• Open Access

Overexpression of T-LAK cell-originated protein kinase predicts poor prognosis in patients with stage I lung adenocarcinoma

Authors

  • Di-Cing Wei,

    1. Department of Life Science, College of Science and Engineering, Fu Jen Catholic University, New Taipei City
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Yi-Chen Yeh,

    1. Department of Pathology, Taipei Veterans General Hospital, Taipei
    2. Institute of Clinical Medicine, National Yang-Ming University, Taipei
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Jung-Jyh Hung,

    1. Division of Thoracic Surgery, Department of Surgery, Taipei Veterans General Hospital, Taipei
    2. School of Medicine, National Yang-Ming University, Taipei
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Teh-Ying Chou,

    1. Department of Pathology, Taipei Veterans General Hospital, Taipei
    2. Institute of Clinical Medicine, National Yang-Ming University, Taipei
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Yu-Chung Wu,

    1. Division of Thoracic Surgery, Department of Surgery, Taipei Veterans General Hospital, Taipei
    2. School of Medicine, National Yang-Ming University, Taipei
    Search for more papers by this author
  • Pei-Jung Lu,

    1. Institute of Clinical Medicine, National Cheng Kung University, College of Medicine, Tainan
    Search for more papers by this author
  • Hui-Chuan Cheng,

    1. Institute of Clinical Medicine, National Cheng Kung University, College of Medicine, Tainan
    Search for more papers by this author
  • Yu-Lin Hsu,

    1. Department of Life Science, College of Science and Engineering, Fu Jen Catholic University, New Taipei City
    Search for more papers by this author
  • Yu-Lun Kuo,

    1. Department of Computer Science and Information Engineering, National Taiwan University, Taipei
    Search for more papers by this author
  • Kuan-Yu Chen,

    1. Division of Pulmonary Medicine, Department of Internal Medicine, National Taiwan University Hospital, College of Medicine, Taipei, Taiwan
    Search for more papers by this author
  • Jin-Mei Lai

    Corresponding author
    • Department of Life Science, College of Science and Engineering, Fu Jen Catholic University, New Taipei City
    Search for more papers by this author

To whom correspondence should be addressed. E-mail: jmlai@mail.fju.edu.tw

Abstract

Tumor recurrence is the most common cause of disease failure after surgical resection in early-stage lung adenocarcinoma. Identification of clinically relevant prognostic markers could help to predict patients with high risk of disease recurrence. A meta-analysis of available lung adenocarcinoma microarray datasets revealed that T-LAK cell-originated protein kinase (TOPK), a serine/threonine protein kinase, is overexpressed in lung cancer. Using stable cell lines with overexpression or knockdown of TOPK, we have shown that TOPK can promote cell migration, invasion, and clonogenic activity in lung cancer cells, suggesting its crucial role in lung tumorigenesis. To evaluate the prognostic value of TOPK expression in resected stage I lung adenocarcinoma, a retrospective analysis of 203 patients diagnosed with pathological stage I lung adenocarcinoma was carried out to examine the expression of TOPK by immunohistochemistry (IHC). The prognostic significance of TOPK overexpression was examined. Overexpression of TOPK (IHC score >3) was detected in 67.0% of patients, and these patients were more frequently characterized with disease recurrence and angiolymphatic invasion. Using multivariate analysis, patient age (>65 years old; = 0.002) and TOPK overexpression (IHC score >3; P < 0.001) significantly predicted a shortened overall survival. Moreover, TOPK overexpression (IHC score >3; P = 0.005) also significantly predicted a reduced time to recurrence in the patients. Our results indicate that overexpression of TOPK could predetermine the metastatic capability of tumors and could serve as a significant prognostic predictor of shortened overall survival and time to recurrence. (Cancer Sci 2012; 103: 731–738)

Lung cancer is the leading cause of cancer-related death worldwide. The incidence of adenocarcinoma, the most common histological subtype of non-small-cell lung carcinoma (NSCLC) in most countries, has increased in both sexes during the past several decades, whereas squamous cell carcinoma has decreased.[1, 2] Surgical resection is the main treatment of choice for early-stage NSCLC.[3] Although patient survival after stage I NSCLC resection is good, postoperative recurrence has been reported to occur in 22–38% of patients.[4-7] The survival rate is poor in patients with recurrence following resection of stage I NSCLC.[6, 8] Many randomized clinical trials have evaluated the role of adjuvant chemotherapy in patients with resected stage I NSCLC and have tried to identify patients with a higher risk of recurrence or poor prognosis after surgical resection.[9-11] Investigation of biomarkers of early-stage NSCLC may be helpful in designing clinical trials for adjuvant chemotherapy in the future.

T-LAK cell-originated protein kinase (TOPK), also known as PDZ-binding kinase (PBK), is a MAPKK-like serine/threonine protein kinase that was identified as an interleukin-2-induced gene in T-LAK cells and as an interaction partner with the human tumor suppressor hDlg, identified by yeast two-hybrid screening.[12, 13] TOPK is barely detected in normal tissue, with the exception of germ cells in the testis and several fetal tissues. TOPK has been shown to be upregulated in various types of cancer.[14-17] The oncogenic role of TOPK has been reported in a panel of cancers. For example, the knockdown of TOPK expression by siRNA inhibits the growth and clonogenicity of breast cancer cells. TOPK has been shown to mediate UVB-induced JNK activation and enhance H-Ras-induced cell transformation.[18] In addition, TOPK serves as an oncogenic MEK that exerts positive feedback on ERK2 to promote colorectal cancer formation in vitro and in vivo.[19] TOPK has also been identified as a downstream target of the EWS-FLI chimeric fusion protein, which plays a significant role in Ewing sarcoma tumorigenesis.[20] Furthermore, TOPK physically interacts with the DBD domain of p53, which might contribute to tumorigenesis by affecting the tumor suppressor function of p53.[21] We have also recently shown that TOPK can stimulate AKT-dependent cell migration/invasion by relieving the PTEN-dependent suppressive effect, indicating its crucial role in facilitating cancer metastasis.[22]

Although accumulating reports have indicated the crucial roles of TOPK in tumorigenesis, its role in lung tumorigenesis and its prognostic value in lung cancer, especially stage I lung adenocarcinoma, have not been investigated. In this report, we analyzed publicly available microarray datasets and found that TOPK was overexpressed in lung adenocarcinoma. By establishing stable cell lines with overexpression or knockdown of TOPK, we have determined the crucial role of TOPK in cell migration, invasion, and clonogenic activity in lung cancer cells. Increased TOPK expression was observed in a significant percentage of patients with stage I lung adenocarcinoma by immunohistochemistry (IHC). Ultimately, overexpression of TOPK was a significant prognostic factor of shortened overall survival and time to recurrence.

Materials and Methods

Microarray data source and analysis

In this study, we used a total of 196 samples from three cohorts of patients with lung adenocarcinoma for microarray and box plot analyses. Microarray dataset 1 (GSE7670) was obtained online from NCBI Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo/) and originated from Su et al.[23] This dataset included 16 stage IA/IB (eight patients) and 22 stage III/IV (11 patients) lung adenocarcinoma pairwise samples and was subjected to microarray analysis on the Affymetrix HG-U133A chip (Affymetrix, Santa Clara, CA, USA).[23] Microarray dataset 2 (GSE10072) was obtained online from GEO and originated from Landi et al.[24] This dataset included 30 stage IA/IB (15 patients) and 18 stage III/IV (nine patients) lung adenocarcinoma pairwise samples and was obtained using the Affymetrix HG-U133A chip. Microarray dataset 3 (GSE19804) was published by Lu et al.[25] and included 62 stage IA/IB (31 patients), 24 stage IIA/IIB (12 patients), and 24 stage III/IV (12 patients) lung adenocarcinoma pairwise samples and was obtained using the Affymetrix HG-U133 plus 2.0 chip. The lung cancer microarray data were analyzed by preprocessing the raw fluorescence intensity data within CEL files. The data were normalized using the Bioconductor Affy package (Bioconductor), which is based on the Robust Multichip Average (RMA) algorithm with R language (http://www.bioconductor.org).

Clinicopathological data and tissue microarray construction

From 1995 to 2007, 203 patients with stage I lung adenocarcinoma were enrolled in the study. Samples were collected and deposited into the surgical pathology archives at the Taipei Veterans General Hospital (Taipei, Taiwan). All patients underwent curative surgical resection, accompanied by complete hilar and mediastinal lymph nodes dissection. No patient received adjuvant chemotherapy or radiotherapy after surgical resection. Determination of disease stage was based on the seventh edition of the TNM classification of the American Joint Committee on Cancer.[26] The pathological slides were reviewed by two pathologists (T.-Y.C. and Y.-C.Y.). Representative parts of the tumor tissue were selected for tissue microarray construction. The protocol regarding tissue microarray was approved by the hospital's institutional review board. The tissue samples and data were anonymized and unlinked to an identifiable person. Waiver of informed consent was also approved by the institutional review board. The tumor tissue retrieved from the paraffin blocks contained at least two 3-mm cores. The patient overall survival rate was calculated from the date of operation to the date of death and was considered censored for patients who were alive at last follow-up. Time to recurrence was measured from the date of the operation to the date of recurrence and was considered censored for patients who were disease-free at the last follow-up or were deceased without evidence of disease recurrence. Relapse-free survival was measured from the date of the operation to the date of recurrence or death from any cause.[27]

Immunohistochemistry staining and scoring

The specimen processing and IHC procedures were carried out as previously described.[28] For TOPK staining, a mAb to TOPK (Cell Signaling Technology, Danvers, MA, USA) was used at the dilution of 1:100 and incubated at 4°C overnight. The sections were then incubated with a biotinylated secondary antibody for 10 min. Streptavidin–HRP conjugate with 3-amino-9-ethylcarbazole (Dako, Carpinteria, CA, USA) was used as the chromogen. All slides were counterstained with hematoxylin.

The intensity of immunoreactivity was graded according to the following scale: 0, negative; 1, weakly positive; 2, moderately positive; and 3, strongly positive. The percentage score was semiquantitatively assessed by the percentage of positive-stained cells: 0, 0%; 1, ≤10%; 2, 11–50%; and 3, 51–100%. The IHC score of each specimen was represented by the value of the intensity score multiplied by the percentage score, which ranged from 0 to 9, according to the method described in a previous report.[29]

Statistical analysis

Three microarray datasets were analyzed by two-tailed t-test and TOPK/PBK was selected to compare the expression in stage I versus stage III/IV, and stage II versus stage III/IV lung adenocarcinoma. The chi-squared-test was used to assess the association between clinicopathological parameters and the TOPK IHC score. Survival curves were plotted using the Kaplan–Meier method and were compared using the log–rank test. Multivariate analyses were carried out by means of the Cox proportional hazards model using spss software (version 18.0; SPSS, Chicago, IL, USA). P-values <0.05 were considered to be significant.

Transfection

For transfection, 50% confluent cells were incubated with 1 μg DNA and lipofetamine reagent (1:6) and mixed according to the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA). Twenty-four hours post-transfection, cells were harvested for detection of the exogenously expressed TOPK or seeded onto Transwell inserts for examining migration/invasion ability. For establishing stable cell lines with overexpression or knockdown of TOPK, 5 × 104 cells were seeded into 6-cm tissue culture plates containing complete growth medium supplemented with G418 (500 μg/mL) at 48 h post-transfection and selected for approximately 3 weeks. SureSilencing shRNA plasmids specific for TOPK (TGACCCTGAGGCTTGTTACAT or TGTGGGAAATGATGACTTTAT) or negative control (GGAATCTCATTCGATGCATAC) were purchased from SABiosciences (Frederick, MD, USA). The expression of TOPK and β-tubulin was determined by Western blot analysis using anti-TOPK antibodies (1:1000; BD Biosciences, Bedford, MA, USA) and anti-β-tubulin antibodies (1:3000; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Cell viability was determined using an MTT assay.

Migration/invasion and clonogenic assay

Cells (1 × 104) were seeded onto Matrigel-coated or uncoated Transwell inserts, and the same medium was added to the lower chamber. The cells were allowed to migrate/invade for 24 h, then fixed in pre-cold methanol and stained with Giemsa (Sigma-Aldrich, St Louis, MO, USA). Cells that remained adherent to the underside of the membrane were counted. The diagrams show the mean values from at least three independent experiments and represented as relative cell migration/invasion fold compared to vehicle cells. For clonogenic assay, cells were seeded in six-well plates with 500 cells/well and cultured for 10 days. Cells were then washed with PBS and the colonies were fixed (acetic acid : methanol, 1:3) and stained with 0.5% crystal violet in methanol.

Results

To systematically investigate the gene expression pattern of TOPK/PBK in NSCLC, we used microarray datasets from three cohorts of patients with NSCLC obtained online from GEO. Figure S1 shows the box plot analysis of TOPK/PBK (Affymetrix probe set ID: 219148_at) gene expression in different stages of lung cancer from various studies. In general, TOPK/PBK is overexpressed in most of the lung cancer tissue samples, as illustrated by the dark gray dashed line in Figure S1, which indicates a twofold change. In addition, higher expression of TOPK/PBK was found to correlate with late-stage lung adenocarcinoma in microarray dataset 3 (stage III/IV vs. stage I, P = 0.013; stage III/IV vs. stage II, P = 0.037) (Fig. 1). Consistent with its increased expression in late-stage lung adenocarcinoma, we recently found TOPK/PBK was also significantly upregulated in 225 secondary metastatic tumors as compared with its expression in 30 benign tumors, indicating it is a potential metastatic biomarker.[22]

Figure 1.

T-LAK cell-originated protein kinase/PDZ-binding kinase (TOPK/PBK) is overexpressed in lung cancer from microarray studies. Box plots show the fold change (tumor/adjacent normal) of TOPK/PBK. Gene expression data were obtained from published lung cancer microarray studies. The x axis denotes dataset source; the y axis denotes the fold change of TOPK/PBK (Affymetrix probe set ID: 219148_at). The dark gray dashed line shows a twofold change in TOPK/PBK expression for each box plot.

In order to gain the mechanistic insights regarding the role of TOPK in cell migration and invasion, we transiently overexpressed TOPK-WT or TOPK-T9E, which is the phosphorylation-mimic form of TOPK and serves as an active form of TOPK in mitosis, in H1299 cells. The result showed that TOPK could promote cell migration/invasion rather than cell proliferation (Fig. 2A). Next, we established stable cell clones with overexpression or knockdown of TOPK to further elucidate the role of TOPK in cell migration/invasion. The results showed TOPK-overexpressed CL1-0 cells displayed a similar ability to enhance cell migration/invasion compared with that in mock-transfected cells (Fig. 2B). Inverse effects were also found in TOPK-knockdown A549 cells (Fig. 2C). Moreover, knockdown of TOPK significantly affected colony formation (Fig. 2D). Taken together, these results showed the role of TOPK in promoting cell migration, invasion, and colony formation in lung cancer cells.

Figure 2.

T-LAK cell-originated protein kinase (TOPK) can promote cell migration, invasion, and clonogenic activity in lung cancer cells. (A) H1299 cells were transiently transfected with vehicle (Veh.) or TOPK expression plasmids as indicated. WT, wild type. (B) Stable clones overexpressing TOPK were established. Cells from (A) and (B) were subjected to detection of TOPK and β-tubulin by immunoblotting or seeded onto Matrigel-coated or uncoated Transwell inserts and 96-well plates to examine migration/invasion and cell growth abilities, respectively. After 24 h, the numbers of migrated (middle panels) and invaded (lower panels) cells were counted and expressed (n = 3) and the growth of cells was quantified by MTT assay. Stable A549 clones with knockdown of TOPK were established and subjected to immunoblotting and migration assay as described above (C). The clonogenic activity of these clones was also determined (D). *P < 0.05; **P < 0.01; ***P < 0.001.

To evaluate the prognostic value of TOPK expression in resected stage I lung adenocarcinoma, a retrospective analysis of 203 patients diagnosed with pathological stage I lung adenocarcinoma was carried out to examine the expression of TOPK by IHC. The demographics of the 203 stage I lung adenocarcinoma patients are summarized in Table 1. The mean follow-up time was 62.5 months. During the follow-up period, 60 (29.6%) patients developed tumor recurrence (local recurrence in 20 patients, distant metastasis in 50 patients).

Table 1. Association of T-LAK cell-originated protein kinase (TOPK) immunohistochemistry (IHC) score and clinicopathological characteristics in 203 patients diagnosed with pathological stage I lung adenocarcinoma
ParametersTOPK IHC scoreP-value
≤3 (n = 67)>3 (n = 136)
  1. P-value for age, follow-up time, and tumor size were derived from a two-tailed Student's t-test; other P-values were derived from a two-tailed Pearson's chi-squared-test. NS, not significant.

Age, years
Mean (±SD)65.5 (±10.7)65.2 (±11.1)NS (0.872)
Range29–8734–88
Sex
Male43 (64.2%)79 (58.1%)NS (0.405)
Female24 (35.8%)57 (41.9%)
Smoking status
Non-smoker30 (50.8%)63 (51.6%)NS (0.920)
Smoker29 (49.2%)59 (48.4%)
Follow-up time, months
Mean (±SD)78.8 (±28.1)54.4 (±29.9)<0.001
Range0.37–126.000.23–125.47
Recurrence status
No recurrence54 (80.6%)89 (65.4%)0.026
Recurrence13 (19.4%)47 (34.6%)
Tumor size, cm
Mean (±SD)3.09 (±1.18)3.00 (±1.03)NS (0.581)
Range1–51–5
Tumor differentiation
Well to moderate47 (70.1%)92 (67.6%)NS (0.718)
Poor20 (29.9%)44 (32.4%)
Tumor necrosis
No41 (61.2%)68 (50.0%)NS (0.133)
Yes26 (38.8%)68 (50.0%)
Angiolymphatic invasion
No54 (80.6%)86 (63.2%)0.012
Yes13 (19.4%)50 (36.8%)

T-LAK cell-originated protein kinase was detected in the cytoplasm of tumor cells by IHC. The intensity of immunoreactivity ranged from 0 to 3 (Fig. 3) and the IHC score ranged from 0 to 9, based on the scoring method described in a previous report.[29] Among the 203 patients, the TOPK IHC score was distributed as follows: 0, 30 patients; 1, 14 patients; 2, 15 patients; 3, 8 patients; 4, 14 patients; 6, 69 patients; and 9, 53 patients. Sixty-seven patients (33.0%) had a TOPK IHC score of less than or equal to 3, and the remaining 136 patients (67.0%) had a TOPK IHC score greater than 3. A TOPK IHC score >3 was considered as TOPK overexpression.

Figure 3.

T-LAK cell-originated protein kinase expression in stage I lung adenocarcinoma. The intensity of immunoreactivity was recorded as: 0, negative (A); 1, weakly positive (B); 2, moderately positive (C); or 3, strongly positive (D).

We investigated the association of various clinicopathological parameters with samples having either a TOPK IHC score ≤3 or a TOPK IHC score >3. The mean follow-up time was significantly shorter in patients with a TOPK IHC score >3 (P < 0.001). Disease recurrence and angiolymphatic invasion were more frequent in patients with a TOPK IHC score >3 (P = 0.026 and 0.012, respectively). There were no significant associations of TOPK expression with any of the other clinicopathological parameters (Table 1).

We further analyzed the prognostic values of age, sex, smoking status, tumor size, pathological tumor differentiation, tumor necrosis, angiolymphatic invasion, and TOPK IHC score on the overall survival rate of the patients. The overall survival rate was significantly worse in patients older than 65 years (P = 0.002), with a tumor size >3 cm (P = 0.032), with the presence of tumor necrosis (P = 0.005), with the presence of angiolymphatic invasion (P = 0.003), and with a TOPK IHC score >3 (P < 0.001) (Table 2). The Kaplan–Meier curves for patients with TOPK IHC scores ≤3 and >3 are shown in Figure 4. Multivariate Cox regression analysis revealed that age (>65 years old) (hazard ratio [HR] = 2.13; 95% confidence interval [CI], 1.31~3.47, P = 0.002), and TOPK IHC score (>3) (HR = 4.05; 95% CI, 2.24~7.03; P < 0.001) were independent predictors of overall survival (Table 3).

Figure 4.

Overall patient survival rate based on T-LAK cell-originated protein kinase (TOPK) immunohistochemistry (IHC) scores. Patients with high TOPK IHC scores (>3) had worse survival rates compared to those with low TOPK IHC scores (≤3) (P < 0.001).

Table 2. Five-year overall survival rates stratified by various clinicopathologic parameters in 203 patients diagnosed with pathological stage I lung adenocarcinoma
Clinicopathologic parameters5-year survival rate ± SE (%)Log–rank P-value
  1. IHC, immunohistochemistry; NS, not significant; TOPK, T-LAK cell-originated protein kinase.

Age, years
≤6574.2 ± 4.90.002
>6555.0 ± 4.6
Sex
Male59.8 ± 4.4NS (0.404)
Female67.3 ± 5.4
Smoking status
Non-smoker66.3 ± 4.9NS (0.300)
Smoker56.8 ± 5.3
Tumor size, cm
≤367.4 ± 4.30.032
>355.0 ± 5.8
Tumor differentiation
Well to moderate66.0 ± 4.1NS (0.189)
Poor55.3 ± 6.3
Tumor necrosis
Yes53.3 ± 5.20.005
No70.7 ± 4.4
Angiolymphatic invasion
Yes47.1 ± 6.30.003
No69.9 ± 4.0
TOPK IHC score
≤385.0 ± 4.4<0.001
>351.4 ± 4.4
Table 3. Multivariate analysis for overall survival in 203 patients diagnosed with pathological stage I lung adenocarcinoma
Clinicopathologic parametersHazard ratio95% confidence intervalP-value
  1. IHC, immunohistochemistry; NS, not significant; TOPK, T-LAK cell-originated protein kinase.

Age >65 years2.131.31–3.470.002
Female0.920.51–1.66NS (0.793)
Smoking history0.980.55–1.77NS (0.965)
Tumor size >3 cm1.360.87–2.12NS (0.167)
Poor differentiation1.200.73–1.96NS (0.453)
Necrosis1.050.63–1.77NS (0.833)
Angiolymphatic invasion1.500.95–2.39NS (0.080)
TOPK IHC score >34.052.24–7.03<0.001

We further analyzed the prognostic factors correlated with time to recurrence. Tumor size (>3 cm) (P = 0.016), the presence of tumor necrosis (P = 0.004), and TOPK IHC score (>3) (P = 0.010) were associated with shortened time to recurrence (Table 4). The Kaplan–Meier curves for time to recurrence in patients with TOPK IHC scores ≤3 and >3 are shown in Figure 5. A TOPK IHC score >3 (HR = 2.85, 95% CI, 1.36~5.96, P = 0.005) remained an independent predictor of shortened time to recurrence after analyses by multivariate Cox regression (Table 5). In addition, we also analyzed the prognostic impact of TOPK expression on relapse-free survival. TOPK IHC score >3 was also strongly associated with poor relapse-free survival (Fig. S2). To strengthen the clinical impact of TOPK, we managed to obtain an additional 21 stage 1 lung adenocarcinoma patient specimens from Kaohsiung Medical University Chung-Ho Memorial Hospital (Kaohsiung, Taiwan) as an independent cohort. The Kaplan–Meier survival curves showed that TOPK expression was correlated with poor outcome. Even though the difference did not reach statistical significance, possibly due to the limited sample size, the expression of TOPK was clearly associated with reduced time to recurrence in the patients. We believe that TOPK can serve as a significant prognostic predictor of shortened overall survival and time to recurrence (Fig. S3).

Figure 5.

Time to recurrence based on T-LAK cell-originated protein kinase (TOPK) immunohistochemistry (IHC) scores. Patients with high TOPK IHC scores (>3) have shortened time to recurrence compared with those with low TOPK IHC scores (≤3) (P = 0.010).

Table 4. Five-year recurrence-free rates stratified by various clinicopathologic parameters in 203 patients diagnosed with pathological stage I lung adenocarcinoma
Clinicopathologic parameters5-year recurrence-free rate ± SE (%)Log–rank P-value
  1. IHC, immunohistochemistry; NS, not significant; TOPK, T-LAK cell-originated protein kinase.

Age, years
≤6574.7 ± 4.8NS (0.142)
>6563.6 ± 4.7
Sex
Male69.1 ± 4.4NS (0.854)
Female67.2 ± 5.4
Smoking status
Non-smoker68.3 ± 5.0NS (0.894)
Smoker69.3 ± 5.1
Tumor size, cm
≤374.5 ± 4.00.016
>358.5 ± 5.9
Tumor differentiation
Well to moderate71.6 ± 4.0NS (0.076)
Poor61.4 ± 6.3
Tumor necrosis
Yes58.9 ± 5.30.004
No76.5 ± 4.2
Angiolymphatic invasion
Yes59.6 ± 6.4NS (0.065)
No72.4 ± 3.9
TOPK IHC score
≤380.2 ± 4.90.010
>362.3 ± 4.4
Table 5. Multivariate analysis for time to recurrence in 203 patients diagnosed with pathological stage I lung adenocarcinoma
Clinicopathologic parametersHazard ratio95% confidence intervalP-value
  1. IHC, immunohistochemistry; NS, not significant; TOPK, T-LAK cell-originated protein kinase.

Age >65 years1.630.88–3.02NS (0.116)
Female1.140.56–2.32NS (0.710)
Smoking history0.830.40–1.74NS (0.639)
Tumor size >3 cm1.460.82–2.58NS (0.189)
Poor differentiation1.380.75–2.56NS (0.298)
Necrosis1.230.64–2.37NS (0.528)
Angiolymphatic invasion1.290.72–2.29NS (0.380)
TOPK IHC score >32.851.36–5.960.005

Discussion

Patients with stage I NSCLC have significant recurrence rates and lower than expected survival rates after surgical resection, indicating that our current staging methods do not adequately predict outcome. Although adjuvant chemotherapy after surgical resection has been shown to improve survival in stage II or IIIA patients, disagreement prevails over whether the benefit is recognized in stage I patients. Thus, it is important to identify new molecular factors for the prediction of prognosis to help select patients who may have subclinical micrometastatic disease and who might benefit from adjuvant treatment. In this cohort, we investigated the prognostic value of TOPK expression in patients with resected stage I lung adenocarcinoma. Older age and TOPK overexpression were significant independent prognostic indicators for overall survival rates by multivariate analysis. Overexpression of TOPK was also a significant prognostic factor for shortened time to recurrence by multivariate analysis, and was associated with significantly worse overall survival and shortened time to recurrence in our study.

A substantial number of studies have tried to identify poor prognostic factors in patients with stage I NSCLC for adjuvant therapy.[9-11] In addition, several clinical trials recently discovered that histological subgroups had different outcomes as a result of targeted therapy and newer chemotherapy regimens.[30-33] Therefore, it is important to identify higher risk early-stage lung adenocarcinoma patients who may benefit from histology-based treatment.

Based on many microarray studies in NSCLC, gene expression profiles have been used to predict patient survival rates in early-stage NSCLC. For example, Beer et al.[34] reported a combination of 50 genes as a risk index for prediction of overall survival rates in stage I adenocarcinoma. Lu et al.[35] reported a combination of 64 genes for prediction of overall survival rates in stage I NSCLC. Chen et al.[36] reported that a combination of DUSP6, MMD, STAT1, ERBB3, and LCK as a five-gene signature was closely associated with improved overall and relapse-free survival in patients with stage I and II NSCLC. Lau et al.[37] identified a three-gene classifier (STX1A, HIF1a, and CCR7) that predicted different prognoses for early-stage I NSCLC. Bianchi et al.[38] proposed a 10-gene model (E2F1, E2F4, HOXB7, HSPG2, MCM6, NUDCD1, RRM2, SERPINB5, SF3B1, and SCGB3A1) for predicting overall survival rates in stage I adenocarcinoma.

In addition to gene expression signatures, various biomarkers of protein expression in lung adenocarcinoma have also been proposed in published reports. In a report by Kojima et al.,[39] vascular endothelial growth factor-C and vascular endothelial growth factor receptor 3 were poor prognostic factors in patients with T1 lung adenocarcinoma. Chen et al.[40] showed that trophinin, which could enhance cell invasion, was a poor prognostic factor for stage I lung adenocarcinoma. Hung et al.[28] proposed a model using a combination of HIF-1α, Twist1, and Snail to predict poor overall survival rates in early-stage lung adenocarcinoma. Seki et al.[41] reported that high eIF4E and low 4E-BP1 were unfavorable prognostic factors in patients with pathological stage I lung adenocarcinoma. Thyroid transcription factor-1 expression was a predictor of better overall survival in patients with lung adenocarcinoma.[42] High expression of ribonucleotide reductase regulatory subunit M1 (RRM1) protein was associated with increased disease-free and overall survival in patients with early-stage NSCLC.[43] However, high RRM1 protein expression correlated with reduced response rates in patients receiving gemcitabine-based regimens.[44, 45] Recently, Liu et al. identified annexin A1 as a pro-invasive and prognostic factor for lung adenocarcinoma.[46] Recently, we identified TOPK as a metastasis-associated protein kinase by its upregulation in secondary metastatic tumors and in brain metastasis from lung tumor.[22] The findings of the present study further establish TOPK as a new prognostic marker for stage I lung adenocarcinoma.

Lung adenocarcinoma patients in Asia often have the epidermal growth factor receptor (EGFR) mutation, therefore, we sequenced 67 lung adenocarcinoma samples in this study to explore the correlation between TOPK expression and EGFR mutation status. Among these 67 patients, 27 patients had the EGFR mutation, including 18 cases with the L858R mutation, eight cases with exon 19 deletion, and one case with the G719S mutation (Table S1). There was no difference in overall survival between the patients with and without EGFR mutation (5-year overall survival rate, 63% vs. 55%; P = 0.628) (Fig. S4). In addition, there was no difference in time to recurrence between the patients with and without the EGFR mutation (5-year recurrence-free rate, 57.4% vs. 71.0%; P = 0.192) (Fig. S5). Finally, TOPK expression was not correlated with positive EGFR mutation status (P = 0.564, chi-squared-test).

In conclusion, our results indicate that overexpression of TOPK is a potential prognostic predictor in stage I lung adenocarcinoma. The expression of TOPK in resected adenocarcinoma of lung could help to identify stage I patients who are at risk for recurrent disease, and may provide a valuable tool in selecting patients for adjuvant treatment. Nevertheless, our finding needs to be validated in a prospective study.

Acknowledgments

We thank Dr. Chi-Ying F. Huang (Yang-Ming University, Taipei, Taiwan) for providing lung cancer microarray data and critical reading of this manuscript. This work was supported by grants from the National Science Council (Taiwan) (NSC95-2311-B-030-002-MY3, NSC99-2627-B-030-001 and NSC100-2627-B-030-001) awarded to J.M. Lai, and from the Center of Excellence for Cancer Research at Taipei Veterans General Hospital (DOH100-TD-C-111-007) awarded to T.Y. Chou and Y.C. Wu.

Disclosure Statement

The authors have no conflicts of interest.

Ancillary