Potentiality of forkhead box Q1 as a biomarker for monitoring tumor features and predicting prognosis in non-small cell lung cancer.

BACKGROUND: This study aimed to explore the correlation of forkhead box Q1 (FOXQ1) with clinicopathological features and survival profiles in patients with non-small cell lung cancer (NSCLC).
METHODS: A total of 238 NSCLC patients with TNM stage I-III who underwent surgical resection were reviewed, and the expression of FOXQ1 in tumor and paired adjacent tissue was detected using immunohistochemistry assays. The clinical data and survival data of patients with NSCLC were retrieved and calculated.
RESULTS: FOXQ1 expression was increased in tumor tissue (61.3% high expression and 38.7% low expression) compared with paired adjacent tissue (37.8% high expression and 62.2% low expression) (P < .001). In addition, high FOXQ1 expression was associated with larger tumor size (P = .042), lymph node metastasis (P = .040), and advanced TNM stage (P = .002). Disease-free survival (DFS) (P = .016) and overall survival (OS) (P = .008) were both reduced in patients with high FOXQ1 expression compared with patients with low FOXQ1 expression. Additionally, high FOXQ1 expression (P = .043), poor pathological differentiation (P = .003), and lymph node metastasis (P < .001) were independent risk factors for DFS, and high FOXQ1 expression (P = .021), tumor size (>5 cm) (P = .014), and lymph node metastasis (P < .001) were independent risk factors for OS.
CONCLUSION: High FOXQ1 expression is associated with advanced tumor features as well as undesirable survival profiles in patients with NSCLC, implying the potential prognostic value of FOXQ1 for NSCLC.

INTRODUCTION

Lung cancer, as the most prevalent cancer type, is responsible for 11.6% of all cancer cases and 18.4% of all cancer‐related mortality.1 Non‐small cell lung cancer (NSCLC) is the predominant type of lung cancer, accounting for approximately 85% of all lung cancer cases.2 Currently, common treatments, including surgical resection, chemotherapy, radiotherapy, and targeted therapy, are proposed to patients depending on the tumor stage as well as other critical concerns, such as cardiac and pulmonary functions.2 For early‐stage patients with NSCLC without surgical contraindications, surgical resection is considered first‐line treatment; however, the majority of patients with NSCLC are diagnosed with intermediate to advanced‐stage disease at hospital admission and have a poor prognosis mainly due to regional recurrence or distant metastases.3 Thus, it is essential to identify candidate prognostic biomarkers and potential therapeutic targets that could improve the clinical outcomes in patients with NSCLC.

Forkhead box Q1 (FOXQ1), as a member of the forkhead box protein family, has many physiological functions, such as serving as a transcription factor, regulating tumor cell differentiation, promoting epithelial differentiation, and activating T cells and autoimmunity.4, 5, 6 Recent studies have shown that FOXQ1 is upregulated in several cancers, including breast cancer, colorectal cancer, and hepatocellular carcinoma, and promotes cell proliferation, invasion, and migration in several cancer pathologies.7, 8, 9 FOXQ1 is also associated with advanced clinicopathological features as well as an unfavorable prognosis in patients with pancreatic cancer, breast cancer, and colorectal cancer.7, 9, 10 Furthermore, cellular experiments have revealed that FOXQ1 knockdown activates the expression of epithelial markers but decreases the expression of several mesenchymal markers in some epithelial‐derived cancers; in particular, FOXQ1 regulates epithelial‐mesenchymal transition (EMT), and its knockdown decreases the proliferation, migration, and invasion of NSCLC cells.11, 12 Regarding the association of FOXQ1 with clinical outcomes in NSCLC, only one clinical study has reported an association of FOXQ1 with clinicopathological features as well as prognosis in patients with NSCLC; however, (a) it is a single‐center investigation that might bring in selection bias; (b) it is of a relatively small sample size (only 103 patients are enrolled); (c) disease‐free survival (DFS) (a key index for assessing prognosis) is not assessed; (d) the cohort included patients with TNM stage IV, which is a highlight confounding factor; and (e) the follow‐up duration is relatively short.13 Therefore, the implication of FOXQ1 in NSCLC is not clear and needs further exploration.

Thus, we conducted this multicenter study with 238 surgical patients with NSCLC (TNM stage I‐III) and aimed to investigate FOXQ1 expression in tumor tissue and adjacent tissue and to further explore the correlation of tumor tissue FOXQ1 expression with clinicopathological properties, DFS, and overall survival (OS) in patients with NSCLC.

MATERIALS AND METHODS

Patients

In the current retrospective study, 238 patients with NSCLC who underwent surgical resection at our hospitals between January 2013 and December 2014 were screened and reviewed. The screening criteria were as follows: (a) confirmed diagnosis of primary NSCLC according to the World Health Organization (WHO) classification, (b) underwent resection without neoadjuvant therapy, (c) tumor and adjacent tissue specimens removed by surgery were preserved and suitable for immunohistochemistry (IHC), and (d) age above 18 years. Patients were excluded if they were complicated with other tumors or if their clinical or follow‐up data were incomplete. The Ethics Committee of our hospitals approved the current study, and all patients or their guardians provided written informed consent or verbal agreement with tape recording.

Clinical data collection and survival assessment

Patients’ clinical data were collected by reviewing the medical records, including age, gender, smoking status, drinking status, pathological differentiation, tumor size, lymph node metastasis, TNM stage (according to the Union for International Cancer Control (UICC) system (7th edition)), and carcinoembryonic antigen (CEA) level, were collected by reviewing medical records. After surgery, patients received adjuvant chemotherapy ± radiotherapy, best supportive care, or observation based on the clinical status and TNM stage. Survival data were collected by reviewing the follow‐up records. The median follow‐up duration was 48.0 months, with the last follow‐up date of 2019/5/31. DFS and OS were calculated according to the follow‐up data.

Specimen collection and detection

Formalin‐fixed and paraffin‐embedded (FFPE) tumor and paired adjacent tissue were collected from the pathology department of our hospital with approval, and all specimens were obtained from surgical resection and treated and stored in a refrigerator (Haier) at 4°C. FOXQ1 expression in the tumor and paired adjacent tissue was detected by IHC assays.

IHC assays

FFPE specimens were sliced into 4 µm sections using a microtome (Leica), deparaffinized with xylene, rehydrated with graded ethanol followed by antigen retrieval in a microwave oven (SUPOR), and incubated with H2O2 to block endogenous peroxidase. The sections were then blocked using normal goat serum. Subsequently, the sections were incubated in a refrigerator (Haier) at 4°C overnight with the indicated primary antibody. The anti‐FOXQ1 rabbit polyclonal antibody (Abcam) was used as the primary antibody. The next day, the sections were incubated with horseradish peroxidase (HRP)‐conjugated goat anti‐rabbit IgG H&L (Abcam). Finally, the sections were stained with diaminobenzidine (DAB) and hematoxylin and then sealed followed by observation under a light microscope (Leica). A semi‐quantitative score was applied to assess the expression of FOXQ1 in the specimens based on the average intensity and density of positively stained cells by IHC.14 The intensity of positively stained cells was scored as follows: 0 (no staining), 1 (weak staining), 2 (moderate staining), and 3 (strong staining); the density of positively stained cells was scored as follows: 0 (0%), 1 (<25%), 2 (26 ~ 50%), 3 (51 ~ 75%), and 4 (>75%). The total IHC score was calculated by multiplying the intensity score by the density score. A total IHC score > 3 was defined as high FOXQ1 expression, and a total IHC score ≤ 3 was defined as low FOXQ1 expression.

Statistical analysis

Data analysis was performed using SPSS 24.0 statistical software (IBM), and graph construction was carried out using GraphPad Prism 7.02 software (GraphPad Software, Inc). Continuous data were assessed by the Kolmogorov‐Smirnov test for normality determination and are described as the mean and standard deviation (SD) if normally distributed and as the median and interquartile range (IQR) if not normally distributed; count data are expressed as the count (percentage). The comparison of continuous data was performed by Student's t test or one‐way analysis of variance (ANOVA) followed by Dunnett's t test; the comparison of count data between independent samples was performed by the chi‐square test or the Wilcoxon rank‐sum test. The comparison of count data for paired samples was performed by McNemar's test. In addition, DFS was calculated from the date of surgery to the date of disease relapse, disease progression or death; patients not known to have relapsed, progressed, or died at the last follow‐up were censored on the date they were last examined. OS was calculated from the date of surgery to the date of death; patients not known to have died at the last follow‐up were censored on the date they were last known to be alive. DFS and OS were illustrated using Kaplan‐Meier curves, and the differences in DFS and OS between different patients were assessed by the log‐rank test. Factors predicting DFS or OS were determined by univariate and forward stepwise (conditional) multivariate Cox's proportional hazards regression model analyses. All tests were two‐sided, and a P value < .05 was considered statistically significant.

RESULTS

Study flow

A total of 471 patients with NSCLC who underwent surgical resection were screened, and 192 were excluded (including 97 patients whose tumor specimens were inaccessible, 56 patients who underwent neoadjuvant treatment, 32 patients with incomplete clinical data and follow‐up records, and 7 patients who were concomitant with other tumors) (Figure 1). The remaining 279 patients with NSCLC were eligible, among which 41 were excluded because they (or their guardians (family members)) were incapable of being contacted for informed consent. Finally, 238 patients with NSCLC were reviewed and analyzed in the study.

Figure 1

Study flow. NSCLC, non‐small cell lung cancer

Comparison of FOXQ1 expression between tumor tissue and adjacent tissue

FOXQ1 expression in tumor tissue and adjacent tissue was evaluated by IHC assays (Figure 2A, Figure S1). The expression levels of FOXQ1 in tumor tissue and adjacent tissue were different (P < .001) (Figure 2B). In tumor tissue, the percentages of tumor tissue with high FOXQ1 expression and low FOXQ1 expression were 61.3% and 38.7%, respectively; in adjacent tissue, the percentages of adjacent tissue with high FOXQ1 expression and low FOXQ1 expression were 37.8% and 62.2%, respectively. These data indicate that FOXQ1 was upregulated in NSCLC tumor tissue compared with adjacent tissue.

Figure 2

FOXQ1 expression in NSCLC tumor tissue and adjacent tissue. Representative IHC images illustrate high FOXQ1 expression in tumor tissue and low FOXQ1 expression in adjacent tissue (A). The numbers (percentages) of high/low FOXQ1 expression in tumor tissue and adjacent tissue (B). The comparison of count data for paired samples was performed by McNemar's test. P < .05 was considered significant. FOXQ1, forkhead box Q1; NSCLC, non‐small cell lung cancer

Correlation of FOXQ1 expression with the characteristics of patients with NSCLC.

According to the cutoff value of the FOXQ1 IHC score at baseline, all patients were divided into patients with high FOXQ1 expression (IHC score >3) (n = 146) and patients with low FOXQ1 expression (IHC score ≤3) (n = 92) (Table 1). High FOXQ1 expression was associated with larger tumor size (P = .042), lymph node metastasis (P = .040), and advanced TNM stage (P = .002) in patients with NSCLC. However, there was no association of FOXQ1 expression with age (P = .169), gender (P = .259), smoking status (P = .747), drinking status (P = .347), pathological differentiation (P = .065), or CEA level (P = .982).

Table 1

Correlation of FOXQ1 expression with patients’ characteristics

Characteristics	NSCLC patients (N = 238)	FOXQ1 expression		P value
		High (n = 146)	Low (n = 92)
Age (years), mean ± SD	62.31 ± 10.49	62.9 ± 10.4	61.3 ± 10.6	.169
Gender, No. (%)
Male	196 (82.4)	117 (80.1)	79 (85.9)	.259
Female	42 (17.6)	29 (19.9)	13 (14.1)
Smoke, No. (%)	134 (56.3)	81 (55.5)	53 (57.6)	.747
Drink, No. (%)	92 (38.7)	53 (36.3)	39 (42.4)	.347
Pathological differentiation, No. (%)
Well	25 (10.5)	14 (9.6)	11 (12.0)	.065
Moderate	155 (65.1)	90 (61.6)	65 (70.7)
Poor	58 (24.4)	42 (28.8)	16 (17.3)
Tumor size (cm), mean ± SD	5.2 ± 2.1	5.4 ± 2.1	4.9 ± 1.9	.042
Lymph node metastasis, No. (%)	81 (34.0)	57 (39.0)	24 (26.1)	.040
TNM stage, No. (%)
I	87 (36.5)	43 (29.5)	44 (47.8)	.002
II	73 (30.7)	46 (31.5)	27 (29.4)
III	78 (32.8)	57 (39.0)	21 (22.8)
CEA (ng/mL), median (IQR)	7.0 (3.1‐32.1)	6.8 (3.2‐30.0)	8.0 (2.7‐42.5)	.982

Comparison was determined by chi‐square test, Student's t test, or Wilcoxon rank‐sum test. The three bold values represent the difference between high FOXQ1 expression and low FOXQ1 expression was considered statistically significant of tumor size/lymph node metastasis/TNM stage gropus. Higher FOXQ1 expression was associated with larger tumor size (P = .042), more lymph node metastasis (P = .040), and advanced TNM stage (P = .002). The comparison of the TNM stage was compared between the overall high FOXQ1 expression population and the overall low FOXQ1 expression populationthe rather than the subgroup, which means that the proportion of advanced patients in the high expression group was higher.

Abbreviations: CEA, carcinoembryonic antigen; FOXQ1, forkhead box Q1; IQR, interquartile range; NSCLC, non‐small cell lung cancer; SD, standard deviation.

Correlation of FOXQ1 expression with DFS and OS

DFS was reduced in patients with high FOXQ1 expression compared with patients with low FOXQ1 expression (P = .016) (Figure 3A). OS was also decreased in patients with high FOXQ1 expression compared with patients with low FOXQ1 expression (P = .008) (Figure 3B). The data above suggested that high FOXQ1 expression was associated with an unfavorable prognosis in patients with NSCLC.

Figure 3

Comparison of DFS and OS between NSCLC patients with high FOXQ1 expression and NSCLC patients with low FOXQ1 expression. Comparison of DFS between patients with high FOXQ1 expression and patients with low FOXQ1 expression (A). Comparison of OS between patients with high FOXQ1 expression and patients with low FOXQ1 expression (B). The survival profiles of NSCLC patients were visualized with a Kaplan‐Meier curve, and the comparison of survival between patients with high FOXQ1 expression and patients with low FOXQ1 expression was performed by a log‐rank test. P < .05 was considered significant. DFS, disease‐free survival; OS, overall survival; FOXQ1, forkhead box Q1; NSCLC, non‐small cell lung cancer

Factors predicting DFS by Cox's proportional hazards regression model

The univariate Cox's regression model revealed that high FOXQ1 expression (HR = 1.447, P = .018), age (HR = 1.447, P = .016), poor pathological differentiation (HR = 1.462, P = .023), tumor size (>5 cm) (HR = 2.016, P < .001), lymph node metastasis (HR = 2.313, P < .001), and TNM stage III (HR = 2.090, P < .001) were associated with decreased DFS in patients with NSCLC (Table 2). Further, the forward stepwise multivariate Cox's regression model showed that high FOXQ1 expression (HR = 1.379, P = .043), poor pathological differentiation (HR = 1.659, P = .003), and lymph node metastasis (HR = 2.261, P < .001) were independent risk factors for DFS in patients with NSCLC.

Table 2

Analysis of factors predicting DFS by univariate and multivariate Cox's proportional hazards regression model

Items	Cox's proportional hazards regression model
	P value	HR	95% CI
			Lower	Higher
Univariate Cox's regression
FOXQ1 expression (high)	.018	1.447	1.065	1.966
Age (>60 y)	.016	1.447	1.072	1.954
Gender (male)	.451	0.862	0.585	1.269
Smoke	.576	1.088	0.810	1.462
Drink	.118	0.784	0.577	1.064
Pathological differentiation (poor)	.023	1.462	1.054	2.027
Tumor size (>5 cm)	<.001	2.016	1.499	2.712
Lymph node metastasis	<.001	2.313	1.705	3.136
TNM stage (III)	<.001	2.090	1.547	2.823
CEA* (abnormal)	.105	1.290	0.948	1.755
Forward stepwise (conditional) multivariate Cox's regression
FOXQ1 expression (high)	.043	1.379	1.011	1.882
Pathological differentiation (poor)	.003	1.659	1.185	2.322
Lymph node metastasis	<.001	2.261	1.655	3.089

Abbreviations: CEA, carcinoembryonic antigen; CI: confidence interval; DFS: disease‐free survival;

FOXQ1, forkhead box Q1; HR: hazard ratio.

Abnormal: CEA > 5 ng/mL, normal: CEA ≤ 5 ng/mL.

Factors predicting OS by Cox's proportional hazards regression model

The univariate Cox's regression model showed that high FOXQ1 expression (HR = 1.573, P = .009), tumor size (>5 cm) (HR = 2.232, P < .001), lymph node metastasis (HR = 2.728, P < .001), and TNM stage III (HR = 2.256, P < .001) were associated with reduced OS in patients with NSCLC (Table 3). Further, the forward stepwise (conditional) multivariate Cox's regression model revealed that high FOXQ1 expression (HR = 1.498, P = .021), tumor size (>5 cm) (HR = 1.567, P = .014), and lymph node metastasis (HR = 2.154, P < .001) were independent risk factors for OS in patients with NSCLC.

Table 3

Analysis of factors predicting OS by univariate and multivariate Cox's proportional hazards regression model

Items	Cox's proportional hazards regression model
	P value	HR	95% CI
			Lower	Higher
Univariate Cox's regression
FOXQ1 expression (high)	.009	1.573	1.119	2.211
Age (>60 y)	.178	1.249	0.904	1.725
Gender (male)	.258	0.789	0.523	1.189
Smoke	.862	0.972	0.706	1.338
Drink	.105	0.757	0.541	1.059
Pathological differentiation (poor)	.074	1.383	0.969	1.974
Tumor size (>5 cm)	<.001	2.232	1.622	3.070
Lymph node metastasis	<.001	2.728	1.969	3.779
TNM stage (III)	<.001	2.256	1.630	3.123
CEA* (abnormal)	.131	1.290	0.927	1.795
Forward stepwise (conditional) multivariate Cox's regression
FOXQ1 expression (high)	.021	1.498	1.064	2.108
Tumor size (>5 cm)	.014	1.567	1.093	2.245
Lymph node metastasis	<.001	2.154	1.491	3.112

Abbreviations: CEA, carcinoembryonic antigen; CI: confidence interval; FOXQ1, forkhead box Q1; HR: hazard ratio; OS: overall survival.

Abnormal: CEA > 5 ng/mL, normal: CEA ≤ 5 ng/mL.

Correlation of FOXQ1 with prognosis in the subgroup analysis

In patients receiving chemotherapy, high FOXQ1 expression was numerically associated with worse OS, although the difference was not significant (P = .145) (Figure S2A). In patients without chemotherapy, high FOXQ1 expression was associated with worse OS (P = .034) (Figure S2B). In patients receiving radiotherapy, high FOXQ1 expression was associated with worse OS (P = .018) (Figure S2C). In patients without radiotherapy, there was no association between FOXQ1 and OS (P = .229) (Figure S2D). These data indirectly indicate that FOXQ1 had influence on radiotherapy sensitivity and might have potential to affect chemotherapy sensitivity to some extent.

DISCUSSION

In the present study, we observed that (a) FOXQ1 was upregulated in NSCLC tumor tissue compared with adjacent tissue, and high FOXQ1 expression was associated with advanced tumor features, including larger tumor size, lymph node metastasis, and advanced TNM stage, in patients with NSCLC and (b) high FOXQ1 expression was an independent risk factor for DFS and OS in patients with NSCLC.

FOXQ1 is a transcription factor, and its gene is located on human chromosome 6p25.3.4 Numerous studies have shown that FOXQ1 mediates all steps of tumor metastasis from initial EMT to ultimate organotrophic colonization and is implicated in regulating tumor invasion and metastasis by regulating its downstream genes, such as zinc finger E‐box binding homeobox (ZEB2), twist‐related protein 1 (TWIST1), and sex‐determining region Y‐box 12 (SOX12).15, 16 Given the key role of EMT in epithelial‐derived cancers, the role of FOXQ1 has recently been investigated in clinical studies that have indicated that FOXQ1 might act as a tumor promoter in several cancers.7, 10, 17 For example, one previous study exhibits that FOXQ1 mRNA expression is upregulated in both pancreatic cancer cell lines and tumor tissue, and its high expression is associated with a higher degree of tumor differentiation in patients with pancreatic cancer.10 Another study illustrates that the expression levels of FOXQ1 mRNA and protein are higher in gastric cancer tissue than in noncancerous tissue, and its elevated expression is associated with larger tumor size, a higher histological grade, lymph node involvement, and tumor‐node‐metastasis stage.17 As for in NSCLC, FOXQ1 has been shown to regulate EMT and correlates with resistance to chemotherapy, but only one clinical study with a small sample (only 103 patients) reports that FOXQ1 is upregulated in NSCLC tissue compared with noncancerous tissue, and high FOXQ1 expression is associated with the downregulation of E‐cadherin, the anomalous positivity of vimentin and S100 calcium binding protein A4, while there are no significant associations of FOXQ1 expression with patient age, gender, tumor diameter, histological grade of the tumor, lymph node metastasis status, or stage grouping with TNM.11, 13 However, the previous study is a single‐center study with relatively small sample size, which might lead to regional selective bias and reduced validation, and it enrolls the patients at TNM stage IV, which may bring in confounding factors and further leads to bias of the results. Therefore, we excluded patients at TNM stage IV in our study. To further validate FOXQ1 expression in NSCLC tumor tissue and its correlation with clinicopathological features in patients with NSCLC, we conducted the present multicenter study with 238 surgical patients with NSCLC at TNM stage I‐III, which revealed that FOXQ1 was upregulated in NSCLC tumor tissue compared with adjacent tissue, and high FOXQ1 expression in tumor tissue was associated with undesirable clinicopathological characteristics, including larger tumor size, lymph node metastasis, and advanced TNM stage, in patients with NSCLC. These findings might be explained by the following reasons: (1) Elevated FOXQ1 expression might activate its downstream oncogenic genes as well as signaling pathways, such as ZEB2, TWIST1, SOX12, and the Wnt signaling pathway, promoting the development of NSCLC and (2) High FOXQ1 expression might promote the proliferation, migration, and invasion of NSCLC cells by enhancing the process of EMT, which might accelerate the development and progression of tumors; thus, its high expression was correlated with larger tumor size, lymph node metastasis and advanced TNM stage in patients with NSCLC.

Existing studies have indicated that FOXQ1 is of prognostic value in several cancers.9, 10, 17 For example, one study suggested that high FOXQ1 expression is an independent risk factor for OS in patients with gastric cancer.17 In another study, high FOXQ1 expression independently predicts poor OS in patients with pancreatic cancer.10 However, only one study with a relatively small sample size reports that high FOXQ1 expression is an independent risk factor for OS in patients with NSCLC; however, this study lacks the analysis of DFS (an important prognostic indictor for NSCLC) and includes patients at TNM stage IV, which may introduce confounding factors and further contributes to bias of the results, and its follow‐up duration is relatively short. These previous studies all suggest that FOXQ1 might represent a potential prognostic biomarker in several epithelial‐derived cancers. Furthermore, according to the previous finding in our study that high FOXQ1 expression in tumor tissue was associated with undesirable clinic‐pathological characteristics, including larger tumor size, lymph node metastasis, and advanced TNM stage, in patients with NSCLC, these factors might be confounding factors, which would affect the correlation of FOXQ1 with prognostic data. Therefore, we used a forward stepwise (conditional) multivariate Cox's proportional hazards regression model to analyze all factors affecting prognosis to weaken the effect of these confounding factors on the results. Our studies demonstrated that high FOXQ1 expression was an independent risk factor for poor survival in patients with NSCLC. The possible reasons might include the following: (a) high FOXQ1 expression was associated with larger tumor size, lymph node metastasis, and advanced TNM stage; thus, FOXQ1 might indirectly influence prognosis by interacting with these clinicopathological properties; (b) FOXQ1 upregulation enhanced the chemoresistance and aggressiveness of NSCLC cells, therefore reducing the responsiveness to treatment, and a poor survival profile might be observed in patients with NSCLC who receive adjuvant chemotherapy 18; and (c) high FOXQ1 expression promoted the progression of EMT, which broke the dormancy of relapse‐initiating cancer stem cells, further leading to a higher risk of relapse and metastasis in patients with NSCLC. Interestingly, we also found that FOXQ1 influenced radiotherapy sensitivity and might have the potential to affect chemotherapy sensitivity to some extent. Thus, NSCLC patients with high FOXQ1 expression had a poor prognosis in a long‐term period.

However, there still exist some limitations in our present study. (a) The present study was a retrospective study in nature; therefore, some selective bias might exist. (b) The underlying mechanism of FOXQ1 in NSCLC was not explored; thus, further cellular experiments were needed in the future. (c) Research on FOXQ1 was still at beginning, and whether the measurement of FOXQ1 was valuable for application in regular clinical practice needed further exploration.

In conclusion, high FOXQ1 expression is associated with advanced tumor features as well as undesirable survival profiles in NSCLC patients, implying the potential prognostic value of FOXQ1 for NSCLC.

Click here for additional data file.

Click here for additional data file.