- Email: [email protected]
neu Expression on Survival in Non–Small-Cell Lung Cancer
o riginal contribution Effect of HER2 /neu Expression on Survival in Non–Small-Cell Lung Cancer Vijaya Korrapati,1 Maryanne Gaffney,2 Lisbeth G. Larsson,2 Lucio Di Nunno,1 Mark Riggs,3 Robert S. Beissner,2 John J. Rinehart,1 Frank E. Mott1 Abstract Major prognostic factors for early-stage non–small-cell lung cancer (NSCLC) are tumor size and nodal status. It has been suggested that HER2/neu overexpression may be related to poor prognosis in NSCLC. We evaluated the significance of HER2/neu overexpression on survival in patients with NSCLC. Data were collected on 239 patients treated surgically for stage I/II NSCLC between 1987 and 1996. None of the patients received adjuvant chemotherapy or radiation. Formalin-fixed, paraffin-embedded tumor tissue samples were stained with p185/HER2 receptor antibody. Results were reported as positive (2+, 3+) or negative (0, 1+) (Group A). A separate analysis considered only 3+ as positive (Group B). HER2/neu overexpression was seen in 18% in Group A (43 of 239) and 6% in Group B (15 of 239). HER2/neu overexpression was highest in bronchoalveolar cell carcinoma and adenocarcinoma. More stage I tumors were positive than stage II in both groups, but this was significant only in Group A (21% vs. 7%, P = 0.02). No difference was seen with age, gender, or grade for either group. In Group A, the relapse rate was 55% for HER2/neu-overexpressing tumors and 31% for HER2/neu-negative tumors (P = 0.003). Median time to relapse in patients with HER2/neu-positive tumors was 2.9 years; it was not reached in patients with HER2/neu-negative tumors. Median survival of patients with HER2/neu-positive tumors was 3.6 years compared to 5 years in patients with HER2/neunegative tumors (P = 0.66). In Group B, the relapse rate was 60% for HER2/neu-overexpressing tumors and 33% for negative tumors (P = 0.036). Median time to relapse was 3.4 years in HER2/neu positive and had not been reached in negative tumors. There was no difference in 5-year survival rates for both groups (47% for HER2/neu positive and 50% for negative, P = 0.66). Clinical Lung Cancer, Vol. 2, No. 3, 216-219, 2001
Key words: HER2/neu expression, Prognosis, Bronchoalveolar cell carcinoma, Adenocarcinoma
Introduction
stage I/II non–small-cell lung cancer (NSCLC) to evaluate the incidence and prognostic significance of HER2/neu overexpression in all histological subtypes.
Lung cancer is the leading cause of cancer-related death in the United States, with 156,900 deaths predicted for the year 2000, accounting for 31% of all cancer-related deaths in men and 25% in women.1 The long-term survival of patients with lung cancer still remains poor and is approximately 13% at 5 years. To improve survival, earlier diagnosis and identification of prognostic factors are important. In human breast cancer, a relationship between HER2/neu overexpression, nodal status, histologic grade, relapse-free survival, and overall survival has been demonstrated.2 In this study, we retrospectively analyzed 239 patients with surgically treated 1 Division
of Hematology and Oncology of Pathology 3 Department of Biostatistics Scott and White Clinic, Temple, TX 2 Department
Submitted: June 20, 2000; Revised: Nov. 27, 2000; Accepted: Dec. 15, 2000 Address for correspondence: Frank Mott, MD, Division of Hematology/ Oncology, Scott and White Clinic, 2401 So 31st Street, Temple, TX 76508 Fax: 254-724-4904; e-mail: [email protected]
216
February 2001
Patient Characteristics We collected data on 239 patients treated with surgical resection at the Scott and White Clinic for pathological stage I/II NSCLC between 1987 and 1996. Staging was according to the tumor-node-metastasis classification set by the International Staging System.3 None of the patients received adjuvant chemotherapy or radiation therapy. Tumors were classified according to their morphological and immunohistochemical features as large-cell carcinoma, squamous cell carcinoma, adenocarcinoma, or bronchoalveolar cell carcinoma. Charts were reviewed to obtain data such as age, gender, time to relapse, and survival.
HER2/neu Staining of Tumor Tissue Formalin-fixed and paraffin-embedded tumor tissue samples were retrieved from the archives of our pathology department. A representative block was chosen for each case and submitted for immunohistochemistry of HER2/neu. The tissue was stained
Figure 1 HER2/neu Staining A
B
1+
0 D
C
3+
2+
Immunohistochemistry of HER2/neu overexpression in non–small-cell carcinoma of the lung. (A) shows no membrane staining (Zero); (B) shows a faint staining of a portion of the cytoplasmic membrane in at least 10% of neoplastic cells (1+); (C) shows a weak, but definite staining of the entire cytoplasmic membrane in at least 10% of the neoplastic cells (2+); and (D) shows strong staining of the entire cytoplasmic membrane in at least 10% of the neoplastic cells (3+).
with antibodies used in our department to assess breast carcinoma HER2/neu receptor status. The slides were evaluated by two pathologists for positive or negative antibody staining. A representative four-micron section from each NSCLC tumor block was mounted on a glass slide and prepared for immunostaining by standard procedures by deparaffinizing (xylene, ethanol), rehydrating (deionized water), and boiling with a citrate buffer for antigen retrieval. The slides were incubated with primary anti-p185/HER2 rabbit antibody (Herceptest, DAKO A/S, Copenhagen, Denmark) and then with secondary peroxidase conjugated goat antirabbit antibody. Staining was developed with a chromogenic substrate (diaminobenzidine), and then the specimen was counter-stained with hematoxylin and eosin. A positive and negative control was run for each test to ensure quality. All procedures were as per DAKO guidelines. Each slide was evaluated by two independent observers and scored on a 0 to 3+ scale. No staining was 0, 1+ indicated faint partial membrane staining in more than 10% of the tumor cells, 2+ indicated weak to moderate complete membrane staining in more than 10% of the tumor cells, and 3+ indicated moderate to strong complete membrane staining in more than 10% of the tumor cells (Figure 1). Any discrepancies in scoring were reviewed by both observers, and a consensus score was determined. The scores were interpreted as per DAKO’s guideline
with a score of 0-1+ being negative and a score of 2+-3+ being positive in Group A and a score of 0-2+ being negative and 3+ only as positive in Group B. HER2/neu overexpression was compared with various patient and tumor characteristics and relapse rates using the chi-square test. Time to relapse and survival times were analyzed using Kaplan-Meier survival curves. The log-rank test was used to assess the effect of HER2/neu overexpression. P values less than 0.05 indicated statistically significant events.
Results Staining Characteristics Overall incidence of 2+-3+ HER2/neu overexpression was 18% (43 of 239, Group A); 3+ HER2/neu overexpression was seen in 6% (15 of 239, Group B). No statistical differences were seen with gender or age. There was a statistically significant difference within histological subgroups. 2+-3+ HER2/neu overexpression was seen in 33% of bronchoalveolar cell carcinomas and 30% of adenocarcinomas compared with 4% in squamous cell carcinomas and 8% in large-cell carcinomas (P = 0.001). This difference was also seen in Group B. Bronchoalveolar cell carcinomas had 3 + HER2/neu overexpression in 15% and adenocarcinomas in 10% of the patients, compared with 1% in squamous cell carcinomas and none in large-cell carcinomas (P
February 2001
217
HER2/neu Expression in Lung Cancer Table 1
HER2/neu Overexpression and Tumor Characteristics Group A (2+-3+ HER2/neu)
Positive Cases
43/239 (18%)
Group B HER2/neu)
(3+
15/239 (6%)
Gender
HER2/neu Overexpression and Relapse Rate
HER2/neu Stain
Number Relapsed
% Relapsed
0, 1+ 2+, 3+
59/192
31%
23/42
55%
73/219
33%
9/15
60%
82/234
35%
Male
26/157 (17%)
9/157 (6%)
Female
17/82 (21%)
6/82 (7%)
0, 1+, 2+ 3+
P = 0.43
P = 0.63
Total
< 60 years
5/42 (12%)
1/42 (2%)
60-69 years
25/110 (23%)
9/110 (8%)
≥ 70 years
13/86 (15%)
5/86 (6%)
P = 0.20
P = 0.41
Large cell
2/24 (8%)
0/24 (0%)
Squamous cell
4/95 (4%)
1/95 (1%)
Adenocarcinoma
24/81 (30%)
8/81 (10%)
Bronchoalveolar
13/39 (33%)
6/39 (15%)
P = 0.001
P = 0.004
7/24 (29%)
4/24 (17%)
Age (one missing)
Grade Well differentiated Moderately differentiated
17/87 (20%)
6/87 (7%)
Poorly differentiated
19/128 (15%)
5/128 (4%)
P = 0.22
P = 0.06
I
39/183 (21%)
12/183 (7%)
II
4/56 (7%)
3/56 (5%)
P = 0.02
P = 0.75
Stage
= 0.004). HER2/neu overexpression was more common in welldifferentiated tumors than in poorly differentiated tumors, but the results were not statistically significant (Group A: 29% vs. 15%, P = 0.22; Group B: 17% vs. 4%, P = 0.06). More patients with stage I tumors were positive for HER2/neu overexpression than were patients with stage II tumors, but only Group A was statistically significant (Group A: 21% stage I vs. 7% stage II, P = 0.02; Group B: 7% stage I vs. 5% stage II, P = 0.75). See Table 1.
Correlation Between HER2/neu Expression and Relapse Rate A correlation between HER2/neu overexpression and relapse rate was seen and reached statistical significance in both groups. In Group A, 55% (23 of 42) of patients with HER2/neu overexpressing tumors relapsed vs. 31% (59 of 192) of patients without HER2/neu overexpression (P = 0.003). In Group B, 60% (9 of 15) of patients with HER2/neu-overexpressing tumors relapsed vs. 33% (73 of 219) of patients without overexpression (P = 0.036). See Table 2. Median time to relapse for patients with overexpressed tumors was 2.9 years in Group A and 3.4 years in Group B (Figure 2), whereas median time to relapse was not reached for the patients with HER2/neu-negative tumors in
February 2001
P Value 0.003
0.036
Data missing in 5 patients
Histology
218
Table 2
either group. These differences in both groups reached statistical significance (Group A: P < 0.01; Group B: P = 0.02). At the end of 6.3 years of median potential follow-up, only a third of patients in the HER2/neu-negative group had relapsed. The actual sites of relapse were not addressed in this study.
Correlation Between HER2/neu Expression and Survival At a median follow-up of 3.4 years, 132 of the 239 patients had died, 73 with relapsed disease and 59 without any evidence of disease. In Group A, the median survival for patients with HER2/neu overexpression was 3.6 years compared to 5 years for patients without overexpression. Five-year survival was 47% for HER2/neu overexpression and 50% for the HER2/neu-negative group. This survival difference did not reach statistical significance (P = 0.66). When just the 3+ HER2/neu overexpression was considered (Group B), the median and 5-year survivals were similar (3 years vs. 4.8 years and 47% vs. 49%, respectively, for overexpression vs. negative; P = 0.61). See Figure 3.
Discussion Prognosis in NSCLC relates to clinical factors such as tumor size and nodal status.3 Other factors may include p53 and ras mutations, angiogenesis factors, and HER2/neu overexpression. HER2/neu overexpression has been demonstrated in breast, ovarian, and colorectal carcinomas and in nonmalignant diseases such as cirrhosis of the liver.4 Its association with nodal status, tumor grade, and survival has been shown in breast cancer. HER2/neu overexpression has been described in NSCLC5 as a poor prognostic factor,6 correlating inversely with survival7 and directly with drug resistance.8 Harpole et al studied patients with stage I NSCLC and reported a 5-year survival of 72% for patients without p53 mutation or HER2/neu overexpression, 58% for patients with either of these two factors, and 38% if both existed.9 In 42 patients with stage I adenocarcinoma of the lung, Hsieh et al showed HER2/neu overexpression in 50% and that overexpression was predictive of early recurrence and decreased survival.10 In a retrospective study, HER2/neu overexpression was predictive of poor survival in 103 patients with either adenocarcinoma or large-cell carcinoma of the lung.11 In one study of 54 patients with stage III NSCLC treated with chemoradiation, HER2/neu correlated with distant metastases and poorer survival,12 while another study of 38 patients with stage IIIA (N2) NSCLC found no correlation.13 Kim et al showed that HER2/neu overexpression in combination with coexpression of Bcl-2, p53, or
Vijaya Korrapati et al Figure 2 HER2/neu Expression vs. Time to Relapse
Figure 3 HER2/neu Expression vs. Survival 100% Group A
HER2/neu level 0, 1+ 2+, 3+ P < 0.01
80% 60%
Percent Surviving
Percent with Relapse
100%
40% 20%
0
2
4
6
8
10
12
80% 60% 40% 20%
14
0
2
Time Since Diagnosis (Years)
4
6
8
10
12
14
Time Since Diagnosis (Years)
100%
100% Group B
HER2/neu level 0, 1+, 2+ 3+ P = 0.02
80% 60%
Percent Surviving
Percent with Relapse
Group A
HER2/neu level 0, 1+ 2+, 3+ P = 0.66
40% 20%
0
2
4
6
8
10
12
14
60% 40% 20%
0
Time Since Diagnosis (Years)
p21 was negatively associated with survival in NSCLC.14 In this study, the incidence of HER2/neu overexpression was 18% (6% if only the 3+ score was considered as positive) and was more common in adenocarcinoma and bronchoalveolar cell carcinoma (greater than 60% combined). We initially designated overexpression as 2+ or 3+ (Group A), consistent with the prevailing standards utilized in breast cancer. However, 3+ has come to be the accepted threshold for “positive” HER2/neu breast cancers; therefore, we made a decision to separately analyze the 3+ overexpressors (Group B). HER2/neu overexpression was significantly associated with rate of relapse and median time to relapse. The median time to relapse exceeded the median survival in Group B (3.4 years vs. 3 years) due to an excess of non–cancer-related deaths in a small sample size (15 patients). A statistically significant survival difference was not seen, which this may be due to the smaller number of HER2/neu overexpressed tumors, especially if only 3+ was considered positive (Group B). Median follow-up (3.4 years) may be too short to detect a difference. Since the majority of cases accessioned came from the later years of the study, this may account for the shorter than expected median follow-up. Overall survival at 8 years was 38% for both groups.
Conclusion Further investigations are needed to determine if more aggressive therapies might benefit patients with tumors that overexpress HER2/neu. Trastuzumab treatment of HER2/neuoverexpressing lung tumors is under clinical investigation. Based on experience in breast cancer, criteria for lung cancer need to
Group B
HER2/neu level 0, 1+, 2+ 3+ P = 0.61
80%
2
4
6
8
10
12
14
Time Since Diagnosis (Years)
be defined regarding the level of HER2/neu overexpression (2+ or 3+) appropriate for trastuzumab use. If 3+ overexpression is required, the actual number of lung cancers eligible for treatment may be relatively small.
References 01. Greenlee RT, Murray T, Bolden S, et al. Cancer statistics, 2000. CA Cancer J Clin 2000; 50:7-33. 02. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 1989; 244:707-712. 03. Mountain CF. A new international staging system for lung cancer. Chest 1986; 89 (suppl):225S-233S. 04. Molina R, Jo J, Filella X, et al. Serum levels of C-erbB-2 (HER-2/neu) in patients with malignant and non-malignant diseases. Tumour Biol 1997; 18:188-196. 05. 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: 421-425. 06. Shi D, He G, Cao S, et al. Overexpression of the c-erbB-2/neu-encoded p185 protein in primary lung cancer. Mol Carcinog 1992; 5:213-218. 07. Kern JA, Schwartz DA, Nordberg JE, et al. p185neu expression in human lung adenocarcinomas predicts shortened survival. Cancer Res 1990; 50:5184-5187. 08. Tsai CM, Chang KT, Perng RP, et al. Correlation of intrinsic chemoresistance of nonsmall-cell lung cancer cell lines with HER-2/neu gene expression but not with ras gene mutations. J Natl Cancer Inst 1993; 85:897-901. 09. Harpole DH Jr., Herndon JE 2nd, Wolfe WG, et al. A prognostic model of recurrence and death in stage I non-small cell lung cancer utilizing presentation, histopathology, and oncoprotein expression. Cancer Res 1995; 55:51-56. 10. Hsieh CC, Chow KC, Fahn HJ, et al. Prognostic significance of HER-2/neu overexpression in stage I adenocarcinoma of lung. Ann Thorac Surg 1998; 66:1159-1164. 11. Nemunaitis J, Klemow S, Tong A, et al. Prognostic value of K-ras mutations, ras oncoprotein, and c-erb B-2 oncoprotein expression in adenocarcinoma of the lung. Am J Clin Oncol 1998; 21:155-160. 12. Thomas M, Rube C, Semik M, et al. Trimodality therapy in stage III non-small cell lung cancer: prediction of recurrence by assessment of p185neu. Eur Respir J 1999; 13: 424-429. 13. Graziano SL, Kern JA, Herndon JE, et al. Analysis of neuroendocrine markers, HER2 and CEA before and after chemotherapy in patients with stage IIIA non-small cell lung cancer: a Cancer and Leukemia Group B study. Lung Cancer 1998; 21:203-211. 14. Kim YC, Park KO, Kern JA, et al. The interactive effect of Ras, HER2, P53 and Bcl-2 expression in predicting the survival of non-small cell lung cancer patients. Lung Cancer 1998; 22:181-190.
February 2001
219
Key words: HER2/neu expression, Prognosis, Bronchoalveolar cell carcinoma, Adenocarcinoma
Introduction
stage I/II non–small-cell lung cancer (NSCLC) to evaluate the incidence and prognostic significance of HER2/neu overexpression in all histological subtypes.
Lung cancer is the leading cause of cancer-related death in the United States, with 156,900 deaths predicted for the year 2000, accounting for 31% of all cancer-related deaths in men and 25% in women.1 The long-term survival of patients with lung cancer still remains poor and is approximately 13% at 5 years. To improve survival, earlier diagnosis and identification of prognostic factors are important. In human breast cancer, a relationship between HER2/neu overexpression, nodal status, histologic grade, relapse-free survival, and overall survival has been demonstrated.2 In this study, we retrospectively analyzed 239 patients with surgically treated 1 Division
of Hematology and Oncology of Pathology 3 Department of Biostatistics Scott and White Clinic, Temple, TX 2 Department
Submitted: June 20, 2000; Revised: Nov. 27, 2000; Accepted: Dec. 15, 2000 Address for correspondence: Frank Mott, MD, Division of Hematology/ Oncology, Scott and White Clinic, 2401 So 31st Street, Temple, TX 76508 Fax: 254-724-4904; e-mail: [email protected]
216
February 2001
Patient Characteristics We collected data on 239 patients treated with surgical resection at the Scott and White Clinic for pathological stage I/II NSCLC between 1987 and 1996. Staging was according to the tumor-node-metastasis classification set by the International Staging System.3 None of the patients received adjuvant chemotherapy or radiation therapy. Tumors were classified according to their morphological and immunohistochemical features as large-cell carcinoma, squamous cell carcinoma, adenocarcinoma, or bronchoalveolar cell carcinoma. Charts were reviewed to obtain data such as age, gender, time to relapse, and survival.
HER2/neu Staining of Tumor Tissue Formalin-fixed and paraffin-embedded tumor tissue samples were retrieved from the archives of our pathology department. A representative block was chosen for each case and submitted for immunohistochemistry of HER2/neu. The tissue was stained
Figure 1 HER2/neu Staining A
B
1+
0 D
C
3+
2+
Immunohistochemistry of HER2/neu overexpression in non–small-cell carcinoma of the lung. (A) shows no membrane staining (Zero); (B) shows a faint staining of a portion of the cytoplasmic membrane in at least 10% of neoplastic cells (1+); (C) shows a weak, but definite staining of the entire cytoplasmic membrane in at least 10% of the neoplastic cells (2+); and (D) shows strong staining of the entire cytoplasmic membrane in at least 10% of the neoplastic cells (3+).
with antibodies used in our department to assess breast carcinoma HER2/neu receptor status. The slides were evaluated by two pathologists for positive or negative antibody staining. A representative four-micron section from each NSCLC tumor block was mounted on a glass slide and prepared for immunostaining by standard procedures by deparaffinizing (xylene, ethanol), rehydrating (deionized water), and boiling with a citrate buffer for antigen retrieval. The slides were incubated with primary anti-p185/HER2 rabbit antibody (Herceptest, DAKO A/S, Copenhagen, Denmark) and then with secondary peroxidase conjugated goat antirabbit antibody. Staining was developed with a chromogenic substrate (diaminobenzidine), and then the specimen was counter-stained with hematoxylin and eosin. A positive and negative control was run for each test to ensure quality. All procedures were as per DAKO guidelines. Each slide was evaluated by two independent observers and scored on a 0 to 3+ scale. No staining was 0, 1+ indicated faint partial membrane staining in more than 10% of the tumor cells, 2+ indicated weak to moderate complete membrane staining in more than 10% of the tumor cells, and 3+ indicated moderate to strong complete membrane staining in more than 10% of the tumor cells (Figure 1). Any discrepancies in scoring were reviewed by both observers, and a consensus score was determined. The scores were interpreted as per DAKO’s guideline
with a score of 0-1+ being negative and a score of 2+-3+ being positive in Group A and a score of 0-2+ being negative and 3+ only as positive in Group B. HER2/neu overexpression was compared with various patient and tumor characteristics and relapse rates using the chi-square test. Time to relapse and survival times were analyzed using Kaplan-Meier survival curves. The log-rank test was used to assess the effect of HER2/neu overexpression. P values less than 0.05 indicated statistically significant events.
Results Staining Characteristics Overall incidence of 2+-3+ HER2/neu overexpression was 18% (43 of 239, Group A); 3+ HER2/neu overexpression was seen in 6% (15 of 239, Group B). No statistical differences were seen with gender or age. There was a statistically significant difference within histological subgroups. 2+-3+ HER2/neu overexpression was seen in 33% of bronchoalveolar cell carcinomas and 30% of adenocarcinomas compared with 4% in squamous cell carcinomas and 8% in large-cell carcinomas (P = 0.001). This difference was also seen in Group B. Bronchoalveolar cell carcinomas had 3 + HER2/neu overexpression in 15% and adenocarcinomas in 10% of the patients, compared with 1% in squamous cell carcinomas and none in large-cell carcinomas (P
February 2001
217
HER2/neu Expression in Lung Cancer Table 1
HER2/neu Overexpression and Tumor Characteristics Group A (2+-3+ HER2/neu)
Positive Cases
43/239 (18%)
Group B HER2/neu)
(3+
15/239 (6%)
Gender
HER2/neu Overexpression and Relapse Rate
HER2/neu Stain
Number Relapsed
% Relapsed
0, 1+ 2+, 3+
59/192
31%
23/42
55%
73/219
33%
9/15
60%
82/234
35%
Male
26/157 (17%)
9/157 (6%)
Female
17/82 (21%)
6/82 (7%)
0, 1+, 2+ 3+
P = 0.43
P = 0.63
Total
< 60 years
5/42 (12%)
1/42 (2%)
60-69 years
25/110 (23%)
9/110 (8%)
≥ 70 years
13/86 (15%)
5/86 (6%)
P = 0.20
P = 0.41
Large cell
2/24 (8%)
0/24 (0%)
Squamous cell
4/95 (4%)
1/95 (1%)
Adenocarcinoma
24/81 (30%)
8/81 (10%)
Bronchoalveolar
13/39 (33%)
6/39 (15%)
P = 0.001
P = 0.004
7/24 (29%)
4/24 (17%)
Age (one missing)
Grade Well differentiated Moderately differentiated
17/87 (20%)
6/87 (7%)
Poorly differentiated
19/128 (15%)
5/128 (4%)
P = 0.22
P = 0.06
I
39/183 (21%)
12/183 (7%)
II
4/56 (7%)
3/56 (5%)
P = 0.02
P = 0.75
Stage
= 0.004). HER2/neu overexpression was more common in welldifferentiated tumors than in poorly differentiated tumors, but the results were not statistically significant (Group A: 29% vs. 15%, P = 0.22; Group B: 17% vs. 4%, P = 0.06). More patients with stage I tumors were positive for HER2/neu overexpression than were patients with stage II tumors, but only Group A was statistically significant (Group A: 21% stage I vs. 7% stage II, P = 0.02; Group B: 7% stage I vs. 5% stage II, P = 0.75). See Table 1.
Correlation Between HER2/neu Expression and Relapse Rate A correlation between HER2/neu overexpression and relapse rate was seen and reached statistical significance in both groups. In Group A, 55% (23 of 42) of patients with HER2/neu overexpressing tumors relapsed vs. 31% (59 of 192) of patients without HER2/neu overexpression (P = 0.003). In Group B, 60% (9 of 15) of patients with HER2/neu-overexpressing tumors relapsed vs. 33% (73 of 219) of patients without overexpression (P = 0.036). See Table 2. Median time to relapse for patients with overexpressed tumors was 2.9 years in Group A and 3.4 years in Group B (Figure 2), whereas median time to relapse was not reached for the patients with HER2/neu-negative tumors in
February 2001
P Value 0.003
0.036
Data missing in 5 patients
Histology
218
Table 2
either group. These differences in both groups reached statistical significance (Group A: P < 0.01; Group B: P = 0.02). At the end of 6.3 years of median potential follow-up, only a third of patients in the HER2/neu-negative group had relapsed. The actual sites of relapse were not addressed in this study.
Correlation Between HER2/neu Expression and Survival At a median follow-up of 3.4 years, 132 of the 239 patients had died, 73 with relapsed disease and 59 without any evidence of disease. In Group A, the median survival for patients with HER2/neu overexpression was 3.6 years compared to 5 years for patients without overexpression. Five-year survival was 47% for HER2/neu overexpression and 50% for the HER2/neu-negative group. This survival difference did not reach statistical significance (P = 0.66). When just the 3+ HER2/neu overexpression was considered (Group B), the median and 5-year survivals were similar (3 years vs. 4.8 years and 47% vs. 49%, respectively, for overexpression vs. negative; P = 0.61). See Figure 3.
Discussion Prognosis in NSCLC relates to clinical factors such as tumor size and nodal status.3 Other factors may include p53 and ras mutations, angiogenesis factors, and HER2/neu overexpression. HER2/neu overexpression has been demonstrated in breast, ovarian, and colorectal carcinomas and in nonmalignant diseases such as cirrhosis of the liver.4 Its association with nodal status, tumor grade, and survival has been shown in breast cancer. HER2/neu overexpression has been described in NSCLC5 as a poor prognostic factor,6 correlating inversely with survival7 and directly with drug resistance.8 Harpole et al studied patients with stage I NSCLC and reported a 5-year survival of 72% for patients without p53 mutation or HER2/neu overexpression, 58% for patients with either of these two factors, and 38% if both existed.9 In 42 patients with stage I adenocarcinoma of the lung, Hsieh et al showed HER2/neu overexpression in 50% and that overexpression was predictive of early recurrence and decreased survival.10 In a retrospective study, HER2/neu overexpression was predictive of poor survival in 103 patients with either adenocarcinoma or large-cell carcinoma of the lung.11 In one study of 54 patients with stage III NSCLC treated with chemoradiation, HER2/neu correlated with distant metastases and poorer survival,12 while another study of 38 patients with stage IIIA (N2) NSCLC found no correlation.13 Kim et al showed that HER2/neu overexpression in combination with coexpression of Bcl-2, p53, or
Vijaya Korrapati et al Figure 2 HER2/neu Expression vs. Time to Relapse
Figure 3 HER2/neu Expression vs. Survival 100% Group A
HER2/neu level 0, 1+ 2+, 3+ P < 0.01
80% 60%
Percent Surviving
Percent with Relapse
100%
40% 20%
0
2
4
6
8
10
12
80% 60% 40% 20%
14
0
2
Time Since Diagnosis (Years)
4
6
8
10
12
14
Time Since Diagnosis (Years)
100%
100% Group B
HER2/neu level 0, 1+, 2+ 3+ P = 0.02
80% 60%
Percent Surviving
Percent with Relapse
Group A
HER2/neu level 0, 1+ 2+, 3+ P = 0.66
40% 20%
0
2
4
6
8
10
12
14
60% 40% 20%
0
Time Since Diagnosis (Years)
p21 was negatively associated with survival in NSCLC.14 In this study, the incidence of HER2/neu overexpression was 18% (6% if only the 3+ score was considered as positive) and was more common in adenocarcinoma and bronchoalveolar cell carcinoma (greater than 60% combined). We initially designated overexpression as 2+ or 3+ (Group A), consistent with the prevailing standards utilized in breast cancer. However, 3+ has come to be the accepted threshold for “positive” HER2/neu breast cancers; therefore, we made a decision to separately analyze the 3+ overexpressors (Group B). HER2/neu overexpression was significantly associated with rate of relapse and median time to relapse. The median time to relapse exceeded the median survival in Group B (3.4 years vs. 3 years) due to an excess of non–cancer-related deaths in a small sample size (15 patients). A statistically significant survival difference was not seen, which this may be due to the smaller number of HER2/neu overexpressed tumors, especially if only 3+ was considered positive (Group B). Median follow-up (3.4 years) may be too short to detect a difference. Since the majority of cases accessioned came from the later years of the study, this may account for the shorter than expected median follow-up. Overall survival at 8 years was 38% for both groups.
Conclusion Further investigations are needed to determine if more aggressive therapies might benefit patients with tumors that overexpress HER2/neu. Trastuzumab treatment of HER2/neuoverexpressing lung tumors is under clinical investigation. Based on experience in breast cancer, criteria for lung cancer need to
Group B
HER2/neu level 0, 1+, 2+ 3+ P = 0.61
80%
2
4
6
8
10
12
14
Time Since Diagnosis (Years)
be defined regarding the level of HER2/neu overexpression (2+ or 3+) appropriate for trastuzumab use. If 3+ overexpression is required, the actual number of lung cancers eligible for treatment may be relatively small.
References 01. Greenlee RT, Murray T, Bolden S, et al. Cancer statistics, 2000. CA Cancer J Clin 2000; 50:7-33. 02. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 1989; 244:707-712. 03. Mountain CF. A new international staging system for lung cancer. Chest 1986; 89 (suppl):225S-233S. 04. Molina R, Jo J, Filella X, et al. Serum levels of C-erbB-2 (HER-2/neu) in patients with malignant and non-malignant diseases. Tumour Biol 1997; 18:188-196. 05. 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: 421-425. 06. Shi D, He G, Cao S, et al. Overexpression of the c-erbB-2/neu-encoded p185 protein in primary lung cancer. Mol Carcinog 1992; 5:213-218. 07. Kern JA, Schwartz DA, Nordberg JE, et al. p185neu expression in human lung adenocarcinomas predicts shortened survival. Cancer Res 1990; 50:5184-5187. 08. Tsai CM, Chang KT, Perng RP, et al. Correlation of intrinsic chemoresistance of nonsmall-cell lung cancer cell lines with HER-2/neu gene expression but not with ras gene mutations. J Natl Cancer Inst 1993; 85:897-901. 09. Harpole DH Jr., Herndon JE 2nd, Wolfe WG, et al. A prognostic model of recurrence and death in stage I non-small cell lung cancer utilizing presentation, histopathology, and oncoprotein expression. Cancer Res 1995; 55:51-56. 10. Hsieh CC, Chow KC, Fahn HJ, et al. Prognostic significance of HER-2/neu overexpression in stage I adenocarcinoma of lung. Ann Thorac Surg 1998; 66:1159-1164. 11. Nemunaitis J, Klemow S, Tong A, et al. Prognostic value of K-ras mutations, ras oncoprotein, and c-erb B-2 oncoprotein expression in adenocarcinoma of the lung. Am J Clin Oncol 1998; 21:155-160. 12. Thomas M, Rube C, Semik M, et al. Trimodality therapy in stage III non-small cell lung cancer: prediction of recurrence by assessment of p185neu. Eur Respir J 1999; 13: 424-429. 13. Graziano SL, Kern JA, Herndon JE, et al. Analysis of neuroendocrine markers, HER2 and CEA before and after chemotherapy in patients with stage IIIA non-small cell lung cancer: a Cancer and Leukemia Group B study. Lung Cancer 1998; 21:203-211. 14. Kim YC, Park KO, Kern JA, et al. The interactive effect of Ras, HER2, P53 and Bcl-2 expression in predicting the survival of non-small cell lung cancer patients. Lung Cancer 1998; 22:181-190.
February 2001
219