At present, most studies are based on imaging findings to predict the growth trend of tumors. For example, Hyun Jung Yoon et al. divided the patients into two groups according to the growth pattern of lung adenocarcinoma, extracted the CT imaging features reflecting marginal features and tumor doubling time and adopted to predict tumor doubling time [3]. YukihiroYoshida et al. also adopted high-resolution computed tomography images to predict the growth and changes of non-invasive adenocarcinoma [4]. However, the biological indicators or test methods that can be adopted to predict the growth of pulmonary nodules have not been systematically expounded and proved. However, there are also studies looking for factors related to nodule growth [16,17,18], which provide some research directions and ideas for predicting the growth of pulmonary nodules.

At present, a great quantity of studies [19,20,21] are based on the immunohistochemical analysis of middle and advanced lung cancer to find out the internal relationship between the relevant immune indexes, related factors and the occurrence, development, invasion, metastasis and prognosis of lung cancer. However, for the early lung cancer with ground glass nodules shown by imaging, there are few studies on tumor-related immune indexes. Some studies have proved that early detection of pulmonary nodules can effectively improve the survival rate of patients with lung cancer [22]. As a consequence, our research draws lessons from the study of immune indexes related to middle and advanced lung cancer, and selects immune indexes that promote tumorigenesis and development. It intends to observe the role of immune indexes in the growth or progression of early lung cancer, that is, pulmonary nodules. It intends to forecast the growth of pulmonary nodules to achieve the ultimate goal of early detection of malignant nodules, timely clinical treatment, and improve the survival rate of patients. The innovation of our research is to study the early lung cancer, that is, pulmonary nodules, and analyze the correlation between immune indexes and them. Through the statistical analysis of the experimental results, it is concluded that VEGF may be one of the important factors affecting early tumor progression.

VEGF plays an important part in the pathology of many cancer patients. Studies have shown that the increase of VEGF is significantly related to tumor proliferation, migration and endothelial cell invasion [10]. As a multi-subtype and multi-family protein first isolated from tumor cells, VEGF is a highly specific vascular endothelial growth factor. Its main role is to mediate angiogenesis in normal and diseased tissues, which is very important for tumor growth and is related to the poor prognosis of lung cancer [23]. The results of our study are basically consistent with the above point of view, suggesting that the high expression of VEGF is strongly related to the growth of lung cancer. It is speculated that the expression of VEGF plays a promoting role in the early growth of tumor, which may lay a foundation for further research on the internal factors of ground glass nodule growth.

Our study indicated that the patients with different expression levels of COX-2 in early lung cancer showed a high level of expression, the growth of pulmonary nodules was more significant. COX-2 is an enzyme which can converts arachidonic acid into prostaglandins (PGs). There is evidence that the overexpression of COX-2 in lung cancer promotes the proliferation, invasion, angiogenesis and anti-apoptosis of tumor cells [7]. Recently, more and more studies have focused on the discovery of transcription factors regulating the expression of COX-2, such as SP1, AP-2, CPSF4, CBP/P300, NF-kappa B and BPTF [24,25,26,27,28,29]. Although there are many regulatory factors, its carcinogenic effect is clear. In addition, the expression of COX-2 can regulate matrix metalloproteinases (MMP-7), leading to proliferation and invasion of lung adenocarcinoma [30]. The nodule growth in patients with pulmonary nodules is closely related to the expression of COX-2. Combined with the experimental results of our study, it can be further inferred that the expression of COX-2 can affect the growth velocity of pulmonary nodules in the early stage of lung cancer. The mechanism of COX-2 promoting the growth of early pulmonary nodules may be established, but it still needs to be proved by experiments, and the factors leading to these mechanisms are also worthy of further study.

Ki67 is a kind of macromolecular protein reflecting the activity of cell proliferation, and it is a widely adopted cell marker in the whole proliferative phase at present. It is expressed in every proliferation cycle except G0 phase. It is one of the best indicators of tumor active proliferation. It plays an important part in judging cell proliferation, analyzing cell growth and metastasis and predicting the prognosis of patients [9]. Our research shows that the increase of Ki67 can predict the growth trend of early pulmonary nodules. However, its correlation with nodular growth was weaker than that of COX-2. Related studies have also shown that Ki-67 can promote cell proliferation by regulating cell cycle [31,32,33,34]. However, the specific regulatory mechanism and the regulatory factors of Ki-67 itself are not very clear, which can be proved by experiments in follow-up studies.

However, the comprehensive consideration of COX-2, Ki67 and VEGF showed a strong correlation with nodular growth. Some studies have shown that COX-2 may be an active stimulator of VEGF in NSCLC tissues [35]. In addition, prostaglandin E2 (PGE2), the main product of tumor COX-2, regulates VEGF through EP3 and EP4, which has an important effect on tumor matrix formation and tumor growth, and regulates VEGF-C and promotes lymphangiogenesis in lung adenocarcinoma through EP1 [36, 37]. Combined with the above related studies, it is concluded that besides regulating VEGF, COX-2 has other regulatory pathways that can directly or indirectly affect tumor growth, but in the real process of tumor growth, there is still some uncertainty about which regulatory pathway is dominant, which may also be one of the reasons why the correlation between COX-2 and early lung cancer growth is more significant than VEGF in this study. In the early stage of pulmonary nodule growth, COX-2 and VEGF have been combined to promote nodule growth. In the study of the relationship between COX-2 and Ki-67, it was found that the average value of Ki-67 labeling index in COX-2 positive tumors was significantly higher than that in COX-2 negative tumors [38]. However, our study is based on patients with gastric cancer as the research object. There is no article to prove that there is a relationship between COX-2 and Ki-67 in the occurrence and development of lung cancer. Similarly, there are no reports about the interaction between VEGF and Ki-67 in the process of tumor growth. Combined with the results of our study, as a marker of tumor growth, the relationship between Ki-67 and COX-2 and VEGF may be due to the performance of COX-2 and VEGF promoting tumor growth, but it can not be ruled out that COX-2 or VEGF have some effects on Ki-67, which needs to be verified by follow-up experiments.

CD44v6 is a soluble intercellular adhesion molecule, which is closely related to the invasion, progression and metastasis of malignant tumors [11]. CD44s and CD44V6 were predominantly present in normal lung tissue [39, 40]. With the activation of receptors on the surface of CD44 cells, some downstream signal events will be expressed accordingly [41, 42]. This interaction can lead to tumor growth. The expression of CD44s and CD44v6 in the origin cells of lung adenocarcinoma is inconsistent [43]. Given the results that we got in our study, it can be concluded that there is no close relationship between CD44V6 and nodule growth in early lung nodules, which may be related to the different origin cells of lung adenocarcinoma. EGFR is a transmembrane receptor tyrosine kinase that regulates cell proliferation and growth-related signal transduction through phosphorylation [12]. At present, studies have shown that EGFR can promote cell division and proliferation, and up-regulated expression in many malignant tumors, including lung cancer [13]. However, regardless of the metastatic status, NSCLC had a higher amplification rate of EGFR gene, and the amplification rate of EGFR gene was related to the differentiation degree of NSCLC [44]. The subjects of this study are patients with early NSCLC, so it can be suspected that the amplification rate of EGFR gene is lower in patients with early pulmonary nodules. The content of MDM2 oncogene as a proto-oncogene in normal tissues is very small, but the positive expression rate of MDM2 oncogene in NSCLC tissues increases with the increase of histological differentiation, and is significantly different from lymph node metastasis and tumor TNM stage [14]. TGF-β1 is a complex cytokine, which is closely related to tumor. In the process of tumor cell proliferation, its regulatory role has two sides [15]. In the early stage of cancer, TGF-β inhibits the growth of tumor cells and induces tumor cell apoptosis. On the contrary, in the later stage of tumor progression, it acts as a promoter to promote tumor invasion and metastasis [45]. As an early stage of lung cancer, the role of TGF-β1 in lung nodules changes from growth inhibition to promotion with the progression of lung cancer, which may also make the correlation between it and the growth of lung nodules unclear. As a consequence, the above factors were not significantly correlated with the growth of early pulmonary nodules in this study.

Combined with our experimental results, postoperative immunohistochemical staining will have a positive impact on clinical treatment. For patients with pulmonary nodules with high expression of COX-2, Ki67 and VEGF, the growth of nodules is in a rapid growth state. For patients with inert pulmonary nodules whose immunohistochemical results showed that COX-2, Ki67 and VEGF were expressed at low levels after operation, the growth rate of nodules was relatively slow. In clinical practice, most patients can determine whether they need surgery through preoperative imaging and liquid examination. However, there are still a few patients whose preoperative examination can not determine whether they can be operated. For example, the imaging findings of patients before surgery suggest that the nodules are large, but the results of multiple examinations reflect that the growth of nodules is not obvious. When the patient is hesitant about the choice of treatment scheme due to consideration of other factors such as basic disease or physical condition, the immunohistochemical results can be obtained through puncture biopsy or tracheoscopy before operation to guide the selection of treatment scheme. According to the analysis of immunohistochemical results, if the expression of COX-2, Ki67 and VEGF is increased, it can be considered that the lung nodule growth rate of these patients is high. We will advise them to take surgical treatment in time. If the immunohistochemical results of patients suggest that the growth of nodules is slow, we can also follow up these patients regularly to observe the changes of pulmonary nodules.

This study can also provide the basis for genetic testing in the blood of patients in the follow-up study, and then screen the genotypes that promote the growth of nodules. We will find out the key factors that promote the growth of nodules through the analysis of gene expression. In the clinical environment, due to difficulties in operation or multiple complications of patients, some patients may not be able to obtain the expected immunohistochemical results through preoperative puncture biopsy or tracheoscopy. For this kind of patients, we can determine whether the patient has a gene that promotes nodule growth by testing before operation, so as to guide clinical treatment. The exploration of this aspect will be the focus of our research in the future.

This study has the following limitations: (1) The experiment in this paper is a retrospective analysis. Due to the failure to carry out the clinical experiment design in advance, our study can only analyze the existing clinical data. The immunohistochemical results are all independent tests after surgical resection of pulmonary nodules, and the relevant test results of patients before surgery are not obtained. Therefore, compared with prospective studies, our experimental results also have certain limitations. In the follow-up study, we will adopt a prospective experimental design. Patients who could obtain immunohistochemical results before operation and found pulmonary nodules for the first time were included, and their test data were obtained for clinical observation. (2) The overall sample is small, which may not fully reflect the situation of all patients. (3) This study only collected preoperative imaging parameters, no relevant hematological indicators, postoperative data, etc., may also affect the final results. (4) This study is limited to the study of the relationship between tumor-related immune indicators and the growth of pulmonary nodules, and will be further studied at the molecular level in the future. As a new type of index to predict the growth of pulmonary nodules, COX-2, Ki67 and VEGF still need to further explore the specific mechanism of their effects on the growth of pulmonary nodules. Large-scale multicenter, prospective clinical studies are needed to verify their clinical value.