Increased high-mobility group A2 correlates with lymph node metastasis and prognosis of non-small cell lung cancer

Abstract

BACKGROUND:

High-mobility group A2 (HMGA2) has been investigated to be associated with tumorigenesis; however, the expression pattern and clinical significance of HMGA2 in non-small cell lung cancer (NSCLC) remains poorly understood. The purpose of this study is to examine the expression of HMGA2 and to analyze its relationships with respect to clinico-pathological features and patient survival in NSCLC.

METHODS:

The expression level of HMGA2 was examined by Western blot and immunohistochemistry in NSCLC cells and tissues. The relationship between HMGA2 expression and survival of NSCLC patients was calculated by a Kaplan-Meier method and the evaluation of risk factor was determined by the multiple regression analysis.

RESULTS:

NSCLC tissues exhibited a higher expression level of HMGA2 compared to normal tissues ( $p<$ 0.05) and the expression level of HMGA2 was significantly associated with poor differentiation of NSCLC ( $p<$ 0.05), lymph node metastasis ( $p<$ 0.05) and advanced clinical stage ( $p<$ 0.05). Besides, HMGA2 was also confirmed to be elevated in NSCLC cells by Western blot. Moreover, increased expression of HMGA2 correlated with decreased survival of NSCLC patients ( $p<$ 0.05).

CONCLUSIONS:

HMGA2 was highly expressed in NSCLC tissues and cells and its overexpression was correlated with low-grade differentiation, lymph node metastasis, advanced clinical stage and poor survival time of NSCLC, which suggested that it could serve as a potential molecular marker and prognostic index for NSCLC.

1. Introduction

The incidence of lung cancer has significantly increased over the last century in China, largely due to environmental pollution and unhealthy lifestyles. Now lung cancer is the leading cause of cancer-related death in this country, with 733,000 expected deaths in China in 2015 and accounts for 28% of all cancer-related deaths among cancer patients in China [1]. Especially, morbidity and mortality of non-small-cell lung cancer (NSCLC) are relatively high, with a 5-year survival rate of less than 20%, which is often due to lung cancer first presenting at late stages and a lack of curative therapeutic options at these later stages. Over the past decade, the discovery of driver mutations has changed the landscape for the treatment of NSCLC [2]. In addition, the findings of abnormal molecule, gene mutation, and genetic phenotype have been suggested to improve the diagnostic accuracy of lung cancer at the early stage, such markers often guide patient treatment options and confer valuable prognostic information [3].

Recently, emerging studies reported that high- mobility group A2 (HMGA2) was correlated with many human malignancies and participated in different cellular activities, including cell cycle regulation, differentiation senescence [4]. The HMGA2 is coded by a gene on chromosome 12q14-15, which is characterized by its ability to bind to nucleotide sites rich in adenine and thymine [5]. Previous study show that HMGA2, the non-histone chromatin binding protein, plays important roles in the repair and protection of genomic DNA in embryonic stem cells and cancer cells [6]. Although over-expression of HMGA2 has been found in many kinds of malignant tumors, including colon, breast, pancreatic, and esophageal cancer; and the research results suggested that the up-regulation of HMGA2 is involved in the malignant phenotype of these cancers, which leads to poor prognosis for these patients with malignant tumors [3, 4]. The data on the natural expression status of the HMGA2 and its relationship with clinicopathological parameters of NSCLC patients remains insufficient. This study was to investigate the significance of HMGA2 expression in NSCLC and its clinical significance for NSCLC diagnosis and prognosis.

2. Material and methods

2.1 Cell lines of NSCLC

Human NSCLC lines NCI-A549 and NCI-H520 were obtained from the Cell Biology Department of Medical School, Xuzhou Medical University, China. Sixteen HBE cell line was purchased from the Tumor Institution, Academy of Chinese Medical Sciences, Beijing, China. The cell lines were treated strictly complying with a process previously described [7], the cells were cultured in RPMI 1640 medium with 10% FBS, 100 units/ml penicillin and 100 $\mu$ g/ml streptomycin (Invitrogen) and maintained at 37 ${}^{\circ }$ C in a humidified 5% CO2, 95% air incubator. We extracted the protein from the cells at the logarithmic phase to detect the HMGA2 expression by Western blot.

2.2 Patients selection

From January 2009 to December 2012, we collected a total 80 surgical tissue specimens from NSCLC patients who accepted surgical resection at the Lianyungang First People’s Hospital, Lianyungang, China. The diagnosis of lung cancer was confirmed by pathologists under the optical microscope, the type of pathology of some patients were determined with immunohistochemical pathology. These patients were not treated with chemotherapy or radiotherapy before surgery. Among these patients received surgery, 77 cases were patients with stage II–III, and only 3 patients belonged to stage IV. Forty adjacent normal lung tissues from NSCLC patients were selected as normal control. After removing the tumor mass via surgery, we carefully separated the cancer-adjacent normal tissues from the cancer tissues. The distance between the edge of cancer-adjacent normal tissues and tumor tissues was greater than 3 cm to avoid contamination of cancer cells. The individuals included in this project were classified according to different clinical features (Table 1). All patients were included in the follow-up program for collecting the data of survival. Definition of survival time: from the date of taking tissues samples (the time of surgery) to the date of patients’ death (or the last date lost to follow-up). In our study, the median follow-up was 36.7 months. This project was approved by the Institutional Review Board (Lianyungang First People’s Hospital, Lianyungang, China), and written informed consent was obtained from all subjects.

Table 1

Clinico-pathological features of lung cancer patients (N $=$ 80)

Group	Characteristics	Number (%)
Sex
	Male	47	(58.8%)
	Female	33	(41.2%)
Age
	$<$ 60	41	(51.3%)
	$⩾60$	39	(48.7%)
Smoking
	Severe	39	(48.8%)
	Moderate	0	(0%)
	Slight	8	(10%)
	Never	33	(41.2%)
Histology
	LAC	35	(39.8%)
	LSCC	45	(56.2%)
Pathologic grade
	Well differentiated	22	(27.5%)
	Moderately differentiated	24	(30%)
	Poorly-differentiated	34	(42.5%)
Clinical staging
	I–II	50	(62.5%)
	III–IV	30	(37.5%)
Lymphatic invasion
	Positive	46	(57.5%)
	Negative	34	(42.5%)

Severe, $>$ 40 pack years; Moderate, 20.1–40 pack years; Slight, 0.1–20 pack years; LAC, adenocarcinoma of the lung; LSCC, squamous cell carcinoma of the lung.

2.3 Tissue microarray (TMA) construction

To produce the tissue microarray, pathologists from Shaanxi Chaoying Biotechnology Company (Li Jun and Wang Mei) selected morphologically representative cancer areas. The TMA was constructed strictly complying with a process previously described with an instrument from Beecher Instruments, Manual Tissue Arrayer, USA. Briefly, they removed a tissue core from the donor block with an inner diameter of approximately 2.0 mm and the each core sample from lung cancer tissue was placed into a recipient block according to a strict order. The array block constructed was sectioned and leveled on the microscope slide and then baked in an oven and finally tested with routine H&E staining and immunohistochemistry (IHC).

2.4 Immunohistochemistry (IHC)

The expression status of HMGA2 in NSCLC tissues was tested by IHC method (Kit of Ultra-sensitive S-P, produced by Zhongshan Jinqiao, Beijing, China), which was strictly performed according to a previous method that was described in published article [3]. The best dilution of anti-human HMGA2 antibody is 1:50 (Shanghai Changdao biological technology company). The positive staining slice of colon cancer served as a positive control; PBS took place of the first antibody as the negative control. The immunostaining of HMGA2 was assessed according to a previous method described [3]. Ten high-magnification fields of each tissue section were randomly selected ( $>$ 1,000 cells) for calculating the expression scores, according to the area of staining and intensity of staining. The criterion of intensity: 0, no staining; 1+, mild staining; 2+, moderate staining; 3+, intense staining. The criterion of area: 0, no staining of cells in any microscopic fields; 1+, $<$ 30% of tissue stained positive; 2+, between 30% and 60% stained positive; 3+, $>$ 60% stained positive. The combined score (area of staining + intensity of staining) of $⩽$ 2 was considered as low expression, between 3 and 4 as moderate expression and between 5 and 6 served as high expression [3].

2.5 Western blot

Western blot was employed to disclose the expression status of HMGA2 in NSCLC cells according to a standard method reported in a previous study [7]. The harvested NSCLC cells were washed once with PBS, lysed with 2 $\times$ sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) sample buffer (20 mM Tris, pH 8.0, 2% SDS, 2 mM dithiothreitol, 1 mM Na3VO4, 2 mM EDTA, and 20% glycerol), and boiled for 5 min. The protein concentrations were quantitated by Bradford assay. A total of 30 $\mu$ g proteins from each sample were loaded on the SDS-PAGE gel (10%) and electrophoretically separated. The proteins were transferred to polyvinylidene difluoride membranes. The membranes were blocked for 2 h at 37 ${}^{\circ }$ C in 20 mM Tris, pH 8.0, 150 mM NaCl, and 0.05% Tween 20 (TBST) containing either 5% BSA or 5% nonfat dried milk. The membranes were incubated with HMGA2 antibody (dilute concentration of HMGA2, 1:150 dilution; Shanghai Changdao biological technology company) overnight at 4 ${}^{\circ }$ C. We used $\beta$ -actin (Santa Cruz Biotechnology, Inc. USA) as the internal control. The primary antibodies were detected using horseradish peroxidase-conjugated secondary antibodies, and after three washes with TBST, positive protein bands were visualized using enhanced chemiluminescence. The gray value of each band was analyzed by Image J analysis software, and the ratio of HMGA2 to the corresponding $\beta$ -actin gray value was calculated. All experiments were performed for three separate times.

2.6 Statistical analysis

The SPSS 19.0 software package was used to perform the statistical analysis (SPSS Institute, version 19.0, Chicago, USA). The expression difference of HMGA2 and its relationships with clinico-pathological features of NSCLC was analyzed using the ${\chi }^2$ , Fisher’s exact. The overall survival was measured from the date of surgery to the date of death from any cause or the date on which the patient was last known to be alive. Survival curves were plotted according to the Kaplan-Meier method. A Cox proportional hazard regression model for multivariate analysis was used to test the confounding effect of the variables that are most closely associated with the expression level of the different expression of HMGA2. All tests were two-sided, and $p$ -values $<$ 0.05 were considered to be statistically significant.

Figure 1.

Protein expression of HMGA2 in NSCLC tissues. (A) Low expression of HMGA2 in adjacent normal lung tissues ( $\times$ 400). (B) High expression of HMGA2 in adjacent normal lung tissues ( $\times$ 400). (C) Low expression of HMGA2 in well differentiated LSCC ( $\times$ 400). (D) Low expression of HMGA2 in well differentiated LAC ( $\times$ 400). (E) Moderate expression of HMGA2 in moderately differentiated LSCC ( $\times$ 400). (F) Moderate expression of HMGA2 in moderately differentiated LAC ( $\times$ 400). (G) High expression of HMGA2 in poorly differentiated LSCC ( $\times$ 400). (H) High expression of HMGA2 in poorly differentiated LAC. LAC, adenocarcinoma of the lung; LSCC, squamous cell carcinoma of the lung; NSCLC, non-small cell lung cancer.

Table 2

Correlation between clinico-pathological features and the expression of HMGA2 in NSCLC

Parameter	Group	N	Expression of HMGA2
			Low (%)		Moderate (%)		High (%)		${\chi }^2$ value	$P$ value
Resources
	Normal	40	21	(52.5)	12	(30)	7	(17.5)	15.635	$P<$ 0.001
	Cancerous	80	18	(22.5)	20	(25)	42	(52.5)
Gender
	Male	47	13	(27.7)	12	(25.5)	22	(46.8)	2.064	$P=$ 0.356
	Female	33	5	(15.2)	8	(24.2)	20	(60.6)
Ages
	$<$ 60	41	6	(14.6)	11	(26.8)	24	(58.5)	3.009	$P=$ 0.222
	$⩾$ 60	39	12	(30.8)	9	(23.1)	18	(46.2)
Smoking
	0	33	6	(18.2)	7	(21.2)	20	( 60.6)	4.568	$P=$ 0.335
	0.1–40	8	1	(12.5)	1	(12.5)	6	(75)
	$>$ 40	39	11	(28.2)	12	(30.8)	16	(41)
Histology
	LAC	35	11	(24.4)	11	(24.4)	23	(51.1)	0.223	$P=$ 0.893
	LSCC	45	7	(20.7)	9	(25.7)	19	(54.3)
Pathological grade
	Well	22	6	(27.3)	4	(18.2)	12	(54.5)	21.616	$P<$ 0.001
	Moderately	24	8	(33.3)	12	(50)	4	(16.7)
	Poorly	34	4	(11.8)	4	(11.8)	26	(76.5) ${}^{*}$
Lymphatic invasion
	Positive	46	2	(4.3)	10	(21.7)	34	(73.9)	25.765	$P<$ 0.001
	Negative	34	16	(47.1)	10	(29.4)	8	(23.5)
pTNM
	I–II	50	17	(34)	17	(34)	16	(32)	22.830	$P<$ 0.001
	III–IV	30	1	(3.3)	3	(10)	26	(86.7)

${}^{*}p<$ 0.001, poorly compared with moderately and well differentiation, respectively; LAC, lung adenocarcinoma; LSCC, lung squamous cell carcinoma.

Figure 2.

Correlation of expression of HMGA2 with clinicopathologic factors of NSCLC patients and its expression in different cell lines. (A) Increased HMGA2 in lung cancer tissues compared with the normal tissues ( $p<$ 0.05). (B) Increased HMGA2 was observed in poorly differentiated NSCLC compared with the moderately and well differentiated NSCLC ( $p<$ 0.05). (C) Lymph node metastasis-positive tissues revealed an increased level of HMGA2 compared with the lymph node metastasis-negative tissues ( $p<$ 0.05). (D) Tissues at staged III–IV had higher HMGA2 expression than those at stages I–II ( $p<$ 0.05). pTNM (pathologic TNM classification); NSCLC, non-small cell lung cancer.

Table 3

Cox proportional hazards regression model analysis of overall survival

Variables (X)	Categories (different groups)	P value	OR value	95% CI for OR
				lower	upper
Gender (X1)	male (X ${}_{1-0}$ ) vs. female (X ${}_{1-1}$ )	0.698	–	–	–
Age (X2)	$<$ 60 (X ${}_{2-0}$ ) vs. $⩾$ 60 (X ${}_{2-1}$ )	0.778	–	–	–
Smoking (X3)	0 (X ${}_{3-0}$ ) vs. 0.1–40 (X ${}_{3{-}_1}$ ) vs. $>$ 40 (X ${}_{3-2}$ )	0.384	–	–	–
Histology (X4)	LAC (X ${}_{4-0}$ ) vs. LSCC (X ${}_{4-1}$ )	0.918	–	–	–
Differentiation (X5)	poor (X ${}_{5-0}$ ) vs. moderate (X ${}_{5-1}$ ) vs. well (X ${}_{5-2}$ )	0.129
Lymphatic invasion (X6)	positive (X ${}_{6-0}$ ) vs. negative (X ${}_{6-1}$ )	$<$ 0.001	3.003	2.076	4.343
TNM (X7)	I–II (X ${}_{7-0}$ ) vs. III–IV (X ${}_{7-1}$ )	$<$ 0.001	2.656	1.560	4.571
HMGA2 (X8)	low (X ${}_{8-0}$ ) vs. moderate (X ${}_{8-1}$ ) vs. high (X ${}_{8-2}$ )	$<$ 0.001	3.381	2.268	5.039
	Risk function:	H(t) $=$ [h ${}_0$ (t)]e ${}^{(1.099\times 6+0.976\times 7+1.218\times 8)}$

LAC, lung adenocarcinoma; LSCC, lung squamous cell carcinoma; Smoking, pack years of smoking. OR, odds ratio; CI, confidence interval.

Figure 3.

Expression of HMGA2 in different cell lines and the relationship between its expression and prognosis of patients with NSCLC. (A and B) Western blot analysis showed that the expression of HMGA2 was higher in A549 and H520 cells than in 16HBE cells ( $p<$ 0.05). (C) Scatter diagram showed that higher expression of HMGA2 correlated with early emergence of death in non-small cell lung cancer ( $p<$ 0.001). (D) Survival curve showed that higher expression of HMGA2 was associated with a shorter post-surgery survival time ( $p<$ 0.001). NSCLC, non-small cell lung cancer.

3. Results

3.1 Comparison of HMGA2 expression in NSCLC tissues and cancer-adjacent normal tissues

HMGA2 protein was expressed in the cytoplasm of NSCLC cells, partially expressed in the nucleus, showing brown and yellow particles with different thickness (Fig. 1A–H). Simultaneously, a small amount of HMGA2 protein was also found in normal lung tissues, located in the cytoplasm and nuclei of cells. The statistical analysis showed that the high expression rate of HMGA2 in NSCLC tissues (42/80, 52.5%) was significantly higher than that in normal lung tissues (7/40, 17.5%), the difference was statistically significant ( $P<$ 0.001), which displayed that NSCLC tissues exhibited a higher expression of HMGA2.

3.2 Correlation of HMGA2 expression with clinico-pathologic factors of NSCLC patients

As shown in Fig. 2A–D, the expression of HMGA2 was correlated with some clinical parameters of NSCLC. Compared with moderately differentiated (4/24, 16.7%) and well-differentiated (12/22, 54.5%) NSCLC, poorly differentiated NSCLC showed a high expression of HMGA2 (26/34, 76.5%). In addition, the expression of HMGA2 in NSCLC tissues with lymph node metastasis (34/46, 73.9%) was significantly higher than that in NSCLC without lymph node metastasis (8/34, 23.5%). Furthermore, over-expression of HMGA2 was observed in NSCLC tissues in stage III–IV (26/30, 86.7%) compared to I–II NSCLC tissues (16/50, 32%). However, the expression of HMGA2 was not associated with gender, age, degree of smoking and histological type in patients with NSCLC ( $P>$ 0.05) (Table 2).

3.3 Expression pattern of HMGA2 in A549 and H520 cells

We performed western blot analysis to determine the expression pattern of HMGA2 in A549, H520 cells and 16HBE cells, and found that the expression of HMGA2 was higher in A549 and H520 cells than in 16HBE cells (Fig. 3A and B). Quantitative analysis showed that the level of HMGA2 expression in the A549 (0.92 $\pm$ 0.22) and H520 (0.86 $\pm$ 0.17) cells was higher than that in 16HBE (0.53 $\pm$ 0.23, $P<$ 0.05).

3.4 Correlation between HMGA2 expression and prognosis in patients with NSCLC

From the follow-up database, we obtained the survival data of NSCLC patients. Of the 80 patients, 71 had complete follow-up records. We sorted out the data and conducted a survival analysis. We drew Kaplan-Meier curves to disclose the correlation between HMGA2 expression and prognosis in patients with NSCLC. As shown in Fig. 3C and D, the results demonstrated that patients with high expression of HMGA2 had shorter survival time than those with moderate and low expression ( $p<$ 0.05), which means that patients with high expression of HMGA2 have a risk of shorter survival.

3.5 Multivariate statistical analysis of clinical parameters

When the survival time was defined as a dependent variable and a variety of clinical parameters were served as independent variables, we performed a Cox-regression analysis. The statistical results showed that the high expression of HMGA2 seemed to be an independent risk factor for the survival of patients with NSCLC ( $p<$ 0.001). Furthermore, lymph node metastasis and TNM stage were also found to be the certain risk factors that correlated with the shorter survival time ( $p<$ 0.001), which means that patients will have a shorter survival as long as they showed lymph node metastasis and advanced clinical stage. The final risk function was H(t) $=$ [h ${}_0$ (t)]e ${}^{(1.099\times 6+0.976\times 7+1.218X8)}$ (Table 3), expounding that lymph node metastasis, advanced clinical stage and the HMGA2 overexpression can shorten the survival time of NSCLC patients.

4. Discussion

The high mobility group A2 (HMGA2) is a member of the high mobility group AT-hook (HMGA) family. HMGA2 plays a critical role in several cellular biological processes, including cell transformation, growth, differentiation, senescence and cycle control [9]. The HMGA2 protein is an architectural transcription factor that has recently been implicated in the development and progression of malignant tumors [4]. HMGA2 over-expression has been detected in several human malignancies, in particular pancreatic [8], gastrointestinal tract [4], and ovarian cancer [9] representing a very useful biomarker of malignancy [10]. Furthermore, highly expression of HMGA2 has been correlated with cancer proliferation, increased metastasis and poor prognosis in multiple types of cancer [5, 8, 9, 10]. In the world, lung cancer continues to be one of the top five causes of cancer-related mortality, with NSCLC being the most common subtype [11]. A few of studies display that HMGA2 is correlated with lung cancer and down-regulation of HMGA2 seems to reduce the proliferation and mobility of lung cancer [10, 12]. However, the data on the natural expression status of the HMGA2 and its relationships with clinicopathological parameters and survival of lung cancer patients remain insufficient.

In this study, we found that HMGA2 protein was highly expressed in the cytoplasm of NSCLC cells, which implied that HMGA2 was likely to be implicated in the occurrence and development of NSCLC. Previous study indicates that HMGA2 is an oncogene important in the pathogenesis of human lung cancer and that targeting HMGA2 can be therapeutically beneficial in lung cancer and other cancers characterized by increased HMGA2 expression [13]. Comparing matching tumor and non-tumorous samples from the same NSCLC patient, HMGA2 is revealed a highly significant overexpression in lung cancer cases, and thus suggests that HMGA2 overexpression is a common event in NSCLC and could serve as molecular marker for lung cancer [12]. We also found that a small amount of HMGA2 protein was also expressed in normal lung tissue, and located in the cytoplasm, suggesting that the function of normal cells was also depended on the function of HMGA2 signal. In our study, the high expression rate of HMGA2 in NSCLC tissues was significantly higher than that in normal lung tissues, and in particular in poorly differentiated lung cancer. Thus, we can infer that higher HMGA2 expression is potentially correlated with the malignant biological behavior of NSCLC. Our results are consistent with numerous prior studies that HMGA2 is up-regulated both in cancer cells and tissues compared with the normal cells and tissues [4, 5, 6]. In addition, we noticed that the expression of HMGA2 was correlated with lymph node metastasis ( $P<$ 0.05) and advanced clinical stage ( $P<$ 0.05). It is well known that metastasis is a major feature of malignant tumors, and lymph node metastasis is a common way of metastasis in lung cancer, especially in NSCLC. In our study, this finding that high expression of HMGA2 was shown to be intimately involved in tumor cell lymph node invasion and advanced clinical stage is in concordance with previous reports showing that a high level of HMGA2 in invasive NSCLC is associated with a poor clinical outcome compared with patients who express lower levels of HMGA2 [4, 14], suggesting that increased expression of HMGA2 in the cytoplasm of the cancer cells may contribute to the progression of NSCLC.

In our study, Western blot was employed to further disclose the expression status of HMGA2 in different NSCLC cells. We found that the level of HMGA2 expression in the NSCLC cells was higher than that in normal bronchial epithelial cells. One newly study shows that HMGA2 is upregulated in lung adenocarcinoma (LAC) and associates with reduced overall survival in patients with LAC. HMGA2 knockdown suppresses the proliferation and motility of H1299 cells, while HMGA2 ectopic expression in A549 cells increases cell proliferation and migration [15]. As NSCLC is a highly malignant tumor, a clear understanding on the mechanism and action of HMGA2 in NSCLC cells will facilitate to disclose the biological mechanisms of NSCLC and help to improve the design of preventive treatment [16, 17]. Long-term follow-up analysis in our study indicated that high expression of HMGA2 showed a significant association with post-surgical survival time, suggesting that patients with high expression of HMGA2 had shorter survival than those with moderate and low expression of HMGA2. Subsequent Cox regression analysis further confirmed that high expression of HMGA2 might be an important risk factor (OR $=$ 3.381) for the post-surgical survival time of lung cancer. The statistical analysis showed that the final inducted variables in regression equation were HMGA2, lymphatic invasion and advanced clinical stage. The final risk function was H(t) $=$ [h ${}_0$ (t)]e ${}^{(1.099\times 6+0.976\times 7+1.218X8)}$ . We can see that lymphatic invasion and advanced clinical stage were also shown to be risks for lung cancer patients’ post-surgical survival time with OR values of 3.003 and 2.656, respectively, indicating that the presence of these clinical features predicted a shorter survival of the patient. Get together, HMGA2 has a potential to become part of an adverse prognostic index that serves to predict the post-surgical survival rate of patients with NSCLC. Previous study finds that high expression of HMGA2 at both mRNA and protein levels is detected in lung cancer and increased expression of HMGA2 correlates with poor survival and is significantly associated with cell proliferation [18]. Another study also shows that HMGA2 is overexpressed in the NSCLC and intimately correlates with disease stage and lymphonode metastasis [19]. More and more studies demonstrate that the up-regulation of HMGA2 is involved in the malignant phenotype of cancers, which leads to poor prognosis for these patients with malignant tumors [4, 5, 16, 20], our findings were similar with these previous studies, and also confirmed that HMGA2 was closely related to the occurrence and development of NSCLC. Taking those findings together, we deduced that the detection of HMGA2 should be served as a useful marker for NSCLC and may contribute to prognosis prediction of NSCLC. Assuredly, there are still many works before it could be used to conduct the diagnosis and prognosis of NSCLC in clinical practice.

Actually, there are several limitations in this study. First, it was carried out in single institute in east China, which may not represent the feature of the wider population. Second, the size of this study was relatively small and did not sufficiently assess difference of NSCLC level between lung cancers at early stages (I–II stage). Therefore, further investigations designing larger samples of patients with lung cancer are still required to get more solid and reliable conclusion.

5. Conclusion

In summary, our study demonstrated that HMGA2 was upregulated and positively associated with the poorly differentiated NSCLC, lymph node metastasis, advanced clinical stage and shorter survival time. The results suggested that the detection of HMGA2 may be served as a useful tumor marker for NSCLC diagnosis and could contribute to prognosis prediction of NSCLC.