Prognostic Significance of HER-2/neu Overexpression in Stage I Adenocarcinoma of Lung

Prognostic Significance of HER-2/neu Overexpression in Stage I Adenocarcinoma of Lung Chih-Cheng Hsieh, MD, Kuan-Chih Chow, PhD, Huei-Jyh Fahn, MD, Chun-Ming Tsai, MD, Wing-Yin Li, MD, Min-Hsiung Huang, MD, and Liang-Shun Wang, MD Division of Thoracic Surgery, Department of Surgery, Cancer Center, Section of Thoracic Oncology, Chest Department, and Department of Pathology, Veterans General Hospital-Taipei, and National Yang Ming University, Taipei, Taiwan, Republic of China Background. Even with early diagnosis and adequate resection, the 5-year survival rate for stage I lung cancer patients is around 60% to 70%. Overexpression of HER- 2/neu protein is associated with poor prognosis in lung cancers. In this study, we evaluated the expression of HER-2/neu in cancer cells of lung and assessed their clinicopathologic and prognostic significance. Methods. From 1986 to 1995, clinical data on 42 consecutive patients who underwent complete surgical resection for stage I lung adenocarcinoma were collected. Expression of HER-2/neu in paraffin-embedded tumor samples was determined by immunohistochemistry and scored with a semiquantitative method. Results. Twenty-one of 42 patients were positive for HER-2/neu overexpression in tumor. Compared with patients with low HER-2/neu expression, patients with HER-2/neu overexpression had a significantly higher incidence of early tumor recurrence (p 0.014). Survival was also significantly better in patients without HER-2/ neu overexpression than in those with HER-2/neu overexpression (p 0.0047). By univariate analysis, HER-2/ neu overexpression and poor cell differentiation are two important factors correlated with poor prognosis. Conclusions. Expression of HER-2/neu oncoprotein in stage I lung adenocarcinoma can predict the tumor s aggressiveness. Early tumor recurrence was frequently detected in patients with HER-2/neu overexpression. We recommend an individualized therapeutic strategy based on the level of HER-2/neu oncoprotein in the tumor cells. (Ann Thorac Surg 1998;66:1159 64) 1998 by The Society of Thoracic Surgeons Lung cancer is the leading cause of cancer mortality worldwide. In Taiwan, nearly 20% of cancer patients died of this disease and the annual mortality was more than 5,000 patients [1]. Most of the patients who died were at the late stage of the disease when they were diagnosed [2]. Even with early diagnosis and an adequate operation, the 5-year survival rate for stage I patients was only 70% [3]. An intrinsic factor that determines the difference between patients who will have early tumor recurrence and metastasis and those who will not could exist. Advances in molecular biology and genetics have demonstrated that expressions of several markers are associated with lung cancer, such as p53, c-myc, c-fos, Ki-ras, c-erbb-1 [4, 5], and HER-2/neu [6 10]. HER-2/neu protooncogene, which encodes a transmembrane glycoprotein with a molecular weight of 185,000 (p185 neu ), is a receptor with tyrosine kinase activity [11, 12]. Since its first identification [13], HER-2/neu also has been shown to be expressed in a wide variety of human malignancies, including carcinoma of the breast, ovary, gastrointestinal Accepted for publication May 5, 1998. Address reprint requests to Dr Wang, Division of Thoracic Surgery, Department of Surgery, Veterans General Hospital-Taipei, No. 201, Sec 2, Shih-Pai Rd, Taipei, Taiwan, ROC (e-mail: gcliu@www.vghtpe.gov.tw). tract, salivary gland, liver, kidney, and bladder [14]. It has been indicated that expression of HER-2/neu could be associated with multidrug resistance, poor prognosis, and short survival [6 7, 9, 15], and measurement of HER-2/neu should be considered as a marker for predicting early recurrence and metastasis. In this study, we used a semiquantitative method to determine HER-2/neu expression in surgical specimens from patients with stage I lung adenocarcinoma, and to assess the correlation between HER-2/neu expression and the clinicopathologic parameters. The prognostic significance of HER-2/neu overexpression in patients with stage I lung adenocarcinoma also was evaluated. Material and Methods Materials From September 1986 to December 1995, samples from 44 consecutive patients who were diagnosed as having stage I lung adenocarcinoma were collected. Stage of the disease was classified according to the new international staging system for lung cancer [16]. In stage I patients, there should be no evidence of metastasis by radionuclide bone scan, brain scan, liver scan, and sonography of the abdomen. All patients had undergone surgical resection and radical ipsilateral mediastinal lymph node dis- 1998 by The Society of Thoracic Surgeons 0003-4975/98/$19.00 Published by Elsevier Science Inc PII S0003-4975(98)00792-9

1160 HSIEH ET AL Ann Thorac Surg HER-2/neu IN STAGE I LUNG ADENOCARCINOMA 1998;66:1159 64 section. All dissected lymph nodes were proved clear without cancer pathologically. Tumor size, lymph node number, differentiation, vascular invasion, and mitotic number also were evaluated. All patients were routinely followed up every 3 to 6 months in the outpatient department. Tumor recurrence and metastasis were identified when blood examination, biochemical studies, chest radiography, abdominal sonography, whole body bone scan, and computed tomographic scan of the chest showed any suspected evidence of the disease. Immunohistochemistry Three paraffin blocks of tumor were prepared for each case. The tissue preparation and staining procedure have been described previously [6, 13]. Briefly, paraffin blocks were sectioned at 4 m. The wax was melted in a 65 C oven overnight. The slides were deparaffinized in xylene, and xylene was removed subsequently with absolute ethanol before immunostaining. The slides were incubated with monoclonal antibody to HER-2/neu (DAKO A/S, Copenhagen, Denmark) and then with peroxidaseconjugated goat anti-mouse immunoglobulin. Diaminobenzidine was used as chromogenic substrate, and brown precipitate was identified as positive staining. The samples were counterstained with hematoxylin and the slides were mounted with glycerol gelatin. Each batch had a positive and a negative control to ensure the quality. Slide Evaluation Each slide was evaluated randomly at four areas that contained tumor cells. A photograph was taken at each evaluated area for a record. In each case, normal lung tissue served as internal negative control. Photographs were read by two independent observers without clinicopathologic knowledge of the case. The signal of stain was scored on a0to2 subjective scale with 0 no expression, 1 intermediate expression, and 2 strong expression (Fig 1). Four areas were examined on three slides from each patient, resulting in 12 scores. An average was taken as the final reading. The average score of HER-2/neu in the normal lung tissue is 0.21 0.17. After the specificity was determined from the internal negative control, an average score more than 0.4 was defined as positive for the overexpression of HER-2/neu and less than 0.4 as negative. All patients were then divided into two groups following this guideline. Statistical Analysis The relationships between metastasis and expression of HER-2/neu oncoprotein and the other quantitative parameters were studied by analysis of variance. Relationships among qualitative parameters were analyzed by 2 test (or two-tailed Fisher s exact test when the expected number in any cell was smaller than 5 cases). The Kaplan-Meier product-limit estimator was used to estimate cancer-specific survival for subgroups of patients with or without HER-2/neu expression. The log-rank test was used to compare these subgroups with respect to the Fig 1. Immunohistochemical staining of HER-2/neu expression in lung adenocarcinoma: (A) strong overexpression, scored as 2 ; (B) intermediate expression, scored as 1 ; (C) negative for HER-2/neu expression. ( 400 before 35% reduction.) cancer-specific survival. The respective influence of different parameters on survival duration was calculated according to the Cox proportional hazards regression method.

Ann Thorac Surg HSIEH ET AL 1998;66:1159 64 HER-2/neu IN STAGE I LUNG ADENOCARCINOMA 1161 Table 1. Comparison of Parameters Between Patients With and Without Metastasis Parameter Results Metastasis (n 11) No Metastasis (n 31) p Value Age (y) 65.9 7.6 65.4 7.2 0.840 a Smoking index 23.6 20.2 0.716 a (pack-years) Mitotic index 5.6 4.0 0.40 a (number/10hpf) Sex Male 8 21 0.759 b Female 3 10 Tumor location Right 6 19 0.695 b Left 5 12 T status T1 2 12 0.214 b T2 9 19 Operation Pneumonectomy 2 3 0.803 b Bilobectomy 0 1 Lobectomy 9 27 Cell differentiation Well 1 3 0.029 b Moderate 4 23 Poor 4 5 Lymphovascular invasion Positive 8 16 0.224 b Negative 3 15 b Two-sided p value deter- a Two-sided p value determined by t test. mined by the 2 test. HPF high-power field. The mean age of the patients was 65.6 years, ranging from 47 to 83 years. Thirty-one patients were male and 13 patients were female. No significant body weight loss was reported by these patients. Eleven patients (25%) were asymptomatic before lung tumors were identified by routine chest radiography. Thirty patients (68.2%) were smokers and 14 patients smoked more than 20 pack-years. Most tumors were found in the left upper lobe (11 cases, 25%), then the right upper lobe (10 cases, 22.7%), right lower lobe (8 cases, 18.2%), left lower lobe (8 cases, 18.2%), and the right middle lobe (7 cases, 15.9%). Among these patients, 37 chose lobectomy, 1 patient chose bilobectomy and 6 patients chose pneumonectomy. Two deaths were related to the operation. One patient died of respiratory failure and the other died of gastric ulcer bleeding and sepsis. The median follow-up time for the remaining 42 patients was 32.2 months, ranging from 3.6 to 102.4 months. Until the date of data analysis, 30 of 42 patients (71.4%) were still well (3 patients had tumor recurrence and 27 patients were tumor-free). Eight patients died of tumor recurrence and 4 patients died of noncancer causes (car accidents and heart attack). Average duration between operation and tumor recurrence was 14.1 6.6 months (ranging from 1.4 to 25.9 months). The cumulative 3-year and 5-year survival rates are 79% and 71%, respectively. Among these 42 patients, 11 patients had evidence of tumor recurrence. As shown in Table 1, no significant difference was observed between patients with and those without tumor recurrence in the following clinicopathologic parameters: age, sex, tumor location, tumor size (T status), surgical modes, smoking index, and histopathologic characters (mitotic index, evidence of lymphovascular invasion). Nevertheless, a statistical difference was found with respect to the cell differentiation. Patients with poor cell differentiation (n 10) had a significantly higher incidence of tumor recurrence than those with well and moderate cell differentiation (p 0.029). Twenty-one patients were negative for HER-2/neu oncoprotein overexpression, and 21 patients were positive for HER-2/neu overexpression. Nevertheless, no significant difference was found between these two groups of patients in terms of the clinicopathologic parameters. Interestingly, among 21 patients who were HER-2/neu positive, 9 patients had tumor recurrence. Tumor recurrence developed in 7 of them within 13 months after the operation. On the other hand, among 21 patients who were negative for HER-2/neu expression, only 2 patients had metastatic lesions. The metastases were respectively found 18.6 and 19.9 months after the operation. The difference is significant (p 0.014) (Table 2). When we grouped the 11 patients with metastasis, the average score of HER-2/neu was 0.62 0.28. For the remaining 31 patients who were tumor-free, the average score was 0.36 0.29. The difference in HER-2/neu expression between these two groups of patients is significant (p 0.016). In all 11 patients with tumor recurrence, the recurrence developed within 26 months after the operation. The survival of HER-2/neu-positive patients was significantly poorer than that of HER-2/neu-negative patients (Fig 2). Differences in tumor-free survival (p 0.0172) (Fig 2A) and cumulative survival (p 0.0047) (Fig 2B) are highly significant. Table 2. Relationship Between Overexpression of HER-2/ neu and Metastasis a Group Overexpression of HER-2/neu Negative for HER-2/neu Overexpression Total Metastasis 9 (42.9%) 2 (9.5%) 11 (26.2%) No metastasis 12 (57.1%) 19 (90.5%) 31 (73.8%) Total 21 (100%) 21 (100%) 42 (100%) a p 0.014 determined by 2 test.

1162 HSIEH ET AL Ann Thorac Surg HER-2/neu IN STAGE I LUNG ADENOCARCINOMA 1998;66:1159 64 Fig 2. (A) Tumor-free survival curves in patients with stage I lung adenocarcinoma (dotted line negative staining; solid line positive HER-2/neu staining.) (B) Cumulative survival curves in patients with stage I lung adenocarcinoma (dotted line negative staining; solid line positive HER-2/neu staining). Survival curves were plotted with method of Kaplan and Meier. Statistical difference of survivals between different groups was compared by the log rank test. Comment The results presented here demonstrate that overexpression of HER-2/neu in stage I lung adenocarcinoma correlated with the tumor recurrence and metastasis. Previous reports have described the p185 neu expression in lung cancer. In some cases, overexpression of HER-2/neu also was shown to correlate with survival [7 9]. However, a limited number of cases, no disease staging, lack of a scoring system for HER-2/neu expression, and the conflicting results made interpretation of these studies difficult. An elegant study by Harpole and associates [10] examined the relationship between disease progression and immunohistopathology along with oncoprotein expression in stage I non small cell lung cancer. They suggested that male sex, presence of symptoms, tumor size, poor cell differentiation, vascular invasion, HER-2/neu expression, p53 expression, and high Ki-67 index could independently be prognostic factors. They further proposed that the outcome of the disease is the dose response of the additive effect of those parameters. It became clear that disease progression is a complex procedure that could be dependent on several clinicopathologic parameters. The cause of early cancer deaths as well as a mechanism for the early relapse and metastasis remains to be determined. Immunohistochemistry is a frequently used method for evaluating gene expression in tumors. However, a great variation in the interpretation of results has been noted. In the case of lung adenocarcinoma, the positive rate of p185 neu expression ranged from 14% to 87.5%. Variation could be caused by the different methods, materials, or subjective biases. The major cause could be the heterogeneous expression of p185 neu in tumor cells. An objective and quantitative evaluation of immunohistochemical results is required, especially when this immunohistochemically based marker is employed for assessing the prognosis. An insensitive cut-off value used to define high- and low-risk groups of patients can disturb the relationship between variables and outcome. Osaki and colleagues [17] have shown that patients with stage IIIB or T4 lung adenocarcinoma had an increased serum level of HER-2/neu. The concentration of HER-2/neu in serum was proportional to that in tissue section. When Diez and coworkers [18] examined the expression of HER-2/neu in the fresh samples of non small cell lung cancer, there was not a significant correlation between p185 level and TNM classification. However, when patients were categorized by p185 levels, the relapse rate, disease-free period, and median time for tumor relapse worsened proportionally with the rise of p185 neu in the tumor tissue. Nevertheless, the results of enzyme-linked immunosorbent assay may not closely correlate with the clinical courses of disease progression. Both techniques can be used complementarily [19]. In patients with early-stage lung adenocarcinoma, lobectomy or pneumonectomy with ipsilateral mediastinal lymph node dissection is considered the best choice of treatment. Postoperative adjuvant therapy is generally not recommended. It is believed that postoperative chemotherapy does not improve the survival rate but increases the risk of chemotherapy complications. From this study, we found that in stage I lung adenocarcinoma, most tumor recurrences developed within 2 years after surgical intervention and were highly associated with overexpression of HER-2/neu oncoprotein. Diez and colleagues [18] have suggested possible individualized patient management based on p185 level and tailoring adjuvant chemotherapy to high-risk patients. In those cases, we would recommend dividing patients into two groups according to the expression of HER-2/neu in the tumor cells. For patients with low HER-2/neu levels, surgical resection may be adequate for cancer control. On the other hand, in patients with high HER-2/neu expres-

Ann Thorac Surg HSIEH ET AL 1998;66:1159 64 HER-2/neu IN STAGE I LUNG ADENOCARCINOMA 1163 sion, or with high risk for recurrence and death, adjuvant chemotherapy after the operation would be suggested. Unfortunately, tumors with HER-2/neu overexpression respond relatively poorly to the standard chemotherapy regimen [20]. Effective chemotherapeutic agents and protocols require further investigation. HER-2/neu protein is homologous to epidermal growth factor receptor with tyrosine kinase activity. Chemoresistance of cells correlated with the activated state of the enzyme. So tyrosine kinase inhibitor could theoretically reverse chemoresistance and increase chemosensitivity [11, 20]. In our preliminary clinical trial, we found that addition of gemcitabine (2,2-difluorodeoxycytidine) or high-dose caffeine could enhance the chemosensitivity of tumors with HER-2/neu overexpression to the standard chemotherapy regimen [21, 22]. In conclusion, our results demonstrated that cell differentiation and HER-2/neu oncoprotein expression in tumor cells are two important parameters to assess the aggressiveness of stage I lung adenocarcinoma. However, use of a similar scoring system in more patients is required to confirm the value of HER-2/neu overexpression and cell differentiation as prognostic factors. Development of prognostic indices based on these general markers will lead to the initiation of adjuvant therapy for patients with high-risk stage I lung adenocarcinoma. The goal of such a regimen is to improve the treatment outcome for these patients. This study was supported, in part, by Lin Rong-San Foundation of Culture and Social Welfare and, in part, by the Department of Health, Executive Yuan, Republic of China (DOH85-HR-524). We thank Ms Jing-Yi Chao at Cancer Center, Veterans General Hospital-Taipei, for statistical analysis and Ms Li-Ling Yang and Wen-Ser Tseng for their excellent technical assistance. References 1. Annual reports of the Department of Health, Executive Yuan, Republic of China, 1995. 2. Melamed MR, Flehinger BJ, Zaman MB. Impact of early detection on the clinical course of lung cancer. Surg Clin North Am 1987;67:909 24. 3. Pairolero PC, Williams DE, Bergstralh EJ, Piehler JM, Bernatz PE, Payne WS. Postsurgical stage I bronchogenic carcinoma: morbid implications of recurrent disease. Ann Thorac Surg 1984;38:331 6. 4. Volm M, Effereth T, Mattern J. Oncoprotein (c-myc, c- erbb-1, c-erbb-2, c-fos) and suppressor gene product (p53) expression in squamous cell carcinomas of the lung. Clinical and biological correlations. Anticancer Res 1992;12:11 20. 5. Rodenhuis S, van de Wetering ML, Mooi WJ, Evers SG, Nico van Zandwijk, Bos JL. Mutational activity of the K-ras oncogene: a possible pathogenic factor in adenocarcinoma of the lung. N Engl J Med 1987;317:929 35. 6. Schneider PM, Hung MC, Chiocca SM, et al. Differential expression of the c-erbb-2 gene in human small cell and non small cell lung cancer. Cancer Res 1989;49:4968 71. 7. Kern JA, Schwarts DA, Nordberg JE, et al. P185neu expression in human lung adenocarcinoma predicts shortened survival. Cancer Res 1990;50:5184 91. 8. Weiner DB, Nordberg J, Robinson R, et al. Expression of the neu gene-encoded protein (p185neu) in human non small cell carcinomas of the lung. Cancer Res 1990;50:421 31. 9. Shi D, He G, Cao S, et al. Overexpression of the c-erbb-2/ neu-encoded p185 protein in primary lung cancer. Mol Carcinogen 1992;5:213 8. 10. Harpole DH Jr, Herndon JE II, Wolfe WG, Iglehart JD, Marks JR. A prognostic model of recurrent and death in stage I non small cell lung cancer utilizing presentation, histopathology, and oncoprotein expression. Cancer Res 1995;55: 51 6. 11. Coussens L, Yang-Feng TL, Liao YC, et al. Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene. Science 1985;230: 1132 9. 12. Akiyama T, Sudo C, Ogawara H, Toyoshima K, Yamamoto T. The product of human c-erbb-2 gene: a 185-kilodalton glycoprotein with tyrosine kinase activity. Science 1986;232: 1644 6. 13. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breast cancer: correlation of relapse and survival with amplification of the erbb-2/neu oncogene. Science 1987;235:177 82. 14. Hynes NE, Stern DF. The biology of erbb-2/neu/her-2 and its role in cancer. Biochim Biophys Acta 1994;1198:165 84. 15. Tsai CM, Chang KT, Perng RP, et al. Correlation of intrinsic chemoresistance of non small cell lung cancer lines with HER-2/neu gene expression but not with ras gene mutations. J Natl Cancer Inst 1993;85:897 901. 16. Mountain CF. A new international staging system for lung cancer. Chest 1986;89:225S 33S. 17. Osaki T, Mitsudomi T, Oyama T, et al. Serum level and tissue expression of c-erbb-2 protein in lung adenocarcinoma. Chest 1995;108:157 62. 18. Diez M, Pollan M, Maestro M, et al. Prediction of recurrence by quantification of p185 neu protein in non small-cell lung cancer tissue. Br J Cancer 1997;75:684 9. 19. Dittadi R, Donisi PM, Brazzale A, et al. Immunoenzymatic assay of erbb2 protein in cancer and non-malignant breast tissue. Relationship with clinical and biochemical parameters. Anticancer 1992;12:2005 10. 20. Tsai CM, Levitzki A, Wu LH, et al. Enhancement of chemosensitivity by tyrphostin AG825 in high-p185 neu expressing non small cell lung cancer cells. Cancer Res 1996;56:1068 74. 21. Tsai CM, Perng RP, Chen MM, et al. Greater enhancement of chemosensitivity by caffeine in high p185neu expression human non small cell lung cancer cell lines. J Natl Cancer Inst 1994;86:1018 20. 22. Tsai CM, Chang KT, Chen JY, et al. Cytotoxic effects of gemcitabine-containing regimens against human non small cell lung cancer cell lines which express different levels of p185neu. Cancer Res 1996;56:794 801. INVITED COMMENTARY This article by Dr Hsieh and associates details the immunohistochemical evidence of overexpression of the HER-2/neu protein in 42 patients with stage I adenocarcinoma lung cancer. They describe a novel technique for quantifying the amount of HER-2/neu expression seen on microscope slides. Their scoring 1998 by The Society of Thoracic Surgeons 0003-4975/98/$19.00 Published by Elsevier Science Inc PII S0003-4975(98)00793-0