could lead to the development of novel therapeutic agents to improve patient outcomes.

Altered growth factor signaling through cell surface receptors is thought to contribute to tumorigenesis through abnormal cell proliferation. The ErbB family of receptor tyrosine kinases consists of the epidermal growth factor receptor (EGFR/ErbB1/Her1), ErbB2 (Her2/neu), ErbB3 (Her3) and ErbB4 (Her4; ref. 12). These receptor tyrosine kinases mediate cellular responses to growth factors through their intracellular domain and interact with downstream pathways important for development, differentiation, and proliferation. The EGFR gene is overexpressed and amplified in non–small cell lung cancer (NSCLC), and the presence of somatic mutations in the tyrosine kinase domain of EGFR confers sensitivity to the agents erlotinib and gefitinib (13, 14). ErbB2 (Her2/neu) overexpression in breast and ovarian cancer is associated with poor clinical outcome (15–18). Treatment with trastuzumab (Herceptin) to specifically target ErbB2 (Her2/neu) in breast cancer has resulted in improved patient outcome for this cancer type (19). ErbB3 expression has been found in varying degrees in cancers of the lung, breast, ovary, prostate, and elsewhere (20). ErbB3 is the only member of this receptor tyrosine kinase family that does not have intrinsic kinase activity (21, 22). ErbB3 therefore must heterodimerize with another ErbB receptor for signal transduction. ErbB4 is required for a number of cellular responses, such as development, differentiation, and proliferation. ErbB4 is expressed mainly in developing neural tissues and myocardium (23, 24). The role of ErbB4 in tumorigenesis is complex and not completely understood. In breast cancer ErbB4 expression is low to moderate and its expression correlates with a favorable prognosis (25). ErbB4 expression in ependymoma is high and correlated with worse patient survival (26). In a study of 80 patients with NSCLC who underwent surgery, tumors that expressed ErbB4 had decreased survival compared with tumors that did not express ErbB4 (27). A positive correlation was also found with lymph node metastasis and ErbB4 expression in the tumors in this series. Overexpression of ErbB4 in a NSCLC cell line resulted in increased cell proliferation in comparison with ErbB4-negative cells (27).

In the present study, we define the expression pattern for the ErbB family of receptor tyrosine kinases from archival tumor specimens of 31 patients with pulmonary carcinoid tumors, 24 with typical carcinoid, and 7 with atypical carcinoid, for the first time. Additionally we sequenced exons 18 to 21 of the EGFR gene and the KRAS to determine whether activating mutations are present in these rare tumors. Finally using H727 lung carcinoid cells that express EGFR, we found that erlotinib reduced proliferation of this tumor cell line by inhibiting EGFR signal transduction. Our findings might have clinical application by inhibiting EGFR signaling for patients with EGFR-positive pulmonary carcinoid tumors.

Immunohistochemistry. These studies were approved by the Mayo Clinic Institutional Review Board. The Mayo Clinic Lung Cancer Specimen Registry was searched for tissue of carcinoid tumors from January 1, 2001 to January 31, 2006. Sufficient-quality frozen tissue and the corresponding archived paraffin-embedded tissue were available in 31 patients. All tumors were reviewed and confirmed as typical and atypical carcinoid tumors based on the 2004 WHO classification by an expert lung pathologist experienced in neuroendocrine lung cancers (3). Immunohistochemical stains were done on representative 4-μm formalin-fixed, paraffin-embedded tissue sections from the lung tumors using antibodies to chromogranin (Millipore; clone LK2H10; 1:500 dilution), EGFR (Dako; EGFR pharmDx), phospho-EGFR (Dako), ErbB2/Her2Neu (Dako; HercepTest), ErbB3/Her3 (Santa Cruz Biotechnology; polyclonal, 1:100), and ErbB4/Her4 (Santa Cruz Biotechnology; polyclonal, 1:100). For chromogranin, ErbB3/Her3, and ErbB4/Her4, epitope retrieval was done in a heated 1 mmol/L EDTA pH 8.0 solution for 30 min. Antigen-antibody reactions were visualized using a polymer-based detection system (Dako) using diaminobenzidine as the chromogen. For antibodies to EGFR and ErbB2/Her2Neu, the procedure was done according to the manufacturer's instructions for EGFR pharmDx and HercepTest for the Dako Autostainer. Appropriate positive and negative controls were employed for all conditions. Stains were scored 0 to 4+ based on the percentage of positive tumor cells as follows: 0, <5%; 1+, 5% to 10%; 2+, 11% to 50%; 3+, 50% to 80%; 4+, >80%.

Cell culture. H727, H720, UMC-11, and HCC827 were purchased from the American Type Culture Collection and were grown at 37°C in 5% CO₂ in RPMI-1640 media with 1% antibiotics (penicillin/streptomycin) and 10% fetal bovine serum.

EGFR and KRAS gene sequencing. DNA was extracted from all of the frozen carcinoid tumor specimens using standard procedures. PCR amplification and sequencing of EGFR exon 18 to 21 was done using the identical conditions and PCR primers described by Lynch and colleagues (13). The PCR primers used were (a) exon 18: 5′-CAA-ATG-AGC-TGG-CAA-GTG-CCG-TGTC-3′, and 5′-GAG-TTT-CCC-AAA-CAC-TCA-GTG-AAAC-3′; (b) exon 19: 5′-GCA-ATA-TCA-GCC-TTA-GGT-GCG-GCTC-3′, and 5′-CAT-AGA-AAG-TGA-ACA-TTT-AGG-ATG-TG-3′; (c) exon 20: 5′-CCA-TGA-GTA-CGT-ATT-TTG-AAA-CTC-3′, and 5′-CAT-ATC-CCC-ATG-GCA-AAC-TCT-TGC-3′; and (d) exon 21: 5′-CTA-ACG-TTC-GCC-AGC-CAT-AAG-TCC-3′, and 5′-GCT-GCG-AGC-TCA-CCC-AGA-ATG-TCT-GG-3′. For KRAS the following PCR primers were used: 5′- GTA-CTG-GTG-GAG-TAT-TTG-AT-3′ and 5′-TGA-AAA-TGG-TCA-GAG-AAA-CC-3′. Amplification was done using 0.2 μmol/L primers and 100 ng DNA with the following conditions: 94°C for 5 min, then 30 cycles of 94°C for 1 min, 58°C for 1 min (55°C for KRAS), 72°C for 1 min. PCR amplicons were resolved by electrophoresis on 2% agarose gels stained with ethidium bromide. The PCR amplicons were cut from gel, gene-cleaned, and sequenced with the forward and reverse PCR primers.

Cell proliferation. Cell proliferation was determined using CyQUANT NF Cell Proliferation Assay (Invitrogen) according to the manufacturer's instructions. The assay is designed to produce a linear analytical response from 100 to 20,000 cells per well. In brief, cells were plated at density of 2,000 cells per well in a 96-well plate. After 48 h the cells were stained with the CyQUANT dye for 30 min at 37°C. The fluorescence intensity of each sample was measured using a fluorescence microplate reader (excitation of 485 nm and emission of 530 nm). The data represent averages of triplicate samples with error bars showing SD.

Cell cycle analysis. Cell cycle determination was analyzed by flow cytometry using ModFitLT software (Verity Software). H727 cells were trypsinized, washed with sterile PBS, and resuspended at 100,000 cells/mL. The cells were fixed in 70% ethanol and stained with propidium iodide (mixture of 2.5 mg/mL propidium iodide with 0.1 mg/mL RNase A and 0.05% Triton X-100) for 40 min at 37°C. The samples were analyzed by flow cytometry (Becton Dickinson FACScan; BD Bioscience).

Immunoblotting. Cellular lysates of H727 and HCC827 cells were prepared by gentle sonication on ice in lysis buffer [50 mmol/L Tris-HCl pH 7.5, 100 mmol/L NaCl, 50 mmol/L NaF, 1% (v/v) NP-40, 5 mmol/L EDTA, 1 mmol/L EGTA, 200 μmol/L sodium orthovanadate, 50 μmol/L β-glycerolpyrophosphate, 100 μmol/L phenylmethylsulfonyl fluoride, 200 μmol/L sodium fluoride, 1 μg/mL of each of leupeptin, aprotinin and pepstatin]. The protein concentration was determined spectrophotometrically using the BCA method (Pierce Biotechnology, Inc.), and equal protein was resolved by 8% SDS-PAGE, and transferred to polyvinylidene fluoride membranes. Nonspecific binding sites were blocked with TBS containing 5% milk prior to addition of the primary antibodies. Bound antibodies were detected with enhanced chemiluminescence (Amersham Biosciences).

Statistical analysis. Experiments were carried out in triplicate for all experimental conditions. Data are shown as mean ± SD. Student's t test was done on the means of two sets of sample data and considered significant if P = 0.05.

Patient characteristics. We examined the clinical characteristics of 31 patients with pulmonary carcinoid tumors diagnosed from January 1, 2001 to January 31, 2006 for which there was available fresh-frozen and paraffin-embedded tissue for study (Table 1). There were 24 patients with pulmonary typical carcinoid tumors and 7 with atypical carcinoid tumors. The entire cohort of subjects included 11 men and 20 women with a mean age of 58 years (range, 31-77 years). Overall follow-up for the entire cohort was a median time of 50 months (range, 1-84 months). Eleven patients (35%) were former cigarette smokers with an average consumption of 30 pack-years. Twenty tumors (64.5%) were located in the right chest. The most common location was the right middle lobe (nine tumors), followed by the right lower lobe (eight tumors), then the left lower lobe (five tumors). One patient had bilateral right and left upper lobe tumors. Evidence of hypercalcemia was present in 5 (16%) patients. Two patients with ectopic production of adrenocorticotropic hormone were diagnosed with Cushing's syndrome 1 month and 11 months prior to the diagnosis of pulmonary carcinoid tumor. There were a total of 16 additional malignancies in 13 (42%) patients, some of which preceded the diagnosis of pulmonary carcinoid tumor, and others that developed afterwards. There were three skin cancers (two melanoma, one squamous cell), two gastrointestinal stromal tumors, two prior pulmonary typical carcinoid tumors, two breast cancers, and prostate cancer, colon cancer, ovarian cancer, papillary thyroid cancer, parathyroid adenoma, polycythemia vera, and multiple uterine leiomyomas. This finding of the association with other malignancies in 42% of the patients has been identified in other cohorts of patients with pulmonary carcinoid tumors and highlights the need for continual cancer surveillance for these patients (28, 29). No patients had MEN 1.

Among the 24 patients with pulmonary typical carcinoid tumors, 11 had stage IA disease (T₁N₀M₀), 6 had stage IB disease (T₁N₁M₀), 2 had stage IIA disease (T₂N₀M₀), 2 had stage IIB disease (T₂N₁M₀), 1 had stage IIIA disease (T₁N₂M₀), and 2 had stage IIIB disease (both T₄N₀M₀). All 24 patients are alive and have no evidence of disease at follow-up with a median time of 50 months.

Among the seven patients with atypical carcinoid tumors, four patients had stage IA disease (T₁N₀M₀), one had stage IB disease (T₁N₁M₀), one had stage IIIA disease (T₁N₂M₀), and one had stage IV disease (T₁N₂M₁) with metastasis to the liver. All seven patients are alive at follow-up with a median time of 48 months. There was no evidence of disease at the time of follow-up in five (71 %) patients. One patient developed liver metastasis 35 months after resection of the primary lung tumor.

The impressive survival in our cohort of patients is likely due to several factors, including the early stage of the majority of the tumors, the expertise of the team of physicians and surgeons caring for these patients, and selection bias. Indeed our study only encompasses tumors resected from 2001 to 2006 where we had matched frozen and paraffin-embedded tissue available to study.

ErbB receptor expression in pulmonary carcinoid tumors. Positive staining for EGFR was observed in 11 of 24 (45.8 %) typical carcinoid tumors and in 2 of 7 (28.6 %) atypical carcinoid tumors (Figs. 1A to D; Table 2). Five typical carcinoid tumors had 0 to 1+ EGFR immunoreactivity, whereas six had 2 to 4+ immunoreactivity. Both of the atypical carcinoid tumors had 2 to 4+ immunoreactivity. The EGFR immunoreactivity was present exclusively in the epithelial cell membranes except in one typical carcinoid tumor which had low-level membranous and cytoplasmic staining for EGFR. Additional slides were prepared for phospho-EGFR staining from the subset of tumors with EGFR immunoreactivity, but we did not observe staining with phospho-EGFR from any of these samples. Immunoreactivity for the ErbB2 receptor was absent in all samples of typical carcinoid and atypical carcinoid tumors. Receptor expression for the ErbB3 and ErbB4 receptors was intensely positive in all samples of typical carcinoid and atypical carcinoid tumors. In contrast to EGFR expression, which was present in the epithelial cell membranes, the immunoreactivity for ErbB3 and c- ErbB4 receptors was entirely cytoplasmic in all of the carcinoid tumor sections.

Pulmonary carcinoid tumors lack activating EGFR tyrosine kinase mutations. In a subset of patients with NSCLC, the presence of somatic mutations in the tyrosine kinase domain of the EGFR gene is associated with sensitivity to the small-molecule EGFR inhibitors gefitinib and erlotinib (13, 14). We examined whether pulmonary carcinoid tumors have mutations in exons 18 to 21 of the EGFR gene by PCR amplification and sequencing of tumor specimen DNA. We found no activating mutations in the EGFR kinase domain from the 24 typical carcinoid and 7 atypical carcinoid tumors. Interestingly the sequences were not all wild-type; 19 (79.2 %) typical carcinoid and 6 (85.7 %) atypical carcinoid tumors had a unique single nucleotide polymorphism (SNP) in exon 20 (Fig. 2). This synonymous SNP, a G to A substitution, does not change the protein coding sequence of EGFR.

We compared the characteristics of the pulmonary carcinoid tumors that express EGFR with those that do not express EGFR. No statistically important correlations were observed, likely due to the small sample size. Specifically there was no difference in age, gender, tobacco use, histology, stage, recurrence, presence of the exon 20 SNP, or outcome.

Pulmonary carcinoid tumors lack KRAS mutations. Activating KRAS mutations are present in 15% to 30% of NSCLC, especially lung adenocarcinomas, and are associated with poor outcome (30). The most frequent mutations are found in codons 12 and 13 in exon 2 and result in constitutively active KRAS. EGFR mutations are common in tumors from patients who have never smoked or have minimal cigarette consumption, whereas KRAS mutations are more frequent in patient tumors with substantial cigarette consumption (31). Further, KRAS mutations have been associated with resistance to anti-EGFR treatment (32). EGFR and KRAS mutations are rarely found in the same tumors, and as discussed above none of the 31 pulmonary carcinoid tumors had EGFR tyrosine kinase domain mutations. We therefore examined whether these tumors have KRAS gene mutations by PCR amplification and sequencing of tumor specimen DNA. We found no mutations present in the KRAS gene from the 24 typical carcinoid and 7 atypical carcinoid tumors.

The reagents available for studying pulmonary carcinoid tumor biology in the laboratory are limited to a few cell lines. We obtained two “typical” lung carcinoid cell lines, H727 and UMC-11, and H720, an “atypical” lung carcinoid cell line. Although these cell lines express neuroendocrine markers, it is uncertain whether these cell lines are actually typical or atypical in histology because they were created prior to the modern WHO classification for pulmonary carcinoid tumors (3). We examined the ErbB receptor expression in H720, H727, and UMC-11 cell lines to determine whether these cell lines could be used as a model of EGFR-positive pulmonary carcinoid tumors. We used HCC827 NSCLC cells as a control, because these cells contain the L858R-activating EGFR tyrosine kinase domain mutation, have amplification of EGFR, and have been shown to be highly sensitive to EGFR inhibition (33). We found that H727 lung carcinoid cells, but not H720 or UMC-11 cells, express EGFR (Fig. 3A). The EGFR gene from H727 cells had previously been sequenced by Lynch and colleagues and shown to be without any tyrosine kinase domain mutations, a finding similar to the 31 tumors in our series (13). In a genetic screen of human cancers and cell lines for mutations in RAS and RAF genes, the H727 cancer cell line was found to have the activating KRAS2 G12V mutation and a normal B-RAF gene (34). In the subsequent experiments we used the H727 cell line as a model for EGFR-positive pulmonary carcinoid tumors.

Phosphorylation status of EGFR and effect of EGFR activation on downstream signaling pathways in H727 cells. H727 cells growing in the absence of exogenous EGF were lysed, immunoprecipitated with an EGFR antibody, and blotted with an EGFR antibody or a phosphotyrosine (pY) antibody. As shown in Fig. 3B, H727 cells express EGFR that has minimal baseline phosphorylation. Because activated EGFR contains multiple tyrosine phosphorylation sites, we used a phospho-specific antibody to Tyr1068, an established EGFR autophosphorylation site, to show that the basal level of EGFR phosphorylation increases significantly with EGF stimulation (100 ng/mL for 10 minutes) of the cells (Fig. 3C). There was no apparent increased EGFR phosphorylation when the H727 cells were treated for 10 minutes with 100 ng/mL heregulin, a ligand for ErbB2-ErbB3 heterodimers and ErbB4, but is not recognized by EGFR.

Next we examined the effect of EGF stimulation of H727 cells in the context of EGFR phosphorylation and downstream signaling pathways (Fig. 3D). EGFR phosphorylation increased from baseline in H727 cells at 10 minutes and persisted for 24 hours, whereas in the control HCC827 cells the activated level of EGFR phosphorylation at baseline increased significantly by 30 minutes and then decreased after 6 hours. In the H727 cells we observed initial rapid extracellular signal-regulated kinase (ERK) phosphorylation at 5 minutes followed by absent ERK activity at all other time points. ERK phosphorylation in the HCC827 cells paralleled the phosphorylation of EGFR, increasing from 30 minutes to 6 hours, then decreasing to low levels. AKT phosphorylation in both the H727 and HCC827 cells showed increased phosphorylation from 30 minutes to 3 hours in the H727 cells and 10 minutes to 1 hour in the HCC827 cells, followed by loss of phosphorylation.

Erlotinib inhibition of EGFR activity in H727 cells results in decreased proliferation. We examined the effect of erlotinib inhibition of EGFR activity and its effect on downstream signaling pathways in H727 cells that were treated with various concentrations of erlotinib for 3 hours followed by stimulation with EGF for 60 minutes (Fig. 4A). We observed a dose- dependent decrease in EGFR phosphorylation in cells treated with increasing concentrations of erlotinib. There seemed to be a decrease in AKT phosphorylation in both the H727 and HCC827 cells. Erlotinib treatment of the H727 cells resulted in increased ERK phosphorylation, whereas in HCC827 cells the level of ERK phosphorylation decreased substantially. Next we examined the effect of erlotinib on the proliferation of these cells following treament for 48 hours. We observed a statistically significant reduction in H727 cell proliferation in a dose-dependent manner with erlotinib (Fig. 4B). H727 proliferation decreased by 32% with 0.1 μmol/L of erlotinib, 48% with 1 μmol/L of erlotinib, and 52% with 10 μmol/L of erlotinib. For any given dose of erlotinib, the HCC827 cells were more sensitive and had a greater reduction in proliferation compared with H727 cells. To investigate the mechanism of reduced cell proliferation in H727 cells due to EGFR inhibition, we examined their cell cycle transition and found that H727 cells treated with erlotinib developed a G₂/M cell cycle arrest (Fig. 4C).

Prior to our study there were two reports evaluating either the expression of EGFR or ErbB2 in pulmonary carcinoid tumors. Rusch and colleagues carried out immunohistochemistry for EGFR and found moderate to high staining (2+ to 4+ representing 20% to >80% positive cells) in 7 of 25 (28%) pulmonary typical carcinoid tumors and in 4 of 7 (57%) atypical carcinoid tumors (35). Wilkinson and colleagues determined there was no expression of ErbB2 in 26 pulmonary typical carcinoid and 23 atypical carcinoid tumors (36). Using the same scoring criteria as Rusch, we found moderate to high staining in 5 of 24 (21%) pulmonary typical carcinoid tumors and in 2 of 7 (29%) atypical carcinoid tumors. Six additional typical carcinoid tumors in our series had low-level staining ranging from 1% to 15% for EGFR. None of the tumors with EGFR staining had phospho-EGFR staining, suggesting that the EGFR pathway is not constitutively active in these tumors. Similarly we observed absent phospho-EGFR in the H727 carcinoid cell line, but erlotinib inhibition of EGFR activity in these cells resulted in decreased proliferation despite the baseline phospho-EGFR activity. Our study results are also concordant with the findings reported by Wilkinson showing no expression of ErbB2. Our report is the first to include staining for ErbB3 and ErbB4, which revealed 100% immunoreactivity in all of the typical and atypical pulmonary carcinoid tumors. The function of ErbB3 and ErbB4 in these tumors is unknown and provides an area for future investigation.

Other novel findings include the absence of EGFR tyrosine kinase domain mutations and the absence of KRAS mutations in all of our patient samples. The exon 20 G to A synonymous SNP was present in a high proportion of these tumors. Most synonymous SNPs create silent mutations without any functional significance, but there have been reports of synonymous SNPs affecting the stability of mRNA (37, 38). The significance of this exon 20 SNP is unknown, and did not correlate with the expression of EGFR or other patient characteristics.

We found that the kinetics of phosphorylation of EGFR in H727 lung carcinoid cells was similar to the control HCC827 cells in that sustained phosphorylation lasted for hours after EGF stimulation. This finding could be the result of altered phosphatase activity in the cells, or other factors involved in receptor recycling. Signaling to downstream kinases after EGF stimulation showed transient ERK phosphorylation in H727 cells, whereas HCC827 cells developed sustained ERK phosphorylation which is an effect of the L858R-activating EGFR tyrosine kinase domain mutation in these cells. In both H727 and HCC827 cells, AKT seemed to be constitutively phosphorylated at baseline with little increase with EGF stimulation.

Erlotinib, an oral reversible inhibitor of EGFR tyrosine kinase activity, showed improved survival compared with placebo in the BR.21 trial as monotherapy in previously treated advanced NSCLC patients (39). Gefitinib, however, failed to show a survival benefit as monotherapy in previously treated advanced NSCLC in the ISEL trial (40). In our study erlotinib treatment revealed important differences in the cellular response of the H727 lung carcinoid cells compared with the HCC827 NSCLC cells. Amann and colleagues previously reported that the mechanism of reduced proliferation in HCC827 cells treated with erlotinib occurred through reduced ERK and AKT phosphorylation, which led to a G₁ cell cycle arrest (33). In H727 cells we observed that treatment with erlotinib reduced EGFR and AKT phosphorylation but resulted in ERK activation and the development of a G₂/M cell cycle arrest. A similar finding of ERK activation resulting in cell cycle arrest has been reported in breast cancer and leukemia (41, 42). H727 lung carcinoid cells do not have EGFR tyrosine kinase domain mutations, yet they have a KRAS G12V mutation. Under normal unstimulated growth conditions in H727 cells, EGFR has minimal constitutive phosphorylation, and although this cell line has a KRAS G12V mutation, we did not observe constitutive ERK phosphorylation. Interestingly we noted that ERK became phosphorylated only when EGFR signaling was inhibited with erlotinib. In the absence of EGFR tyrosine kinase domain mutations, the absolute level of EGFR has been found to be an important independent factor for the response to treatment with EGFR tyrosine kinase inhibitors (43, 44). We found moderate to high staining for EGFR in 21% of the pulmonary typical carcinoid tumors and in 29% of the atypical carcinoid tumors from the 31 patient samples, and high expression of EGFR in the H727 cell line by Western blotting. Although the patient samples and H727 cells lack EGFR tyrosine kinase domain mutations, the high expression of EGFR may be responsible for the sensitivity to erlotinib.

Surgical resection of the tumor mass remains the optimal treatment for pulmonary carcinoid tumors. Complete surgical excision, including dissection of regional lung and mediastinal lymph node metastases, should be attempted to remove all of the tumor tissue for a cure. Depending upon the location of the tumor, however, this may not be possible. Lung preservation surgery, utilizing bronchial sleeve procedures, requires specialized surgical expertise (45–47). The role of adjuvant therapy following the complete surgical resection of a pulmonary carcinoid tumor is undefined due to the small series and retrospective nature of the reports. There are no prospective trials that directly address the benefit of chemotherapy or radiation therapy for patients with pulmonary typical carcinoid or atypical carcinoid tumors. Our findings might have clinical application in the treatment of patients with pulmonary carcinoid tumors. The inhibition of EGFR signaling may represent a potential therapeutic strategy in the treatment of patients with EGFR-positive pulmonary carcinoid tumors, especially for those patients whose cancer cannot be surgically resected, or in patients presenting with advanced disease.

No potential conflicts of interest were disclosed.