The Abnormal Expression of miR-205-5p, miR-195-5p, and VEGF-A in Human Cervical Cancer Is Related to the Treatment of Venous Thromboembolism

Background Low molecular heparin (LWMH) therapy can prevent the occurrence of VTE in tumor patients and may have a direct antitumor effect. However, the expression pattern of VEGF-A and microRNAs was less reported in cervical cancer subjects who received concurrent chemoradiotherapy (CCRT) or received anticoagulant treatment with low molecular weight heparin (LWMH) after CCRT (CCRT+LWMH). Methods In this study, 30 cervical cancer subjects treated with CCRT and 30 cervical cancer patients treated with CCRT+LWMH were enrolled. We screened five miRNAs (miR-15a-5p, miR-16-5p, miR-29a-3p, miR-195-5p, and miR-205-5p), which have multiple binding sites with VEGF-A and are highly expressed in serum of patients with cervical cancer, by RT-qPCR. The expression level of VEGF-A was also detected by RT-qPCR and ELISA. Statistical methods were used for difference and correlation analyses. Results We observed the curative effect in the two treatment methods. In the CCRT group, the total effective rate was 60.00%, and in the CCRT+LWMT group, the total effective rate was 83.33% (P = 0.013, χ2 = 6.129). Additionally, the serum levels of VEGF-A in the CCRT+LWMH group were downregulated, relative to the CCRT group (P < 0.05), and VEGF-A in serum was significantly positively correlated with venous thromboembolism (VTE) (r = 2.134, P = 0.035). Only miR-205-5p and miR-195-5p were upregulated in CCRT+LWMH, relative to CCRT (P < 0.05). In serum of patients with cervical cancer after CCRT+LWMH treatment, there was no significant correlation between VEGF-A and miR-15a-5p (r = −0.132, P = 0.209), miR-16-5p (r = −0.205, P = 0.311), or miR-29a-3p (r = −0.029, P = 0.662), but VEGF-A was significantly negatively correlated with miR-195-5p (r = −0.396, P = 0.040) and miR-205-5p (r = −0.315, P = 0.032). Furthermore, VTE was also significantly negatively correlated with miR-195-5p (r = −0.412, P = 0.031) and miR-205-5p (r = −0.123, P = 0.044). Conclusion These data revealed roles for VEGF-A and these miRNAs as potential biomarkers in cervical cancer patients with VTE, which exhibited usage potential in the treatment of venous thromboembolism.


Background
Venous thromboembolism (VTE) includes deep vein thrombosis (DVT) and pulmonary embolism (PE), which is based on a hypercoagulable state, vascular wall damage, and tumor compression vessels ( [1]). VTE is a common complication and the second cause of death in patients with malignant tumors, second only to the tumor itself, and a malignant tumor is an important cause of VTE [2]. According to statistics, 20% of VTEs occur in patients with malignant tumors and about 20% of patients with malignant tumors can be diagnosed with VTE, but this result is as high as 50% at autopsy [3]. VTE can be the first clinical manifestation of cancer patients. About 10% of patients with primary thrombosis are diagnosed with malignant tumors several years after the thrombotic event, and about 40% of the patients had distant metastasis by the time of diagnosis [4]. According to different pathological types, the overall incidence of VTE in the first hospitalized adult malignant tumors ranged from 2.7% to 12.1% [5], and the highest incidence of several tissue sources were the brain, uterus, bladder, pancreas, and stomach. In the advanced patients with distant metastases, the incidence of VTE is 4-13 times higher than that of lesions [6]. Thus, VTE may be closely related to the occurrence and development of tumors.
Many epidemiological studies have demonstrated that anticoagulant therapy not only prevents the occurrence of VTE in cancer patients but also may have a direct antitumor effect, thereby inhibiting tumor metastasis and prolonging patient survival [7]. Low molecular heparin (LWMH) is currently a commonly used anticoagulant drug with a strong antithrombotic effect, long half-life, less bleeding side effects, and high safety; clinical application and research show that in addition to its anticoagulant effect, LWMH also has antiangiogenic and antitumor effects ( [8]). It is well known that tumor angiogenesis is a necessary condition for tumorigenesis, growth, invasion, and metastasis. Moreover, the proliferation of vascular endothelial cells is a key link in the process of angiogenesis. Vascular endothelial growth factor (VEGF) is the most important proangiogenic factor ( [9]). Furthermore, radiation can induce tumor cells to secrete vascular endothelial growth factor (VEGF), promote pathological neovascularization, and cause a hypoxic state of tumor cells [10]. Hypoxia can upregulate VEGF-A levels in the tumor microenvironment again, thus increasing the vicious circle [11]. More importantly, LMWH can inhibit the secretion of VEGF-A by different tumor cell lines, inhibit the proliferation of vascular endothelial cells, and thus inhibit tumor angiogenesis. Marchetti et al. [12] confirmed that LMWH can inhibit the migration of vascular endothelial cells and the formation of luminal structures. Moreover, it has been found that LMWH and its sulfated derivative S-NACH, in addition to inhibiting tumor growth, can also promote the uptake of chemotherapeutic drugs by tumor cells, which may improve tumor resistance [13].
Cervical cancer is one of the most common gynecological malignancies that seriously threaten women's health [14]. Chemotherapy alone was a risk factor for VTE, and there are many reports about the relationship between cervical cancer and thrombotic disease [15]. However, there are few studies on preventive anticoagulant therapy for cervical cancer patients. In the present study, we assessed the curative and anticoagulant effects of LWMH on cervical cancer patients who received concurrent chemoradiotherapy (CCRT). The levels of VEGF-A and microRNAs (miR-15a-5p, miR-16-5p, miR-29a-3p, miR-195-5p, and miR-205-5p), which have multiple binding sites with VEGF-A, were analyzed in the serum of cervical cancer patients, in order to verify if they could be valuable biomarkers for rapid diagnosis of cervical cancer, as well as promising therapeutic agents.

Patient Enrollment.
The study population included 30 cervical cancer patients who received concurrent chemoradiotherapy (CCRT) and 30 cervical cancer patients who received anticoagulant treatment with low molecular weight heparin (LWMH) after CCRT (CCRT+LWMH) ( Table 1) from 2017 to 2018 in People's Hospital of Xinjiang Uyghur Autonomous Region.
Exclusion criteria were as follows: (1) the following anticoagulation contraindications-(i) recent central nervous system hemorrhage and intracranial or spinal cord high-risk hemorrhage, (ii) active hemorrhage (hemorrhage): 24-hour transfusion more than 2 units, (iii) chronic and clinical significance measurable bleeding > 48 hours, (iv) thrombocytopenia (platelets < 50000/mcL), (v) severe platelet dysfunction, (vi) large-scale surgery with high risk of bleeding recently, (vii) basic disease of coagulopathy, (viii) fall, and (ix) spinal anesthesia/lumbar puncture; (2) metastatic tumor; (3) history of other malignant tumors or any malignant tumor; (4) combination with other serious systemic diseases and inability to tolerate antitumor treatment; (5) those who received chemotherapy or surgery before concurrent radiotherapy and chemotherapy; (6) those who did not complete the planned treatment plan in this study for various reasons.

Ethical
Approval. This study was approved by the ethics committee of People's Hospital of Xinjiang Uyghur Autonomous Region (KY2016052011). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. , and no patient (0 patient) had progressive disease (PD). The total effective rate was 83.33%. The difference between the two was statistically significant (P = 0:013, χ 2 = 6:129) ( Table 2).

The Expression of VEGF-A in the Serum of Patients with
Cervical Cancer Is Correlated with VTE after CCRT+LWMH Treatment. The changes of VEGF-A levels in the serum were measured by ELISA. Firstly, age and clinical grade had no significant effect on VEGF-A levels (P > 0:05, Figures 1(a)  and 1(b)). Furthermore, compared with the CCRT group, the levels of VEGF-A in the CCRT+LWMH group were downregulated significantly (P < 0:05, Figure 1(c)). The correlation between VEGF-A and VTE in patients with cervical cancer was assessed. In serum of patients with cervical cancer after CCRT+LWMH treatment, the level of VEGF-A was significantly positively correlated with VTE (r = 2:134, P = 0:035).

Discussion
In the present study, CCRT+LWMH treatment could effectively slow the progress of the VTE and decrease the serum VEGF-A levels in cervical cancer patients. Moreover, miR-195-5p and miR-205-5p, which have binding sites with VEGF-A and are highly expressed in serum of patients with cervical cancer, were also upregulated after CCRT+LWMH treatment. Furthermore, VTE was remarkably correlated with VEGF-A, miR-195-5p, and miR-205-5p. Our findings suggested that the abnormal expression of miR-205-5p, miR-195-5p, and VEGF-A in human cervical cancer is related to the VTE. VTE is a recognized complication of cervical cancer and has high morbidity and mortality in cervical cancer patients ( [16]). However, the pathophysiological mechanisms of VTE in patients with cervical cancer are complex, which include activation of coagulation and fibrinolysis systems, inflammatory response, acute-phase response, apoptosis, and cytokine production ( [17]). Many epidemiological studies have demonstrated that anticoagulant therapy not only prevents the occurrence of VTE in cancer patients but also may have a direct antitumor effect, thereby inhibiting tumor metastasis and prolonging patient survival [18]. LWMH is currently a commonly used anticoagulant drug with strong antithrombotic effect, long half-life, less bleeding side effects, and high safety [19]. In the present study, in the course of treatment for remission of VTE, we found that the total effective rate was 83.33% in CCRT with LWMH treatment, but the total effective rate was only 60% in CCRT without LWMH treatment. Our study confirmed that LWMH injection could be an efficient way to alleviate VTE with CCRT treatment.
The VTE-related molecular mechanisms were further investigated. It was found that tumor cells can produce tissue factor (TF), cancer procoagulant (CP), inflammatory factors, and cytokines to directly activate blood coagulation [20]. Studies [21] have shown that overexpression of TF is a major factor in the development of cancer-associated VTE. TF not only activates the exogenous coagulation cascade but also stimulates tumor angiogenesis via upregulating VEGF and downregulating thrombospondin [22]. Importantly, it has been reported that LMWH can inhibit the secretion of VEGF by different tumor cell lines, inhibit the proliferation of vascular endothelial cells, and thus inhibit tumor angiogenesis ( [23]). Consistent with the above studies, we confirmed that LMWH inhibits the serum levels of VEGF-A from cervical cancer patients. VEGF-A, a key subtype of VEGF which is a member of the platelet-derived growth factor (PDGF) superfamily, is produced and secreted by a variety of cells and can promote vascular hyperplasia and regulate vascular   Figure 2: The expression levels of miR-15a-5p, miR-16-5p, miR-29a-3p, miR-195-5p, and miR-205-5p are tested by using RT-qPCR in serum of patients with cervical cancer: (a) levels of miRNAs in control and cervical cancer groups ( * P < 0:05 vs. control); (b) the binding sites between miRNAs and VEGF-A; (c-g) the expressed changes of the miRNAs after different treatments ( * P < 0:05 vs. CCRT). CCRT: concurrent chemoradiotherapy. CCRT+LWMH: low molecular weight heparin (LWMH) treatment after CCRT.   [24]. Waltham et al. [23] found that VEGF can promote the lumen recanalization and thrombus organization. Similarly, overexpression of VEGF-A could inhibit the thrombotic occlusion and incidence of mural thrombus in a balloon injury model of the femoral artery of New Zealand white rabbits [25]. These results suggested that the abnormal expression of VEGF-A plays a key role in the thrombogenesis.
VEGF and its functions have been confirmed to be regulated by different miRNAs. For example, the serum levels of VEGF in transient and acute ischemic stroke patients were downregulated compared to control which was associated with the expression levels of miR-195-5p [26], and miR-195-5p and miR-451a have been shown to target VEGF-A in several experimental settings [27][28][29]. Additionally, miR-205-5p suppressed VEGF expression in human keloid fibroblasts [30]. Moreover, a number of studies focus on the role of miRNA-mediated VEGF in cervical cancer. It was reported that miR-144 inhibits growth and metastasis of cervical cancer cells by suppressing VEGF-A and VEGF-C [31]. In the present results, we also verified that the abnormal expression of miR-205-5p and miR-195-5p in the serum of cervical cancer patients was negatively related to serum VEGF-A.
Interestingly, both miR-195-5p and miR-205-5p levels were significantly upregulated in CCRT+LWMH-treated patients compared to the CCRT group. This elevation could be related to the VET event during cervical cancer development, especially when associated with lower VEGF-A serum levels. Although we cannot discern the exact time point at which miR-195-5p and miR-205-5p were upregulated, the negative relation between the two miRs and VEGF-A may be used to mark when the VTE happens or is cured. In patients with cervical cancer, the present results found that VTE was significantly negatively correlated with miR-195-5p and miR-205-5p. We have predicted that VEGF-A could be a target gene of miR-195-5p or miR-205-5p. Previous studies have identified the hypothesis in different cell lines, and a negative correlation between miR-195-5p/miR-205-5p and VEGF was demonstrated [32,33].
Our studies suggest a role for VEGF-A, miR-195-5p, and miR-205-5p as potential biomarkers in cervical cancer patients treated for venous thromboembolism.

Limitations
This study does not further discuss why the VET was positively correlated with VEGF-A or negatively related to miR-195-5p/miR-205-5p. Our data were collected in a small group of patients and at a single institution.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors declare no potential conflicts of interest with respect to the research, authorship, and publication of this article.