Abstract

Background

Biological complexity leads to significant variation in the survival of patients with stage I non-small-cell lung cancer (NSCLC). DNA damage response (DDR) pathways play a critical role in maintaining genomic stability and in the progression of NSCLC. Therefore, the development of a prognostic biomarker focusing on DDR pathways is an intriguing issue.

Patients and methods

Expression of several proteins (ATM, ATMpS1981, γH2AX, 53BP1, 53BP1pS25, Chk2, Chk2pT68, MDC1, MDC1pS964, BRCA1pS1423, and ERCC1) and overall survival were investigated in 889 pathological stage I NSCLC patients.

Results

Low expression of BRCA1pS1423 or ERCC1 was significantly associated with worse survival in the whole cohort of patients. Analysis performed based on histology revealed that low expression of γH2AX, Chk2pT68, or ERCC1 was a poor prognostic factor in squamous cell carcinoma patients [adjusted hazard ratio (aHR), Cox P: 1.544, 0.012 for γH2AX; 1.624, 0.010 for Chk2pT68; 1.569, 0.011 for ERCC1]. The analysis of the interaction between two proteins showed that this effect was more pronounced in squamous cell carcinoma patients. However, these effects were not detected in adenocarcinoma patients.

Conclusions

The proteins involved in DDR pathways exhibited differential expression between squamous cell carcinoma and adenocarcinoma and were important determinants of survival in stage I squamous cell carcinoma patients.

introduction

Lung cancer is the most common cause of cancer-related death in many countries, but the 5-year survival of patients with non-small-cell lung cancer (NSCLC) has been slowly improving; it stands currently at 16% [1, 2]. Lung cancer screening for preclinical disease, the development of biomarkers that predict treatment efficiency, or new drugs for advanced stage have been major issues for improving the dismal survival of this disease [3–5]. The development of a prognostic biomarker for stage I NSCLC is another focus of research in this field, as the 5-year survival of these patients remains disappointing at 58%–73% [6] and two recent meta-analyses indicated that adjuvant chemotherapy did not yield a survival advantage and was even harmful to the patients [7, 8].

The DNA damage response (DDR) is a complex network of signaling pathways. Proteins involved in DDR pathways play crucial roles in maintaining genomic stability, thus suppressing tumor development or the progression of cancer as well as regulating cell cycle checkpoints [9–11]. Ataxia telangiectasia mutated (ATM), γH2AX, p53 binding protein 1 (53BP1), checkpoint kinase 2 (Chk2), mediator of DNA damage checkpoint protein 1 (MDC1), breast cancer susceptibility gene 1 (BRCA1), or excision repair cross-complementation 1 (ERCC1) are activated as a network and represent important assets in the repair of double-strand breaks when damage occurs [11–13]. The knowledge about DDR pathways is progressing rapidly; however, the exact mechanisms underlying this network require further elucidation. Among the many proteins involved in DDR pathways, the prognostic relevance of ERCC1 for stage I NSCLC was tested [14]. Notably, the information on the prognostic significance of proteins involved in DDR pathways is very limited.

In this study, we hypothesized that proteins involved in DDR pathways have effects on the survival of patients with pathological stage I NSCLC.

methods

patients and tissue specimens

Consecutive patients with pathological stage I squamous cell carcinoma or adenocarcinoma were enrolled retrospectively from 11 university hospitals in South Korea (details provided in supplemental Figure S1, available at Annals of Oncology online). Patients who received surgical resection between January 1997 and December 2005 were considered for this study. They were pathologically staged using the criteria of AJCC, sixth edition [15]. Inclusion criteria were as follows: (i) complete surgical removal of tumor (R0 resection); (ii) availability of formalin-fixed, paraffin-embedded (FFPE) primary lung cancer tissue blocks; and (iii) follow-up of >24 months after surgical resection. Patients who received limited resection, such as wedge resection or segmentectomy, any anticancer treatment prior to surgical resection, or had history of other cancers were excluded.

Information regarding smoking habits, Charlson comorbidity index score (CCIS), types of surgery, tumor size, and lymphatic or blood vessel invasion on the pathological specimens was collected. To increase the quality of the information, the reports of radiological examinations (chest computed tomography scan, positron emission tomography, etc.), surgical pathology, and clinical information for all patients were reviewed independently. All FFPE blocks were sent to a central laboratory (SuperBioChips, Seoul, South Korea) to construct tissue microarrays and for immunohistochemical staining. The study protocol was approved by the Institutional Review Boards of the university hospitals and informed consent by the patient was waived.

construction of tissue microarrays and immunohistochemistry

Slides of tumor samples stained with hematoxylin and eosin and matched FFPE samples with a marked representative area were reviewed. Core tissue biopsy specimens (2 mm in diameter) were obtained from individual FFPE samples (donor blocks) and arranged in a new recipient paraffin block (tissue array block) using a trephine apparatus. Each tissue array block contained up to 50 specimens, which allowed all 984 specimens to be contained in 20 array blocks. Serial sections (4 μm) from FFPE blocks were analyzed for protein expression. Among the proteins involved in DDR pathways, 11 commercially available antibodies against the wild-type and phosphorylated forms of each protein were chosen for ATM, ATMpS1981, γH2AX, 53BP1, 53BP1pS25, Chk2, Chk2pT68, MDC1, MDC1pS964, BRCA1pS1423, and ERCC1.

All immunohistochemical stainings were performed at the SuperBioChips Laboratories. Immunohistochemistry was performed using a Leica Bond-max automated immunostainer (Leica Microsystems, Newcastle, UK), according to the manufacturer's protocols. For antigen retrieval, slides were heated for 20 min at 100°C in antigen retrieval buffer (pH 6.0 citrate buffer and pH 9.0 EDTA buffer). Endogenous biotin was blocked using the appropriate kit. Subsequently, slides were incubated with the primary antibody for 15 min. The dilution ratio and optimal concentration of retrieval buffer for each antibody are shown in supplemental Table S1 (available at Annals of Oncology online). Antibody binding was detected using the Bond Polymer Refined Detection kit (Leica Microsystems). Mayer's hematoxylin was used as the counterstain.

evaluation of immunohistochemistry

Samples containing >1000 malignant cells were considered for evaluation. For comparable semiquantitative estimation, the H score method used for ERCC1 was applied to the proteins included in this study [16]. Briefly, the percentage of tumor cells with positive nucleus staining on each tumor slide was calculated and was attributed a score of 0, 0.1, 0.5, or 1 for 0%, 1%–9%, 10%–49%, or >50% of stained tumor cells, respectively. The intensity of staining was recorded and given a score of 0, 1, 2, or 3. H scores were obtained by multiplying the proportion by the intensity of the staining. Each sample was reviewed independently by two pathologists (HJJ and LK) who were blinded to the clinicopathological variables and survival status. The mean H score was calculated for each sample. Subsequently, the median value of all mean H scores was used as a criterion for the classification of the expression of the 11 proteins as ‘low’ and ‘high’.

survival measurements

The end point of this study was the estimation of overall survival from the date of surgery to death due to any cause. Dates of death were obtained principally from the revision of medical records. To increase the quality of the information regarding the survival status of patients lost to follow-up, this missing information was obtained by contacting the patients or their relatives by phone or mail (56 patients) or from the Ministry of Public Administration and Security, South Korea (87 patients).

statistical analysis

Expression of the proteins was dichotomized as low versus high. The chi-square test for heterogeneity was used to compare the distributions of the clinicopathological variables or levels of protein expression between squamous cell carcinoma and adenocarcinoma. Spearman's correlation test was used to evaluate potential relationships among the expression levels of the proteins. The effect of each clinicopathological variable or protein on overall survival was estimated using the Kaplan–Meier method and the log-rank test. Hazards ratio (HR) adjusted for all clinicopathological variables with a log-rank P value <0.05 and the 95% confidence interval (CI) were estimated using the Cox proportional hazards model. To verify if the interaction of two proteins with a Cox P value <0.05 potentiated the effect, the effect of the combinations was analyzed after adjustment for all the variables. All significance testing was performed at a two-sided 0.05 level. All analyses were performed using the SAS software package (v. 9.1.3).

results

characteristics of the patients

Table 1 shows the clinical characteristics of the 889 patients (401 squamous cell carcinoma and 488 adenocarcinoma cases) (supplemental Table S2, available at Annals of Oncology online). The median age of the patients was 64.5 years and the median survival time was 136 months. As adjuvant chemotherapy, platinum doublets were given to 98 patients (11% of total of patients) and included platinum plus vinorelbine (54 patients), vinorelbine and ifosfamide (20 patients), paclitaxel (Taxol, Bristol-Myers Squibb, NY) (15 patients), etoposide (3 patients), cyclophosphamide and doxorubicin (5 patients), and gemcitabine (1 patient). Among the whole cohort, 316 patients (35.5%) died and 573 patients (64.5%) survived. The distributions of gender, smoking habits, size of tumor, type of surgery, and lymphatic vessel invasion were significantly different between the squamous cell carcinoma and adenocarcinoma groups. There were 345 ever smokers (87.1%) among the patients with squamous cell carcinoma, whereas 256 of the patients with adenocarcinoma were never smokers (52.5%).

Table 1.

Baseline characteristics of the 889 patients with pathological stage I non-small-cell lung cancer according to histology

Variables SQC ADC P 
Age    
 Median (range) 66 (34–85) 63 (30–81)  
Gender   <0.001 
 Men 370 (57.6) 272 (42.4) 
 Women 31 (12.6) 216 (87.4) 
Smoking habit   <0.001 
 Never 51 (16.6) 256 (83.3) 
 Ever 345 (59.9) 231 (40.1) 
CCIS   0.814 
 0 169 (46.6) 194 (53.4) 
 1 126 (47.5) 139 (52.5) 
 ≥2 106 (44.7) 131 (55.3) 
Tumor size, cm   <0.001 
 ≤2 77 (41.2) 110 (58.8) 
 >2 to ≤3 97 (35.3) 178 (64.7) 
 >3 to ≤5 155 (51.8) 144 (48.2) 
 >5 72 (57.1) 54 (42.9) 
Type of surgery   <0.001 
 Lobectomy 312 (41.1) 447 (58.9) 
 Bilobectomy 37 (57.8) 27 (42.2) 
 Pneumonectomy 52 (78.8) 14 (21.2) 
Lymphatic vessel invasion   0.049 
 Yes 67 (53.2) 59 (46.8) 
 No 330 (43.8) 424 (56.2) 
Blood vessel invasion   0.054 
 Yes 66 (52.8) 59 (47.2) 
 No 326 (43.5) 423 (56.5) 
Adjuvant chemotherapy   0.389 
 Yes 59 (41.8) 82 (58.2) 
 Platinum doublets 53 (54.1) 45 (45.9) 
 UFT 6 (13.9) 37 (86.1) 
 No 342 (45.8) 405 (54.2) 
MST, months (95% CI) 123.8 (100.6–147.0) NR 0.133a 
Deaths 155 161  
Variables SQC ADC P 
Age    
 Median (range) 66 (34–85) 63 (30–81)  
Gender   <0.001 
 Men 370 (57.6) 272 (42.4) 
 Women 31 (12.6) 216 (87.4) 
Smoking habit   <0.001 
 Never 51 (16.6) 256 (83.3) 
 Ever 345 (59.9) 231 (40.1) 
CCIS   0.814 
 0 169 (46.6) 194 (53.4) 
 1 126 (47.5) 139 (52.5) 
 ≥2 106 (44.7) 131 (55.3) 
Tumor size, cm   <0.001 
 ≤2 77 (41.2) 110 (58.8) 
 >2 to ≤3 97 (35.3) 178 (64.7) 
 >3 to ≤5 155 (51.8) 144 (48.2) 
 >5 72 (57.1) 54 (42.9) 
Type of surgery   <0.001 
 Lobectomy 312 (41.1) 447 (58.9) 
 Bilobectomy 37 (57.8) 27 (42.2) 
 Pneumonectomy 52 (78.8) 14 (21.2) 
Lymphatic vessel invasion   0.049 
 Yes 67 (53.2) 59 (46.8) 
 No 330 (43.8) 424 (56.2) 
Blood vessel invasion   0.054 
 Yes 66 (52.8) 59 (47.2) 
 No 326 (43.5) 423 (56.5) 
Adjuvant chemotherapy   0.389 
 Yes 59 (41.8) 82 (58.2) 
 Platinum doublets 53 (54.1) 45 (45.9) 
 UFT 6 (13.9) 37 (86.1) 
 No 342 (45.8) 405 (54.2) 
MST, months (95% CI) 123.8 (100.6–147.0) NR 0.133a 
Deaths 155 161  

P values were obtained by two-sided chi-square test.

aP values were obtained by log-rank test.

ADC, adenocarcinoma; CCIS, Charlson comorbidity index score; CI, confidence interval; MST, median survival time, months; NR, not reached; SQC, squamous cell carcinoma; UFT, uracil and tegafur.

Age, gender, smoking habits, CCIS, size of tumor, type of surgery, and lymphatic or blood vessel invasion were significantly associated with the survival of the patients. As expected, adjuvant chemotherapy was not a prognostic factor.

expression of proteins involved in DDR pathways according to histology

The level of expression of proteins (low versus high) was evaluated in the two distinct histologies (Table 2). High expression of ATMpS1981, γH2AX, 53BP1, 53BP1pS25, Chk2, Chk2pT68, MDC1, BRCA1pS1423, and ERCC1 was observed in squamous cell carcinoma compared with adenocarcinoma. The expression of most proteins showed a significant correlation to each other (P < 0.001) (supplemental Table S3, available at Annals of Oncology online).

Table 2.

Expressions of proteins involved in DNA damage response pathways according to histology

Variables Protein expression SQC ADC P 
ATM Low 345 (44.7) 427 (55.3) 0.550 
High 56 (47.9) 61 (52.1) 
ATMpS1981 Low 187 (40.4) 276 (59.6) 0.003 
High 214 (50.2) 212 (49.8) 
γH2AX Low 182 (36.7) 314 (63.3) <0.001 
High 219 (55.7) 174 (44.3) 
53BP1 Low 130 (32.3) 273 (67.7) <0.001 
High 271 (55.9) 214 (44.1) 
53BP1pS25 Low 136 (34.3) 261 (65.7) <0.001 
High 265 (54.0) 226 (46.0) 
Chk2 Low 243 (39.8) 368 (60.2) <0.001 
High 158 (56.8) 120 (43.2) 
Chk2pT68 Low 250 (38.0) 408 (62.0) <0.001 
High 151 (65.7) 79 (34.3) 
MDC1 Low 170 (36.2) 299 (63.8) <0.001 
High 231 (55.1) 188 (44.9) 
MDC1pS964 Low 225 (47.6) 248 (52.4) 0.123 
High 176 (42.4) 239 (57.6) 
BRCA1pS1423 Low 188 (41.4) 266 (58.6) 0.022 
High 213 (49.1) 221 (50.9) 
ERCC1 Low 132 (32.5) 274 (67.5) <0.001 
High 267 (55.5) 214 (44.5) 
Variables Protein expression SQC ADC P 
ATM Low 345 (44.7) 427 (55.3) 0.550 
High 56 (47.9) 61 (52.1) 
ATMpS1981 Low 187 (40.4) 276 (59.6) 0.003 
High 214 (50.2) 212 (49.8) 
γH2AX Low 182 (36.7) 314 (63.3) <0.001 
High 219 (55.7) 174 (44.3) 
53BP1 Low 130 (32.3) 273 (67.7) <0.001 
High 271 (55.9) 214 (44.1) 
53BP1pS25 Low 136 (34.3) 261 (65.7) <0.001 
High 265 (54.0) 226 (46.0) 
Chk2 Low 243 (39.8) 368 (60.2) <0.001 
High 158 (56.8) 120 (43.2) 
Chk2pT68 Low 250 (38.0) 408 (62.0) <0.001 
High 151 (65.7) 79 (34.3) 
MDC1 Low 170 (36.2) 299 (63.8) <0.001 
High 231 (55.1) 188 (44.9) 
MDC1pS964 Low 225 (47.6) 248 (52.4) 0.123 
High 176 (42.4) 239 (57.6) 
BRCA1pS1423 Low 188 (41.4) 266 (58.6) 0.022 
High 213 (49.1) 221 (50.9) 
ERCC1 Low 132 (32.5) 274 (67.5) <0.001 
High 267 (55.5) 214 (44.5) 

Expression of the proteins was classified into low and high by using a criterion of median value of all mean H scores. P values were obtained by two-sided chi-square test.

ADC, adenocarcinoma; SQC, squamous cell carcinoma.

effect of the proteins on the survival of stage I NSCLC patients

Patients with low expression of BRCA1pS1423 or ERCC1 exhibited a significantly worse survival compared with those with high expression of these proteins, after adjustment for age, gender, smoking habits, CCIS, size of tumor, type of surgery, and lymphatic or blood vessel invasion (log-rank P = 0.002 for BRCA1pS1423 and 0.017 for ERCC1). The separation of the cohort into two groups according to histology revealed that patients with low expression of γH2AX, Chk2pT68, or ERCC1 had a significantly worse survival compared with those with high expression of these proteins, after adjustment for the clinicopathological variables in squamous cell carcinoma [aHR (95% CI) and Cox P, 1.544 (1.100–2.168) and 0.012 for γH2AX; 1.624 (1.125–) and 0.010 for Chk2pT68; and 1.569 (1.110–2.217) and 0.011 for ERCC1] (Table 3). Expression of BRCA1pS1423 was marginally prognostic for patients [aHR (95% CI) and Cox P: 1.353 (0.965–1.892) and 0.080]. However, these effects of the proteins on survival were not observed in adenocarcinoma patients.

Table 3.

Expressions of proteins involved in DNA damage response pathways and survival of the 889 patients of pathological stage I non-small-cell lung cancer according to histology

Variables Protein expression SQC
 
ADC
 
aHR (95% CI) Cox P aHR (95% CI) Cox P 
ATM Low 0.798 (0.474–1.346) 0.398 1.150 (0.698–1.896) 0.583 
High 
ATMpS1981 Low 1.292 (0.914–1.827) 0.147 1.082 (0.774–1.512) 0.646 
High 
γH2AX Low 1.544 (1.100–2.168) 0.012 0.998 (0.702–1.419) 0.993 
High 
53BP1 Low 0.960 (0.674–1.367) 0.821 0.930 (0.665–1.301) 0.672 
High 
53BP1pS25 Low 1.071 (0.759–1.511) 0.697 0.913 (0.657–1.269) 0.586 
High 
Chk2 Low 1.242 (0.871–1.772) 0.232 1.057 (0.720–1.554) 0.777 
High 
Chk2pT68 Low 1.624 (1.125–2.343) 0.010 0.725 (0.478–1.101) 0.131 
High 
MDC1 Low 1.066 (0.766–1.482) 0.705 0.939 (0.652–1.312) 0.713 
High 
MDC1pS964 Low 1.216 (0.860–1.719) 0.269 1.169 (0.837–1.633) 0.358 
High 
BRCA1pS1423 Low 1.353 (0.965–1.892) 0.080 1.262 (0.896–1.778) 0.183 
High 
ERCC1 Low 1.569 (1.110–2.217) 0.011 1.023 (0.729–1.437) 0.894 
High 
Variables Protein expression SQC
 
ADC
 
aHR (95% CI) Cox P aHR (95% CI) Cox P 
ATM Low 0.798 (0.474–1.346) 0.398 1.150 (0.698–1.896) 0.583 
High 
ATMpS1981 Low 1.292 (0.914–1.827) 0.147 1.082 (0.774–1.512) 0.646 
High 
γH2AX Low 1.544 (1.100–2.168) 0.012 0.998 (0.702–1.419) 0.993 
High 
53BP1 Low 0.960 (0.674–1.367) 0.821 0.930 (0.665–1.301) 0.672 
High 
53BP1pS25 Low 1.071 (0.759–1.511) 0.697 0.913 (0.657–1.269) 0.586 
High 
Chk2 Low 1.242 (0.871–1.772) 0.232 1.057 (0.720–1.554) 0.777 
High 
Chk2pT68 Low 1.624 (1.125–2.343) 0.010 0.725 (0.478–1.101) 0.131 
High 
MDC1 Low 1.066 (0.766–1.482) 0.705 0.939 (0.652–1.312) 0.713 
High 
MDC1pS964 Low 1.216 (0.860–1.719) 0.269 1.169 (0.837–1.633) 0.358 
High 
BRCA1pS1423 Low 1.353 (0.965–1.892) 0.080 1.262 (0.896–1.778) 0.183 
High 
ERCC1 Low 1.569 (1.110–2.217) 0.011 1.023 (0.729–1.437) 0.894 
High 

Expression of the proteins was classified into low and high by using a criterion of median value of all mean H scores.

ADC, adenocarcinoma; aHR, hazards ratio adjusted with age, gender, smoking habit, Charlson comorbidity index score, tumor size, type of surgery, and lymphatic or blood vessel invasion; CI, confidence interval; MST, median survival time, months; NR, not reached; SQC, squamous cell carcinoma.

combined effect of the proteins on the survival of the patients according to histology

The interactions between two proteins that were shown to have a significant prognostic effect in squamous cell carcinoma patients were analyzed to determine whether their combinations exert a synergistic or discriminative effect on the survival of the patients. The combination of any two of the γH2AX, Chk2pT68, or ERCC1 proteins potentiated the prognostic effect in squamous cell carcinoma patients [aHR (95% CI) and Cox P: 1.982 (1.271–3.090) and 0.003 for γH2AX and Chk2pT68; 2.081 (1.334–3.248) and 0.001 for γH2AX and ERCC1; and 2.048 (1.303–3.221) and 0.002 for Chk2pT68 and ERCC1] (Figure 1; supplemental Table S4, available at Annals of Oncology online). Similarly, these proteins had no effect on survival in adenocarcinoma patients [aHR (95% CI) and Cox P: 0.721 (0.424–1.226) and 0.721 for γH2AX and Chk2pT68; 0.915 (0.578–1.448) and 0.704 for γH2AX and ERCC1; and 0.785 (0.470–1.311) and 0.355 for Chk2pT68 and ERCC1].

Figure 1

Combined effect of γH2AX, Chk2pT68, and ERCC1 on the survival of the 889 patients with pathological stage I non-small-cell lung cancer according to histology: squamous cell carcinoma (A) γH2AX/Chk2pT68, (B) γH2AX/ERCC1, (C) ERCC1/Chk2pT68 and adenocarcinoma, (D) γH2AX/Chk2pT68, (E) γH2AX/ERCC1, and (F) ERCC1/Chk2pT68.

Figure 1

Combined effect of γH2AX, Chk2pT68, and ERCC1 on the survival of the 889 patients with pathological stage I non-small-cell lung cancer according to histology: squamous cell carcinoma (A) γH2AX/Chk2pT68, (B) γH2AX/ERCC1, (C) ERCC1/Chk2pT68 and adenocarcinoma, (D) γH2AX/Chk2pT68, (E) γH2AX/ERCC1, and (F) ERCC1/Chk2pT68.

discussion

The present study demonstrated that proteins involved in DDR pathways, such as γH2AX, Chk2pT68, and ERCC1, are strong prognostic factors for patients with pathological stage I squamous cell carcinoma, but not for adenocarcinoma patients.

Proteins involved in DDR pathways play an important role in suppressing carcinogenesis or the progression of lung cancer, as well as maintaining genomic stability [10, 11]. Therefore, they can be good candidates for a druggable target or a biomarker to predict prognosis or treatment response [17].

Regarding the distribution of the expression of proteins between squamous cell carcinoma and adenocarcinoma, a study using 187 patients reported that expression of ERCC1 did not vary according to histology, gender, or smoking habits [14]. However, the result of another study using 130 patients with early stage NSCLC showed that high expression of ERCC1 was observed in squamous cell carcinoma compared with adenocarcinoma [18]. This finding was also reproduced in other studies using patients with operable or advanced stage NSCLC [16, 19]. In this study, the expression of proteins, including ERCC1, was explored, for the first time, in the 889 patients with pathological stage I disease. High expression of all the proteins, with the exception of ATM and MDC1pS964, was observed in squamous cell carcinoma compared with adenocarcinoma (Table 2). A study using cell lines showed that smoking induced ATM activation and phosphorylation of H2AX on Ser 139 and Chk2 on Thr68 [20]. This can be regarded as similar to the process of repairing double-strand breaks and is in accordance with the results of this study showing that 87% of patients with squamous cell carcinoma were ever smokers. This indicates the presence of biological heterogeneity between the two histologies, in terms of DDR pathways. The significant correlations observed among the expression (low versus high) of almost all proteins listed in supplemental Table S3 (available at Annals of Oncology online) may represent clinical evidence for the role of the network of DDR pathways in stage I NSCLC.

Histology is well recognized as an important phenotype in NSCLC patients [5, 21, 22]. As a phenotype, it also can be more reproducible and consistent compared with information on smoking habits, especially in this kind of retrospective study [23]. In this study, we evaluated the effect of the expression of DNA repair proteins in patients with squamous cell carcinoma or adenocarcinoma.

ERCC1 was shown to be prognostic in stage I NSCLC patients [14]. However, the prognostic relevance of most other DDR proteins has never been examined in patients. The consideration of the efficiency of the development of a biomarker may lead to a pathway-based approach that is more robust rather than that developed to use a single protein. This study showed that the proteins, or their combinations, had a significant prognostic effect that was significantly dependent on histology. In squamous cell carcinoma patients, low expression of γH2AX, Chk2pT68, or ERCC1 was associated with a 1.5–1.6 times higher risk for death compared with those exhibiting high expression (Table 3). The analysis of the combined effect of γH2AX, Chk2pT68, or ERCC1 showed that the patients with low/low expression levels exhibited a risk for death that was two times higher than that of patients with high/high expression levels (Figure 1; supplemental Table S4, available at Annals of Oncology online). To know if cumulative effect of the three proteins is present, the patients were classified into scores of 0, 1, 2, or 3 of indicating numbers of the proteins with low expression (supplemental Table S5, available at Annals of Oncology online). The risk for death increased as proportion as the score increased in squamous cell carcinoma. These results seem to be plausible, as low expression of these proteins can be speculated to confer insufficient response to DNA damage, which is critical to maintaining genomic stability or for the progression of lung cancer. The stratum holding prognostic relevance from the combined effect, i.e. the percentage of patients with low/low or high/high expression levels of γH2AX, Chk2pT68, or ERCC1, was 62%–68% of the total patients.

Squamous cell carcinoma remains a major histology of lung cancer. However, it has been relatively neglected in the development of new drugs or biomarkers compared with adenocarcinoma. This may be attributed to the fact that it is considered a ‘smart cancer’ characterized by biological complexities compared with ‘stupid cancer’ characterized by oncogenic addiction such as the activation of EGFR mutation or EML4-ALK translocation, which is observed mostly in adenocarcinoma [5, 21]. In this regard, this study is important for the development of biomarkers that are specific for squamous cell carcinoma.

These results must be interpreted with some caution. This study may be limited potentially by its retrospective nature, as well as by not being validated using an independent cohort. We could not evaluate potential changes in smoking habits after surgery, which could affect the survival of patients. Nevertheless, we used every endeavor to increase the quality of the information on clinicopathological variables and identify the survival status of the patients. Moreover, we evaluated comorbidity (which is an important determinant of survival) as a confounding variable [24]. A large number of consecutive, pathological stage I patients were enrolled in this study. The sample size would be enough for this kind of study or for the interpretation using the subgroups of squamous cell carcinoma or adenocarcinoma. We used 8F1 as an antibody against ERCC1. Regarding the appropriateness of antibodies for ERCC1, 8F1 was compared with FL297; however, this ongoing issue warrants further validation in clinical samples [25].

In summary, proteins involved in DDR pathways showed biological heterogeneity between squamous cell carcinoma and adenocarcinoma. Proteins such as γH2AX, Chk2pT68, and ERCC1 may be helpful for the classification of the squamous cell carcinoma patients into groups of low or high risk for death.

funding

Korea Healthcare Technology R&D Project, Ministry for Health, Welfare & Family Affairs, Republic of Korea (A084452 and A110518).

disclosure

The authors declare no conflicts of interests.

references

1
Jemal
A
Siegel
R
Xu
J
Ward
E
Cancer statistics, 2010
CA Cancer J Clin
 , 
2011
, vol. 
60
 (pg. 
277
-
300
)
2
Jung
KW
Park
S
Kong
HJ
, et al.  . 
Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2008
Cancer Res Treat
 , 
2011
, vol. 
43
 (pg. 
1
-
11
)
3
Felip
E
Cedrés
S
Checa
E
Martinez
P
How to integrate current knowledge in selecting patients for first line in NSCLC?
Ann Oncol
 , 
2010
, vol. 
21
 
(Suppl 7)
(pg. 
vii230
-
vii233
)
4
Konopa
K
Do we have markers to select patients for adjuvant therapies of non-small-cell lung cancer?
Ann Oncol
 , 
2010
, vol. 
21
 
(Suppl 7)
(pg. 
vii199
-
vii202
)
5
Postmus
PE
Screening for lung cancer, an ongoing debate
Ann Oncol
 , 
2008
, vol. 
19
 
(Suppl 7)
(pg. 
vii25
-
vii27
)
6
Goldstraw
P
Crowley
J
Chansky
K
, et al.  . 
The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM Classification of malignant tumours
J Thorac Oncol
 , 
2007
, vol. 
2
 (pg. 
706
-
714
)
7
Arriagada
R
Auperin
A
Burdett
S
, et al.  . 
Adjuvant chemotherapy, with or without postoperative radiotherapy, in operable non-small-cell lung cancer: two meta-analyses of individual patient data
Lancet
 , 
2010
, vol. 
375
 (pg. 
1267
-
1277
)
8
Pignon
JP
Tribodet
H
Scagliotti
GV
, et al.  . 
Lung adjuvant cisplatin evaluation: a pooled analysis by the LACE Collaborative Group
J Clin Oncol
 , 
2008
, vol. 
26
 (pg. 
3552
-
3559
)
9
Ciccia
A
Elledge
SJ
The DNA damage response: making it safe to play with knives
Mol Cell
 , 
2010
, vol. 
40
 (pg. 
179
-
204
)
10
Nuciforo
PG
Luise
C
Capra
M
, et al.  . 
Complex engagement of DNA damage response pathways in human cancer and in lung tumor progression
Carcinogenesis
 , 
2007
, vol. 
28
 (pg. 
2082
-
2088
)
11
Gorgoulis
VG
Vassiliou
LV
Karakaidos
P
, et al.  . 
Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions
Nature
 , 
2005
, vol. 
434
 (pg. 
907
-
913
)
12
Bartkova
J
Horejsi
Z
Sehested
M
, et al.  . 
DNA damage response mediators MDC1 and 53BP1: constitutive activation and aberrant loss in breast and lung cancer, but not in testicular germ cell tumours
Oncogene
 , 
2007
, vol. 
26
 (pg. 
7414
-
7422
)
13
Bonner
WM
Redon
CE
Dickey
JS
, et al.  . 
GammaH2AX and cancer
Nat Rev Cancer
 , 
2008
, vol. 
8
 (pg. 
957
-
967
)
14
Zheng
Z
Chen
T
Li
X
, et al.  . 
DNA synthesis and repair genes RRM1 and ERCC1 in lung cancer
N Engl J Med
 , 
2007
, vol. 
356
 (pg. 
800
-
808
)
15
Mountain
CF
Revisions in the International System for Staging Lung Cancer
Chest
 , 
1997
, vol. 
111
 (pg. 
1710
-
1717
)
16
Olaussen
KA
Dunant
A
Fouret
P
, et al.  . 
DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy
N Engl J Med
 , 
2006
, vol. 
355
 (pg. 
983
-
991
)
17
Lapenna
S
Giordano
A
Cell cycle kinases as therapeutic targets for cancer
Nat Rev Drug Discov
 , 
2009
, vol. 
8
 (pg. 
547
-
566
)
18
Lee
KH
Min
HS
Han
SW
, et al.  . 
ERCC1 expression by immunohistochemistry and EGFR mutations in resected non-small cell lung cancer
Lung Cancer
 , 
2008
, vol. 
60
 (pg. 
401
-
407
)
19
Holm
B
Mellemgaard
A
Skov
T
Skov
BG
Different impact of excision repair cross-complementation group 1 on survival in male and female patients with inoperable non-small-cell lung cancer treated with carboplatin and gemcitabine
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
4254
-
4259
)
20
Zhao
H
Albino
AP
Jorgensen
E
, et al.  . 
DNA damage response induced by tobacco smoke in normal human bronchial epithelial and A549 pulmonary adenocarcinoma cells assessed by laser scanning cytometry
Cytometry A
 , 
2009
, vol. 
75
 (pg. 
840
-
847
)
21
Lynch
TJ
Bell
DW
Sordella
R
, et al.  . 
Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib
N Engl J Med
 , 
2004
, vol. 
350
 (pg. 
2129
-
2139
)
22
Scagliotti
GV
Parikh
P
von Pawel
J
, et al.  . 
Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-cell lung cancer
J Clin Oncol
 , 
2008
, vol. 
26
 (pg. 
3543
-
3551
)
23
Lee
E
Ryu
J
Cho
J
, et al.  . 
Comparison of face-to-face interview questionnaires and medical record data for smoking habit in lung cancer patients
Tuberc Respir Dis
 , 
2007
, vol. 
62
 (pg. 
27
-
32
)
24
Jacot
W
Colinet
B
Bertrand
D
, et al.  . 
Quality of life and comorbidity score as prognostic determinants in non-small-cell lung cancer patients
Ann Oncol
 , 
2008
, vol. 
19
 (pg. 
1458
-
1464
)
25
Olaussen
KA
Soria
JC
Validation of ERCC1-XPF immunodetection-letter
Cancer Res
 , 
2010
, vol. 
70
 (pg. 
3851
-
3852
)

Author notes

Both authors contributed equally to this work.