High preoperative plasma vasohibin‐1 concentration predicts better prognosis in patients with non–small cell lung carcinoma

Abstract Background and Aim Vasohibin‐1 (VASH1) is an angiogenesis inhibitor synthesized and secreted by endothelial cells, whose expression is induced by angiogenic stimuli such as vascular endothelial growth factor. We have previously demonstrated that VASH1 is immunohistochemically evident in endothelial cells in the tumor microenvironment of patients with non–small cell lung cancer (NSCLC) and is positively correlated with that of vascular endothelial growth factor in cancer cells. Here, we determined the preoperative plasma concentration of VASH1 in patients with NSCLC and evaluated the association between the preoperative VASH1 levels and certain outcomes. Methods We analyzed presurgical plasma VASH1 concentrations in a total of 79 lung cancer patients (51 males and 28 females; 34‐83 y of age; 46 adenocarcinomas, 27 squamous cell carcinomas, and 6 other types) who underwent lung resection. The impact of preoperative VASH1 level was analyzed using clinical characteristics and prognosis. Results Plasma VASH1 concentration ranged from 34.1 to 1190.4 fmol/mL. We divided the patients into 3 groups according to plasma VASH1 level for this assessment: low VASH1 group (n = 26), medium VASH1 group (n = 27), and high VASH1 group (n = 26). The death and recurrence rates of the high, medium, and low VASH1 groups were 5.5, 16.2, and 12.7 per 100 person‐years, respectively. Multivariate adjusted hazard ratio of death and recurrence of the high VASH1 group was lower than that of the low VASH1 group (hazard ratio 0.42; 95% CI 0.17‐0.99). Conclusion The present analysis suggests that high preoperative plasma VASH1 concentration is associated with better prognosis in patients with NSCLC. We propose preoperative VASH1 level as a biomarker for the prognosis of patients with non–small cell lung carcinoma.


| INTRODUCTION
The leading cause of cancer-related deaths in the world is lung cancer. 1 Non-small cell lung cancer (NSCLC) represents the majority of lung cancer cases, and up to 20% of the total cancer-related deaths is derived from NSCLC. 2 Surgical treatment is a standard treatment for early-stage NSCLC. 3 The outcome of surgical treatment has been improved due to better operation procedures, optimal postoperative care, and advances in chemotherapy. 4,5 However, the 5-year survival rate of NSCLC remains only around 50%. 5 Solid tumor is composed of tumor cells, tumor vessels, infiltrated stromal and inflammatory cells, and accumulated extracellular matrices. 6 Tumor vessels lined with endothelial cells (ECs) are critical for tumor progression, as they supply oxygen and nutrients to the cancer cells and also act as a doorway for cancer cells for metastatic dissemination. Tumor vessels are formed by a process known as angiogenesis or neovascularization, 7 and this process is controlled by a local balance between its stimulators and inhibitors. 8 Our analysis has identified vasohibin-1 (VASH1) as an endogenous angiogenesis inhibitor synthesized and secreted by ECs, whose expression is induced by representative angiogenesis stimulators such as vascular endothelial growth factor (VEGF) and fibroblast growth factor 2. 9 Thus, we proposed VASH1 as a negative feedback regulator of angiogenesis. 9 Indeed, our subsequent analyses have revealed that VASH1 is immunohistochemically detectable in ECs of tumor vessels, but not in ECs of normal vessels in human, 10 and inhibits tumor growth and metastasis in experimental murine models. [10][11][12] Due to the characteristics of VASH1 as an angiogenesis inhibitor, increased expression of VASH1 is expected to be beneficial and should contribute to a better prognosis of cancers. However, contrary to this expectation, immunohistochemical analyses of pathological human sections have revealed that increased VASH1 immunostaining in ECs of tumor vessels is predictive of poor clinical outcome in multiple cancer types including NSCLC [13][14][15][16][17][18][19][20] ; an exception to this is renal cell carcinoma. 21 The mature VASH1 protein is composed of 365 amino acids and has a predicted molecular weight of 44 kDa. This protein is posttranslationally degraded after its secretion into at least 2 truncated forms: a 36-kDa protein and a 29-kDa protein. 22 The 36-kDa VASH1 protein is an N-terminally truncated form retaining antiangiogenic activity, whereas the 29-kDa VASH1 protein is truncated at both the N-terminal and C-terminal ends and loses antiangiogenic activity. 22 Immunohistochemical staining of VASH1 protein within ECs may thus not immediately correlate with antiangiogenic activity of VASH1 after secretion.
We have developed a VASH1 enzyme-linked immunosorbent assay (ELISA) that can detect the 44-and 36-kDa VASH1 forms that preserve antiangiogenic activity. 11,22 Here, we applied this system, determined the preoperative plasma VASH1 concentration in patients with lung cancer, and evaluated its potential value as a prognostic factor in patients after surgery. A total of 147 patients with lung tumors underwent operation at Tohoku University Hospital, Japan, from April 2005 to January 2007. Among them, 99 patients who consented to this study were subject to plasma VASH1 level quantitation before surgery. Twenty patients were excluded for the following reasons: benign disease (6), metastatic diseases (2), small cell lung cancer (1), stage IV (7), and insufficient data (4).

| Patient enrollment and sample and data collection
A total of 79 patients were followed up by experienced clinicians with standard physical examination, including chest X-rays, chest computed tomography scans, and tumor marker assessments, according to histological types for at least 5 years after surgery, unless recurrences were detected or the patients dropped out from the follow-up. Data were collected through in-person interviews using a standard epidemiological questionnaire, including age, sex, smoking status, performance status, and permanent residence. Clinical information including pathological stage, sex, smoking status, performance status, and prognosis were assessed using medical records. This staging was based on Classification of Lung Cancer, Seventh Edition, by The Japan Lung Cancer Society. 22 Date of death and causes of death were obtained by mail questionnaires or permission from the Ministry of Justice after the follow-up. This study was approved by the ethical review board of the Tohoku University Hospital and was conducted in accordance with the principles specified in the Declaration of Helsinki.

| Enzyme-linked immunosorbent assay for VASH1
Enzyme-linked immunosorbent assay for VASH1 was performed as described previously. 11,23 A peptide corresponding to Gly286-Arg299 (VC) and Ala217-Lys229 (VR1) of human VASH1 protein (Q7L8A9) were conjugated with keyhole limpet hemocyanin, and these antigens were immunized to A/J mice, and several monoclonal antibodies (mAbs) were prepared. 9 Among them, we found that a particular combination of VC-derived clones, recognizing Gly286-Arg299, and VR1-derived clones, recognizing Ala217-Lys229, was the best for the VASH1 ELISA. 11 We used VC1-derived clone 12F6 and VRderived clone 12E7 for plate coating and horseradish peroxidase (HRP) labeling, respectively. A 100-μL aliquot of anti-VASH1 mAb VR1 (10 μg/mL) in phosphate-buffered saline (PBS) was added to each well of an immunoplate and incubated overnight at 4°C. Each well was washed with saline and blocked with 200 μL of PBS containing 0.5% bovine serum albumin. An aliquot of the sample (25 μL) or standard diluted with assay buffer (100mM PBS, pH 7.0, containing 0.5% bovine serum albumin) and 75-μL assay buffer were added to each well. The plate was left at 4°C overnight and then washed 3 times with washing buffer. Next, 100 μL of antivasohibin mAb-HRP (1 μg/mL VC-HRP in assay buffer) was added to each well, and the samples were incubated at room temperature for 2 hours. The plate was washed 3 times with washing buffer, and the immunoreactivity was visualized by adding 100 μL of substrate solution (Color-Burst Blue, AlerCHECH, Springvale, Maine), leaving the plate to stand for 30 minutes at room temperature. The reaction was stopped by addition of stop buffer, and the absorbance at 450 nm was measured using ARVO HTS (PerkinElmer Life Science, Yokohama). This ELISA system detects the 44-and 36-kDa VASH1 form, but not the 29-kDa VASH1 form. 23

| Statistical analysis
The association between serum VASH1 levels and baseline characteristics was evaluated by χ 2 test, t test, or one-way analysis of variance, as appropriate. To compare the relapse-free survival (RFS), we used Kaplan-Meier survival plots and log-rank statistics. Multivariate analyses were performed using the Cox proportional hazards regression analysis. We counted the number of person-years of follow-up for each patient from the operation day until the date of death, the relapse of the lung cancer, or the last confirmation of existence, whichever occurred first.   Table 1.
Among 79 patients, 51 were male and 28 were female. The age ranged from 34 to 83 (median, 67). The numbers of patients according to pathological stages were as follows: IA 31, IB 20, IIA 13, IIB 6, and IIIA 9. The numbers for each pathological type were as follows: 45 adenocarcinomas, 28 squamous cell carcinomas, and 6 others  group and medium VASH1 group intersected at 3 years after surgery.
As such, we analyzed the data before 3 years after surgery and after 3 years after surgery. The effects of VASH1 levels may differ before 3 years after surgery and after 3 years after surgery. Overall survival showed similar trend as RFS ( Figure 3 and Table 3).
The association between VASH1 and RFS was analyzed using

| DISCUSSION
The mature 44-kDa VASH1 protein is degraded into a 36-kDa protein and a 29-kDa protein after secretion. The 36-kDa protein retains antiangiogenic activity, whereas the 29-kDa protein loses it. 11 We have recently shown that this degradation is enhanced in the presence of cancer cells, which indicates that the VASH1 protein secreted by ECs is inactivated in the tumor microenvironment. 24 Here, we applied a VASH1 ELISA assay, which detects antiangiogenic 44-and 36-kDa protein, but not inactive 29-kDa protein, 11,23 and evaluated the potential value of preoperative plasma VASH1 concentration as a prognostic factor in patients with resected lung cancer. Our analysis revealed that the high plasma VASH1 group showed a lower relapse and death rate than did the low VASH1 group throughout the observation period. We performed multiple comparisons and confirmed this result. This is the first demonstration that the determination of Relapse-free survival rates were 61.2% in high VASH1 group, 47.6% in medium VASH1 group, and 32.1% in low VASH1 group. High VASH1 group had better prognosis than had low VASH1 level patients. P = .056, log-rank test. VASH1 indicates vasohibin-1 preoperative plasma VASH1 concentration can be used for prognostic assessment of lung cancer patients after surgery.
Multiple studies, including ours, have examined whether the immunohistochemical staining of VASH1 can be used as a biomarker in various cancer types. Contrary to expectations, most studies have shown that an increased intensity of VASH1 immunostaining in tumor vessels was associated with poor clinical outcomes. [13][14][15][16][17][18][19][20] Immunohistochemical analysis has shown the VASH1 protein in the cytoplasm of ECs, and the intensity of VASH1 immunostaining is positively correlated with that of VEGF in cancer cells. 10,13 As the degradation and inactivation of VASH1 after secretion are enhanced in the tumor microenvironment, 24 immunohistochemical staining of VASH1 in ECs may not correlate with its antiangiogenic activity. Rather, it simply reflects the response of ECs to angiogenic stimulation by cancer cells. The source of plasma VASH1 is not determined at the moment. While it can be assumed to derive from tumor vasculature, it is also possible that VASH1 proteins in systemic circulation are derived from the normal vascular bed. Nevertheless, our present analysis is the first to propose the value of determining plasma VASH1 concentrations in lung cancer patients.
Blood sampling, or so-called liquid biopsy, is now widely being tested for its practicality in early diagnosis, treatment monitoring, and/or prognosis assessment of cancer patients, including NSCLC. 25 Such blood samples contain circulating tumor cells, circulating free DNA, exosomes, and tumor-educated platelets, among other components. 25 In particular, the value of circulating tumor cells in prognostic assessment of NSCLC is currently being investigated but is not yet applied in routine practice. 25 We would like to suggest the implementation of determining plasma VASH1 concentrations as routine practice in regular clinics.
It has been shown that tail vein injection of adenovirus encoding human VASH1 gene in experimental murine models increased plasma VASH1 concentration and inhibited tumor growth and metastasis. 10,11 The challenging question then is whether it would be possible to increase the plasma VASH1 concentration in medium and low VASH1 patients so as to improve their prognosis. Consequently, it is of critical importance to determine how to manipulate the plasma concentration of VASH1. We have recently demonstrated that persistent physical exercise raises plasma VASH1 concentrations in patients with peripheral vascular disease. 23 Preoperative physical exercise has been reported to reduce the mortality of patients with NSCLC. 26 In this context, it would be noteworthy to investigate whether preoperative physical exercise does indeed increase plasma VASH1 concentrations of patients with NSCLC, and if it is associated with an improvement in prognosis.
In summary, high preoperative plasma VASH1 concentration is associated with better prognosis in patients with lung cancer. We propose plasma VASH1 concentration as a suitable biomarker to assess the prognosis of patients with lung cancer after surgery.   Multivariate HR (95% CI) has been adjusted for age, sex, baseline performance status (0, 1, or more), and pathological stage (stages IA, IB, and IIA/IIB/IIIA).