Original article. Epidermal growth factor receptor overexpression correlates with a poor prognosis in completely resected non-small-cell lung cancer

Transcription

1 Original article Annals of Oncology 15: 28 32, 2004 DOI: /annonc/mdh011 Epidermal growth factor receptor overexpression correlates with a poor prognosis in completely resected non-small-cell lung cancer G. Selvaggi 1 *, S. Novello 1, V. Torri 2, E. Leonardo 3, P. De Giuli 3, P. Borasio 4, C. Mossetti 4, F. Ardissone 4, P. Lausi 4 & G. V. Scagliotti 1 1 Thoracic Oncology Unit, Department of Clinical and Biological Sciences, University of Torino; 2 Mario Negri Institute, Milano and Departments of 3 Pathology and 4 Thoracic Surgery, Azienda Ospedaliera San Luigi, Orbassano, Torino, Italy Received 26 February 2003; revised 20 May 2003; accepted 13 August 2003 Background: We designed a prospective study to test epidermal growth factor receptor (EGFR) expression by immunohistochemistry (IHC) in resected stage I IIIA non-small-cell lung cancer (NSCLC) and to correlate overexpression with survival. Patients and methods: EGFR expression was evaluated in 130 consecutive NSCLC patients after radical surgery (60 squamous cell carcinomas, 48 adenocarcinomas, 22 large cell carcinomas: stage I, 41 (31%); stage II, 37 (29%) and stage IIIA, 52 (40%). Results: Overall, 101 of 130 (78%) specimens expressed EGFR, and with a cut-off value of 10% positive cells 48 cases (37%) were classified as positive. At univariate analysis, EGFR was significantly more expressed in stage III (50%) than stage I (20%) and stage II (25%) (P <0.03). No correlation with histotype was found. After a median follow-up of 84 months, both median survival time (18 versus 50 months), 2-year (43% versus 70%) and 5-year (31% versus 46%) survival rates of positive cases were significantly lower than negative ones [P <0.001; hazard ratio 1.96; 95% confidence interval (CI) ]. At the multivariate analysis, EGFR overexpression and stage emerged as independent factors for cancer-related mortality. Conclusion: In patients with radically resected stage I IIIA NSCLC, EGFR overexpression predicts shorter survival, thus representing a valuable prognostic factor. Key words: epidermal growth factor receptor, non-small-cell lung cancer, prognostic factors, survival Introduction *Correspondence to: G. Selvaggi, Thoracic Oncology, Department of Clinical and Biological Sciences, University of Torino, Azienda Ospedaliera S. Luigi, Regione Gonzole, 10, Orbassano, Torino, Italy. Tel: ; Fax: ; gselvaggi@inwind.it The family of epidermal growth factor receptors (HER-1, HER-2/ neu, HER-3 and HER-4) includes cell membrane receptors with intrinsic tyrosine kinase activity which can transduce a proliferation signal in response to the binding of different ligands. These receptors play a key role in malignant proliferation of cells in a variety of human tumors [1]. After ligand receptor binding, epidermal growth factor receptor (EGFR) undergoes dimerization and activation of tyrosine kinase occurs, with receptor autophosphorylation, downstream signal transduction through activation of Ras and MAP kinase and ultimately gene activation leading to cell proliferation. EGFR is a 170 kda membrane glycoprotein, with an extracellular ligand binding domain, a transmembrane lipophilic part, and an intracellular domain with tyrosine kinase activity [2]. EGFR was found to be overexpressed in both cell lines and samples of non-small-cell lung cancer (NSCLC) [3 5], and associated with increased tumor proliferation, poor differentiation, higher incidence of metastases to lymph nodes and a worse prognosis [6]. Higher rates of EGFR expression were also found in stage III disease [7] and in squamous cell histotype [8, 9]. The mechanism responsible for EGFR overexpression is largely unknown and gene amplification is only rarely involved in NSCLC [10]. The prognostic significance of EGFR remains to be defined: most of the studies analyzing molecular markers are affected by either a retrospective design or by a small sample size with limited statistical power. Therefore, we conducted a prospective study, starting in 1991, to evaluate EGFR expression by immunohistochemistry (IHC) on 130 consecutive surgical specimens from early stage NSCLC patients and to correlate overexpression with long-term survival. In a previous paper [11] we also analyzed overexpression of HER-2/neu in the same population of 130 patients. Patients and methods All consecutive patients undergoing resection for NSCLC at the Department of Thoracic Surgery, San Luigi Hospital, Orbassano, Italy from January 1991 to February 1992 were included in this prospective study designed to evaluate surgical specimens for the presence and significance of EGFR and HER-2/neu. Surgery was intended to be radical for all patients and was performed by the same surgical team. Histology was determined according to World Health Organization (WHO) criteria [12]. Post-surgical pathological stage was 2004 European Society for Medical Oncology

2 29 determined according to the TNM (tumor node metastasis) staging system in use when the study started [13]. None of the patients received pre-operative chemotherapy or radiotherapy. Adjuvant treatment was limited to thoracic radiotherapy in pn2 cases. All patients were followed up every 3 months for the first 2 years, then on a 6-monthly basis for 3 years. Treatment at relapse was decided individually and included standard chemotherapy or radiotherapy according to progression patterns. EGFR assay EGFR expression was determined by IHC (avidin-biotin complex). Specimens from surgically removed tumors were 10% formalin-fixed, then paraffinembedded, and 5 µm sections were cut from tissue blocks, placed on pretreated glass slides, then kept at 60 C for 30 min, dewaxed and rehydrated with graded alcohol. An antigen retrieval procedure was performed using a 0.1% pepsin solution at ph 2.25, at 37 C for 20 min. After inhibition of endogenous peroxidase, which occurred after a 30-min treatment with 3% hydrogen peroxide in methanol, EGFR status was assessed using a 1/40 diluted mab anti-egfr (Ab-1, Oncogene Sciences, USA) at room temperature for 1 h. Then, 100 µl of a rabbit anti-mouse biotinylated antibody (LSAB, Dako, USA) were added and incubated for 30 min at room temperature. Subsequently, sections were incubated for 45 min with a preformed avidin biotinylated horseradish peroxidase macromolecular complex (Vectastain, Vector Laboratories, USA). Final staining was provided by means of 3,3 diaminobenzidine tetrahydrochloride. Finally, sections were counterstained with hematoxylin. Negative controls were obtained omitting anti-egfr antibody and using anti-sheep IgG as a primary antibody. It was considered as positive only cell membrane staining. An average number of 1500 cells per section was evaluated utilizing a semiquantitative grading system based on four stages (0, no staining; 1+, staining in 1 10% of considered cells; 2+, staining in 11 25% of considered cells; 3+, staining in >25% of considered cells). Microscopic analyses were all performed by the same pathologists (E.L. and P.D.G.). Samples of macroscopically normal lung tissue taken from surgery were also examined for EGFR expression. A cut-off value of 10% positive cells was used in order to avoid inclusion of scattered positivity of the same intensity found in normal bronchial tissue (see Results); this cut-off value was chosen before analyzing the data on the basis of preliminary data on EGFR expression from normal bronchial tissue. Statistical analysis The clinical pathological variables considered were sex, age, histotype and pathological TNM stage. Overall survival curves were calculated from the date of surgery to the time of death related to NSCLC or time of last follow-up observation. Kaplan Meier curves were calculated for each relevant variable and for EGFR expression. Univariate and multivariate analyses were performed with the Cox proportional hazard regression model [14]. In the multivariate model, a backward selection procedure on variable, describing patients and disease characteristics was chosen in order to identify the most parsimonious model for predicting cancer-related mortality. The variable EGFR was then added to the model in order to verify the association between EGFR and mortality after controlling for the baseline variables associated with prognosis. The number of events at the time of analysis allowed 80% power to detect a statistically significant result (5%, two tails) for a hazard ratio of three associated to the group with >10% EGFR positive cells. Results From January 1991 to February 1992, 130 consecutive patients entered the study and underwent resection for curative intent. All resections (75 lobectomies, 46 pneumonectomies, seven bilobectomies and two atypical resections) were performed by the same Table 1. Patients and disease characteristics No. of patients 130 Male/female 120/10 Median age, years (range) 60 (39 76) Stage I 41 (31%) II 37 (29%) III 52 (40%) Histology Squamous carcinoma 60 (46%) Adenocarcinoma 48 (37%) Large cell 22 (17%) Type of resection Lobectomy 75 (58%) Pneumonectomy 46 (35%) Other 9 (7%) surgical team. Patients characteristics are summarized in Table 1. At the time of the analysis 51 patients (39%) were still alive, 79 had died (61%): five patients (4%) died in the post-operative period, 10 patients (8%) died of unrelated causes during follow up, and in 69 patients (88%) the cause of death was tumor progression. Normal bronchial tissue samples taken from the resected lobe (or lung) at the time of surgery were either negative or weakly positive for EGFR only in the basal layer of the cell membrane (<10% of examined cells). Overall, 101 tumors out of 130 (78%) expressed EGFR. When adopting a cut-off value of >10% positive cells (2+, 3+), 48 patients (37%) overexpressed EGFR. EGFR overexpression according to the pathological stage was 20% in stage I patients, 25% in stage II and 50% in stage IIIA: the higher incidence in stage IIIA patients was statistically significant (P <0.03). No correlation was found between EGFR overexpression and histology: it was detected in 30% of squamous cell carcinomas, 38% of adenocarcinomas and 56% of large cell carcinomas. After a median follow-up time of 84 months, overall 5-year survival was 39%, with a median survival time of 37 months. A higher pathological stage significantly predicted decreased overall survival (P <0.001). In patients overexpressing EGFR, median survival time was significantly shorter: 18 versus 50 months [P <0.02; hazard ratio 1.96; 95% confidence interval (CI) ]. Twoyear and 5-year survival were, respectively, 43% and 31% for EGFR 2+, 3+ patients and 70% and 46% for EGFR 0, 1+ patients (P <0.001) (Figure 1). At univariate analysis, association of stage III, adenocarcinoma, pneumonectomy and EGFR overexpression (2+, 3+) with shorter survival was statistically significant. At multivariate analysis, stage III (P <0.03) and EGFR overexpression (2+, 3+) (P <0.03) were independent prognostic factors for reduced survival (Table 2). Concomitant overexpression of EGFR and HER-2/neu was found in four patients (3%), 44 patients (34%) overexpressed only

3 30 Figure 1. Survival curve according to epidermal growth factor receptor (EGFR) expression. EGFR, 11 patients (8%) overexpressed only HER-2/neu and 71 patients (55%) expressed both receptors below the cut-off values. Discussion In 2003, lung cancer is still the primary cause of death from malignancy in both sexes [15]; 5-year survival does not go beyond 70% even in the most favorable group of radically resected stage I patients [16]. Adjuvant therapies have not shown any impact on overall survival so far. Early diagnosis and careful selection of higher-risk subgroups will enable therapies to be tailored to individual patients. Thus, prognostic factors are strongly needed to define more aggressive tumors: a wide array of biomarkers have been studied recently and EGFR has been the focus of most researchers attention [17]. In 1991, we designed a prospective study to evaluate EGFR expression in radically resected patients with stage I III NSCLC. In the case of strong positivity at IHC, scored as 2+ to 3+, EGFR overexpression is significantly predictive of a worse prognosis. In our previous paper [11] we reported that overexpression of HER-2/neu, a closely related receptor of the family of EGFR, is a negative prognostic factor in the same population. Concomitant overexpression of the two receptors was observed in 3% of patients from our series: overexpression of either one receptor maintained its negative predictive power (survival data not shown). In our series no correlation was found between EGFR overexpression and histology: no difference was seen between adenocarcinomas and squamous cell carcinomas in contrast with studies in which a higher expression of EGFR was observed in squamous cell carcinomas [7, 18]. However, EGFR was predictive of a worse outcome in adenocarcinomas as well [19]. We found a significantly higher expression of EGFR in stage III when compared to earlier stages. These findings have been reported in other series [7, 20]. A tentative hypothesis could be that expression increases stepwise from pre-cancerous lesions to more advanced stages of cancer [21]. In a recently published meta-analysis [22], over 2000 patients were included and results of 11 studies were compared. IHC was the most frequently used method: a wide range of monoclonal antibodies was used with different dilutions. In eight of the 11 studies using IHC, EGFR overexpression confirmed a worse prognostic significance (hazard ratio 1.13) even if cut-off values are usually resulting from arbitrary choices of investigators. In the review from Nicholson et al. [23] EGFR overexpression confirmed its prognostic value in multiple tumor types, but evidence was weaker in NSCLC. However, as Nicholson et al. suggest [23],

4 31 Table 2. Univariate and multivariate analysis for variables considered (Cox proportional hazard regression model) Univariate analysis P value Hazard ratio (95% CI) Male ( ) Stage III ( ) Adenocarcinoma versus squamous carcinoma Large cell versus squamous carcinoma ( ) 2.13 ( ) EGFR >10% (2+, 3+) ( ) Multivariate analysis Stage III ( ) EGFR >10% (2+, 3+) ( ) CI, confidence interval; EGFR, epidermal growth factor receptor. the true prognostic significance of EGFR might be underestimated by the fact that in published studies EGFR is assessed as total cellular level rather than its activated form, which is probably the only form affecting prognosis [24]. A further bias is the lack of standardized cut-off points of normal receptor levels and the inclusion of both early and late stages of disease in patient populations. The heterogeneity of available reports could also be explained by differences in interpreting the intensity of expression and the localization of receptors and by the wide range of methods in use for EGFR detection. IHC relies on subjective judgment which represents an intrinsic limit of the technique: with IHC some authors reported only cell membrane staining [25] as opposed to cytoplasmic staining [19, 26], while others did not report any preferential localization of the receptor [27]. As already seen from HER-2/neu studies [28, 29], differences may arise from using IHC or FISH or probing of DNA or mrna, which measures either the protein level or the gene amplification. Even a quantitative detection cannot fully define the real drive of EGFR on the tumor proliferation in vivo as mentioned above. Standardization of techniques to determine EGFR overexpression must therefore become a priority in the near future; IHC remains in our opinion the best choice for routine clinical use, even if a universal scoring system is still needed to better compare research results. Long-term prognosis could also be affected by multiple downstream steps involved in the EGFR signal transduction pathway, as Ras, raf, or MAP kinases. EGFR is activated by the binding of its ligands: EGF, transforming growth factor alpha (TGFα) and HB-EGF (amphiregulin). An autocrine loop sustained by TGFα is another mechanism involved in uncontrolled cellular proliferation [30]. We found that overexpression of EGFR correlates with a more aggressive behavior of the tumor, leading to shorter survival. However, it is not clearly defined if this will translate into a higher sensitivity to agents targeting EGFR [31]. In a series of 169 patients, EGFR overexpression was not found to be prognostic but became predictive of a worse outcome when associated with expression of matrix metalloproteinase 9 (MMP-9) [32]. A possible mechanism to explain this finding is that in vitro EGFR stimulation leads to activation of ras with subsequent up-regulation of MMP-9 [33]. At the time of the study, we did not plan to investigate the impact of mutant EGFR (EGFRvIII), which seems to define a more aggressive phenotype [34]. The EGFRvIII is a truncated EGFR that lacks extracellular domains I and II with a constitutively activated tyrosine kinase independent of ligand interaction [35]. Multiple variables affect the transforming power of EGFR, ranging from modulation of phosphorylation of EGFR kinase substrates to the presence of receptor mutant phenotypes, to the impact of heterodimeric coupling [36] or of more distal steps of the signal transduction pathway. The most recent clinical data showed that EGFR targeted therapies could not be dependent on intensity of expression to be effective [37, 38], as opposed to previous insights from head and neck cancer cell lines [39]. In conclusion, EGFR seems to be a valuable prognostic marker in stage I III NSCLC. Evaluation of EGFR expression, as well as other members of the EGF family of receptors, could contribute to a molecular classification of lung cancer and could define a subset of patients at higher risk for relapse after radical resection, who could hopefully benefit from adjuvant treatment with targeted agents against EGFR. Acknowledgements The authors acknowledge the work of Dr L. Gubetta, Head of the Department of Pathology, Azienda Ospedaliera San Luigi, Orbassano, for reviewing the discrepancies between tissue slides. References 1. Wells A. The epidermal growth factor receptor (EGFR) a new target in cancer therapy. Signal 2000; 1: Ullrich A, Schlessinger J. Signal transduction by receptor with tyrosine kinase activity. Cell 1990; 61: Brabender J, Danenberg KD, Metzger R et al. Epidermal growth factor receptor and HER2-neu mrna expression in non-small cell lung cancer is correlated with survival. Clin Cancer Res 2001; 7: Pastorino U, Andreola S, Tagliabue E et al. Immunohistochemical markers in stage I lung cancer: relevance to prognosis. J Clin Oncol 1997; 15: Rusch V, Klimstra D, Venkatyraman E et al. Overexpression of the epidermal growth factor receptor and its ligand TGF alpha is frequent in resectable non-small cell lung cancer but does not predict tumor progression. Clin Cancer Res 1997; 3: Volm M, Dring P, Wodrich W. Prognostic significance of the expression of c-fos, c-jun, and c-erbb1 oncogene products in human squamous cell lung carcinomas. J Cancer Res Clin Oncol 1993; 119: Veale D, Ashcroft T, Marsch C et al. Epidermal growth factor receptors in non-small cell lung cancer. Br J Cancer 1987; 55: Fontanini G, De Laurentiis M, Vignati S et al. Evaluation of epidermal growth factor-related growth factors and receptors and of neoangiogenesis in completely resected stage I IIIA non-small cell lung cancer: amphiregulin and microvessel count are independent prognostic indicators of survival. Clin Cancer Res 1998; 4: Ohsaki Y, Tanno S, Fujita Y et al. Epidermal growth factor receptor expression correlates with poor prognosis in non-small cell lung cancer patients with p53 overexpression. Oncol Rep 2000; 7:

5 Reinmuth N, Brandt B, Kunze WP et al. Ploidy, expression of erbb1, erbb2, P53 and amplification of erbb1, erbb2, erbb3 in non-small cell lung cancer. Eur Resp J 2000; 16: Selvaggi G, Scagliotti GV, Torri V et al. HER2/neu overexpression in patients with radically resected non-small cell lung carcinoma. Cancer 2002; 94: World Health Organization. Histological typing of lung tumors. Tumori 1981; 67: Mountain CF. The new international staging system for lung cancer. Surg Clin North Am 1987; 67: Cox DR. Regression models and life tables (with discussion). J R Stat Soc B 1972; 34: Jemal A, Thomas A, Murray T, Thun M. Cancer statistics CA Cancer J Clin 2002; 52: Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997; 111: Mendelsohn J, Baselga J. The EGF receptor family as targets for cancer therapy. Oncogene 2000; 19: Hendler FJ, Ozanne BW. Human squamous cell lung cancers express increased epidermal growth factor receptors. J Clin Investig 1984; 74: Pavelic K, Banjac Z, Pavelic J, Spaventi S. Evidence for a role of EGF receptor in the progression of human lung carcinoma. Anticancer Res 1994; 13: Fontanini G, Vignati S, Bigini D et al. Epidermal growth factor receptor (EGFr) in non-small cell lung carcinomas correlates with metastatic involvement of hilar and mediastinal lymph nodes in the squamous subtype. Eur J Cancer 1995; 31: Piyathilake CJ, Frost AR, Manne U et al. Differential expression of growth factors in squamous cell carcinoma and precancerous lesions of the lung. Clin Cancer Res 2002; 8: Meert AP, Martin B, Delmotte P et al. The role of EGF-R expression on patient survival in lung cancer: a systematic review with meta-analysis. Eur Respir J 2002; 20: Nicholson RI, Gee JMW, Harper ME. EGFR and cancer prognosis. Eur J Cancer 2001; 37: S9 S Magne N, Fischel JL, Dubeuil A et al. Influence of epidermal growth factor receptor (EGFR), p53 and intrinsic MAP kinase pathway status of tumor cells on the antiproliferative effect of ZD1839 ( Iressa ). Br J Cancer 2002; 86: Rusch V, Baselga J, Cordon-Cardo C et al. Differential expression of the epidermal growth factor receptor and its ligands in primary non-small cell lung cancers and adjacent benign lung. Cancer Res 1993; 53: Pfeiffer P, Clausen PP, Andersen K, Rose C. Lack of prognostic significance of epidermal growth factor receptor and the oncoprotein p185- HER-2 in patients with systemically untreated non-small cell lung cancer: an immunohistochemical study on cryosections. Br J Cancer 1996; 74: Gorgoulis V, Aninos D, Mikou P et al. Expression of EGF, TGF alpha and EGFR in squamous cell lung carcinomas. Anticancer Res 1992; 12: Cox G, Vyberg M, Melgaard B et al. Herceptest: HER2 expression and gene amplification in non-small cell lung cancer. Int J Cancer 2001; 92: Hirsch FR, Varella-Garcia M, Franklin WA et al. Evaluation of HER2/ neu gene amplification and protein expression in non-small cell lung carcinomas. Br J Cancer 2002; 86: Ciardiello F, Tortora G. A novel approach in the treatment of cancer: targeting the epidermal growth factor receptor. Clin Cancer Res 2001; 7: Arteaga CL. The epidermal growth factor receptor: from mutant oncogene in nonhuman cancers to therapeutic target in human neoplasia. J Clin Oncol 2001; 19: S32 S Cox G, Jones JL, O Byrne KJ. Matrix metalloproteinase 9 and the epidermal growth factor signal pathway in operable non-small cell lung cancer. Clin Cancer Res 2000; 6: O Byrne KJ, Cox G, Swinson D et al. Towards a biological staging model for operable non-small cell lung cancer. Lung Cancer 2001; 34: S83 S Garcia de Palazzo IE, Adams GP, Sundareshan P et al. Expression of mutated epidermal growth factor receptor by non-small cell lung carcinomas. Cancer Res 1993; 53: Voldborg BR, Damstrup L, Spang-Thomsen M et al. Epidermal growth factor receptor (EGFR) and EGFR mutations, function and possible role in clinical trials. Ann Oncol 1997; 8: Klapper LN, Kirschbaum MH, Sela M, Yarden Y. Biochemical and clinical implications of the erbb/her signaling network of growth factor receptors. Adv Cancer Res 2000; 77: Sirotnak FM, Zakowski MF, Miller VA et al. Efficacy of cytotoxic agents against human tumour xenografts is markedly enhanced by coadministration of ZD1839 (Iressa), an inhibitor of EGFR tyrosine kinase. Clin Cancer Res 2000; 6: Saltz L, Rubin M, Hochster H et al. Cetuximab (IMC-C225) plus irinotecan (CPT-11) is active in CPT-11-refractory colorectal cancer (CRC) that expresses epidermal growth factor receptor (EGFR). Proc Am Soc Clin Oncol 2001; 20: 3a (Abstr 7). 39. Magnè N, Fischel JL, Dubreuil A et al. Influence of epidermal growth factor receptor (EGFR), p53 and intrinsic MAP kinase pathway status of tumour cells on the antiproliferative effect of ZD1839 ( Iressa ). Br J Cancer 2002; 86: