Soluble B7-H5 Is a Novel Diagnostic, Severity, and Prognosis Marker in Acute Pancreatitis

Xu, Ruoxin; Gong, Ju; Chen, Wei; Jin, Yakang; Huang, Jian

doi:https://doi.org/10.1155/2021/1223850

BioMed Research International

On this page

Abstract Introduction Materials and Methods Results Discussion Conclusions Data Availability Conflicts of Interest Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2021 | Article ID 1223850 | https://doi.org/10.1155/2021/1223850

Soluble B7-H5 Is a Novel Diagnostic, Severity, and Prognosis Marker in Acute Pancreatitis

Ruoxin Xu,¹Ju Gong,²Wei Chen,¹Yakang Jin,³and Jian Huang¹

Academic Editor: Gian Paolo Caviglia

Received08 Jun 2021

Revised06 Aug 2021

Accepted13 Sept 2021

Published08 Oct 2021

Abstract

As an important ligand in T lymphocyte costimulatory pathways, B7-H5 is involved deeply in the immune response in various diseases. However, its clinical usefulness as an early indicator in acute pancreatitis (AP) remains unclear. In this study, the levels of sB7-H5 and cytokines in plasma samples of 75 AP patients, 20 abdominal pain patients without AP, and 20 healthy volunteers were determined. Then, the correlation of sB7-H5 and clinical features, cytokines, the Ranson score, APACHE II score, Marshall score, and BISAP score was analysed, and the value of sB7-H5 for diagnostic, severity, and prognosis of AP was evaluated. We found that the levels of sB7-H5 were specifically upregulated in AP patients. Receiver operating characteristic (ROC) analysis revealed that sB7-H5 can identify AP patients from healthy or abdominal pain patients with 78.9% or 86.4% sensitivity and 93.3% or 90.0% specificity. Further analysis showed that the levels of sB7-H5 were significantly correlated with WBC (), GLU (), LDH (), Ca²⁺ (), AST (), PLT (), IL-6 (), IL-10 (), and TNF-α (). And levels of sB7-H5 were gradually increased among patients with mildly acute pancreatitis (MAP), moderately severe acute pancreatitis (MSAP), and severe acute pancreatitis (SAP). It can distinguish the severity of AP with good sensitivity and specificity. Moreover, when dividing the patients into two groups according to the median level of sB7-H5, the local complication and length of stay of low levels of the sB7-H5 group were significantly less than those in high levels of the sB7-H5 group. And the levels of sB7-H5 in AP patients were significantly correlated with the Ranson score (), APACHE II score (), Marshall score (), and BISAP score (). The AUCs of assessing local complications of sB7-H5 at day 1 and day 3 were 0.704 () and 0.727 (). These results showed the potential value of sB7-H5 as a diagnostic, severity, and prognosis marker of AP.

1. Introduction

Acute pancreatitis (AP) is a severe inflammation of the pancreas caused by local pancreatic damage, which may lead to multiorgan failure or death. The incidence of acute pancreatitis ranges from 13 to 45 per 100,000 population years, and it is rising worldwide probably due to a combination of risk factors, such as obesity, alcohol abuse, and gallstone disease [1–3]. According to the Revised Atlanta Classification, acute pancreatitis could be classified as mild AP (MAP), moderately severe AP (MSAP), and severe AP (SAP). Although the majority of patients (80%–85%) suffer from MAP with a mortality rate less than 3%, the patients developing severe acute pancreatitis (SAP) have a mortality rate of 13–35% [4]. Numbers of researches have shown that the first 48 hours after symptom onset are critical for effective treatment and good outcome. Thus, a good accurate predictor of the clinical course of AP to identify patients with risk of developing complications or death is essential in AP therapy [5].

HHLA2, also named as B7-H5 or B7-H7, is a newly discovered member of B7 family. Several studies showed that HHLA2 was potentially involved in cancer progression through immune regulation. Upregulation of HHLA2 was associated with poor clinical features and outcomes in breast cancer, clear cell renal cell carcinoma, colorectal cancer, and osteosarcoma, whereas downregulation of HHLA2 might be beneficial for gastric cancer and epithelial ovarian cancer [6, 7]. It can regulate a variety of immune cells, including B cells, T cells, and endothelial cells by binding to its putative receptor(s) on these cells [8].

Although the pathogenesis of AP is still unclear, the role of T lymphocyte–costimulatory pathways has been implicated in the development of this condition [9, 10]. Different studies showed that the members of the B7 family, which function as ligands in costimulatory pathways, had significant clinical significance in AP. Both sB7-H2 and membrane B7-H2 on CD14⁺CD16⁺ cells were specifically increased in AP patients when compared with healthy control, and upregulation of sB7-H2 might be a useful marker in the clinical diagnosis of AP [11, 12]. The level of sB7-H4, another member of the B7 family, was significantly related to the severity and poor prognosis of AP. Besides, upregulation of B7-H3 was also observed in AP patients, and applying anti-B7-H3 monoclonal antibody could ameliorate l-arginine-induced acute pancreatitis in mice via attenuating the inflammatory response [13]. However, according to our knowledge, the role of B7-H5 in AP has not been studied so far.

In this study, we measured the levels of soluble B7-H5 (sB7-H5) in AP patients and analysed the correlation between the levels of sB7-H5 with clinical features and cytokines. We found the levels of sB7-H5 were continually increased in AP patients within 72 hours, and the increasing of sB7-H5 might be a useful diagnostic, severity, and prognosis biomarker for AP patients.

2. Materials and Methods

2.1. Patients

This retrospective study was approved by the Ethics Committee of First Affiliated Hospital of Soochow University (Ethics Inspection No: 2021 No. 132). 30 MAP patients, 25 MSAP patients, 20 SAP patients, 20 patients with other abdominal pain without AP, and 20 healthy control individuals (recruited from physical examination) were collected from 2019.1 to 2020.12. Oral and written consents were provided from all patients. The patients were diagnosed according to the revised Atlanta classification for AP in 2012. The venous blood samples were collected on admission and after 72 hours. Plasma was separated by centrifugation and stored at -80°C until further analysis.

2.2. sB7-H5 and Cytokine Measurement

sB7-H5 and cytokines were detected using ELISA following the Manufacturer’s manual. The Human Platelet receptor Gi24 (C10orf54) ELISA kit (CSB-EL002961HU, Cusabio technology, Wuhan, Hubei, China), Human IL-6 ELISA Kit (70-EK106/2-96, MultiSciences Biotech, Hangzhou, Zhejiang, China), Human IL-10 ELISA Kit (70-EK110/2-96, MultiSciences Biotech, Hangzhou, Zhejiang, China), and Human TNF-α ELISA Kit (70-EK182-96, MultiSciences Biotech, Hangzhou, Zhejiang, China) were used.

2.3. Statistical Analysis

Data are collected and analysed using Graphpad prism (version 8, Graphpad Software, San Diego, CA, USA) and shown as ; the difference between 2 groups was evaluated using the Student’s -test; differences among multiple groups were evaluated by ANOVA and followed by the least significant difference (LSD) -test. The relationships between sB7-H5 with cytokines were analysed via the Pearson correlation analysis. The receiver operating characteristic (ROC) curves with calculation of the sensitivity and specificity were calculated. A value < 0.05 was considered as statistical significance.

3. Results

3.1. Levels of sB7-H5 Were Upregulated among AP Patients

To determine the changes of sB7-H5 in AP patients, the healthy group (), abdominal pain without AP group (), and the AP group () were enrolled. The basic clinical parameters of the AP patients, abdominal pain group, and healthy group are shown in supplement Table 1. There were no significant differences in age and sex among the three groups, and the levels of cytokines such as IL-6, IL-10, and TNF-α were significantly upregulated in AP patients, which were in line with previous studies [14]. The levels of sB7-H5 levels in the AP group () were significantly higher than those in the healthy group (), whereas the sB7-H5 levels in the abdominal sepsis group () were significantly lower than those in the healthy group (Figure 1(a)). The results indicated that sB7-H5 levels were specifically increased in AP patients. Furthermore, we determined whether the levels of sB7-H5 were changed with time in AP patients. The results showed that the levels of sB7-H5 at 72 hours () after admission were still increased than those on admission (Figure 1(b)). These results indicated that the levels of sB7-H5 were upregulated and continuously increasing in AP patients.

(a)

(b)

3.2. Levels of sB7-H5 Function as a Diagnostic Marker

To assess the value of sB7-H5 in differentiating AP patients among the three groups, ROC curve analyses were used. The areas under the ROC curves (AUCs) were 0.823 and 0.976 in differentiating between the AP with the healthy group and AP with abdominal pain (Figures 2(a) and 2(b)). When both using 1.36 ng per mL as the cut-off value, the sensitivity and specificity of sB7-H5 distinguishing AP from the healthy group and AP from abdominal pain were 78.9% and 86.4%, 93.3%, and 90.0%, respectively. Considering the stable expression of sB7-H5 in AP patients, sB7-H5 is a potential stable AP diagnostic marker with good performance.

(a)

(b)

3.3. Levels of sB7-H5 Were Correlated with the Severity of AP

The stable upregulation of sB7-H5 in AP patients indicated that it specifically responded to the pathological processes of AP. Considering the clinical parameters and cytokines function predictors of severity in AP, we firstly analysed the correlations between sB7-H5 and the inflammatory response in AP patients. As shown in Table 1, there were significant correlations between the serum sB7-H5 expression level and the levels of WBC, GLU, LDH, Ca²⁺, AST, PLT, IL-6, IL-10, TNF-α, IFN-γ, and scoring systems. The significant correlation between sB7-H5 with clinical parameters and cytokines suggested it could function as a predictor for the severity of AP. To further confirm the value of sB7-H5 in differentiating between MAP, MSAP, and SAP groups, ROC curve analyses were used. The AUCs were 0.863 and 0.972 in differentiating between and SAP, MAP, and groups on admission, respectively. Using 1.90 ng/mL or 2.538 ng/mL as the cut-off values for distinguishing from SAP or MAP from , the sensitivity and specificity of sB7-H5 were 82.8% and 80%, 100% and 90.9%, respectively (Figures 3(a) and 3(b)). Moreover, the AUCs of sB7-H5 for differentiating those groups at 72 hours were 0.923 and 0.863, respectively, and the sensitivity and specificity were 82.6%, 100% and 91.7%, 70% (Figures 3(c) and 3(d)). These results indicated that sB7-H5 could identify the severity of AP.

(a)

(b)

(c)

(d)

3.4. Levels of sB7-H5 Were Correlated with the Prognosis of AP

To determine the relationship between the levels of sB7-H5 with the prognosis of AP, we first divided the AP patients into two groups according to the median level of sB7-H5 to compare the local complication and length of stay in the two groups. As shown in Table 2, the local complication and length of stay of low levels of the sB7-H5 group were significantly less than those in high levels of the sB7-H5 group. The results indicated the levels of sB7-H5 were related with the outcome of AP. Considering the scoring systems function as an early assessment of the prognosis of acute pancreatitis [5], we further determined the correlation between levels of sB7-H5 with the outcomes of AP patients. Levels of sB7-H5 were significantly correlated with Ranson score (, ), BiSAP score (, ), Marshall score (, ), APACHE II score (, ), and length of stay (, ) (Figures 4(a)–4(e)). Furthermore, the ROCs of sB7-H5, the scoring systems, and cytokines in the prediction of local complications were plotted. The AUCs of sB7-H5 Day1, sB7-H5 Day3, Ranson score, BiSAP score, Marshall score, APACHE II score, IL-6, IL-10, and TNF-α were 0.704 (), 0.727 (), 0.796 (), 0.888 (), 0.786 (), 0.751 (), 0.790 (), 0.687 (), and 0.813 () (Figure 5). The results indicated that the levels of sB7-H5 were correlated with the outcomes of AP.

(a)

(b)

(c)

(d)

(e)

4. Discussion

AP is a progressive systemic inflammatory response (SIRS), and the mortality of SAP could be as high as 30% if without timely and appropriate treatment. Early diagnosis of AP and identifying SAP patients is the key point for improved prognosis, but still difficult, thus, there is a need for exploring rapid, long-lasted, and useful biomarkers. Although the pathogenesis of AP is still not fully understood, activation of T cells not only played an important role in the early local as well as systemic immune responses in AP but also directly involved in organ damage, especially in SAP [15]. The severe form of AP is closely related to the extent and systemic immune response [15]. The B7 family is one of the most important families in the costimulatory signal pathways which function as the second signal for T cell activation [16]. Massive studies focused on the immune checkpoint roles for cancer treatment of the B7 family and showed the importance of this family in immune response regulation [7, 17]. Members of the B7 family might also participate in the development of acute pancreatitis [13].

A few studies on members of this family in AP disclosed that sB7-H2, sB7-H4, and PD-L1 (sB7-H1) could function as biomarkers for diagnosis and severity identification. Huang et al. (2015) showed sB7-H2 could distinguish from MAP with 77.8% sensitivity and 80.0% specificity, whereas it distinguished SAP from with 92.0% sensitivity and 86.0% specificity [11]. Cheng et al. (2020) revealed that sB7-H4 could differentiate between health and AP, MAP and MSAP, and MSAP and SAP groups. The AUC for those groups were 0.958, 0.811, and 0.826, respectively, and the sensitivity and specificity could reach 86.5% and 96%, 81.8% and 67.9%, and 71.8% and 83.3%, respectively [18]. According to Chen et al. (2017), sPD-L1 was an independent risk factor for infectious complications in AP, and the AUC of sPD-L1 for predicting infectious complications was 0.721, which were similar with lymphocyte count [19, 20]. In this study, we evaluated the role of sB7-H5 in AP, and this is the first study of sB7-H5 in AP according to our knowledge.

We first found the levels of sB7-H5 were able to distinguish AP patients from the healthy group or abdominal pain group. Interestingly, the sensitivity and specificity of sB7-H5 to distinguish AP from abdominal pain were better than it for AP from healthy. The contrary trends of sB7-H5 in abdominal pain and acute pancreatitis indicated the different roles of sB7-H5 in these diseases. Second, our results also revealed that levels of sB7-H5 could function as markers of inflammation and identify severe AP patients. Levels of sB7-H5 were positively with the indicators of inflammatory response, including WBC, GLU, LDH, AST, IL-6, IL-10, and TNF-α; and further ROC analysis showed the AUC of sB7-H5 to identify the SAP increased from 0.863 on admission to 0.923 at 72 hours. These results indicated sB7-H5 is a stable biomarker for AP severity, and its stability provides a relative long window for detection. Since the outcomes of AP patients are closely related with the severity, we further compared the value of sB7-H5 in prognosis prediction. We also found the AP patients with higher levels of sB7-H5 had significantly more local complication and length of stay. And the AUC of sB7-H5 to predict the local complication increased from 0.704 on admission to 0.727 at 72 hours. These results revealed the prediction value of sB7-H5 in diagnostic, severity, and prognosis of acute pancreatitis.

Considering the clinical features of AP, SAP patients should be identified as soon as possible. Thus, great efforts have been made in the identification of biomarkers (such as acute phase proteins, cytokines, activation peptides of pancreatic proteases, and leukocyte-derived enzymes) that might be used for early risk and severity stratification in AP patients [5, 21]. However, all of these biomarkers and scoring systems have limits. For example, acute phase proteins, such as C-reactive protein (CRP), serum amyloid A protein (SAA), lipopolysaccharide-binding protein (LBP), and pentraxin 3 (PTX 3), are nondisease specific, whereas the scoring systems are unsuitable in clinical practice due to their complex and time-consuming [22–24]. In this study, we found the levels of sB7-H5 could identify SAP with good sensitivity and specificity. Using the levels of sB7-H5 for at 72 hours, the sensitivity and specificity in differentiating between SAP from MAP and groups could reach 82.6% and 100%. The high sensitivity indicated that using sB7-H5 for severity prediction could minimize the false negative results, which is important for AP treatment [25].

Since there are few studies about sB7-H5 in AP, the mechanisms of sB7-H5 in AP are still unclear. According to the role of sB7-H5 in various cancers, it may have both coinhibitory functions in inhibiting the proliferation of human CD4 and CD8 T cells and the production of T cell cytokines or costimulatory functions in immune signalling [26]. In our study, the positive relationship between sB7-H5 and cytokines suggested a costimulatory function of sB7-H5 contributing to the progress of inflammation and severity in AP.

5. Conclusions

This study reveals that upregulation of sB7-H5 in patients with AP could function as a biomarker for diagnosis, severity, and prognosis. However, this study is limited by the relatively small enrolled number of patients and the lack of patients with organ failure or in-hospital mortality. Further studies with multicentre, large samples may strengthen the value of sB7-H5 in AP diagnostics, severity, and prognosis. Besides, the stable upregulation of sB7-H5 indicates an important role of sB7-H5 in AP, and further studies on the role and mechanism of sB7-H5 may deepen our study on the pathogenesis of AP. Disclose the immune function of sB7-H5 might provide a potential therapy target of AP.

Data Availability

All experiment methods and the result generated or used during the study appear in the submitted article.

Conflicts of Interest

The authors declare that there is no conflict of interest regarding the publication of this paper.

Acknowledgments

This research is supported by Science and Technology of Tongren City (no. 2018-1-5) and Suzhou University (no. P112200519 and no. P112200719).

Supplementary Materials

Supplement Table 1: basic clinical parameters of the AP patients, abdominal pain without AP group and healthy group. (Supplementary Materials)

References

G. Ouyang, G. Pan, Q. Liu et al., “The global, regional, and national burden of pancreatitis in 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017,” BMC Medicine, vol. 18, 2020.
View at: Publisher Site | Google Scholar
P. J. Lee and G. I. Papachristou, “New insights into acute pancreatitis,” Nature Reviews Gastroenterology & Hepatology, vol. 16, no. 8, pp. 479–496, 2019.
View at: Publisher Site | Google Scholar
R. Pezzilli, F. Caputo, G. Testino et al., “Alcohol-related chronic exocrine pancreatic insufficiency: diagnosis and therapeutic management. A proposal for treatment by the Italian Association for the Study of the Pancreas (AISP) and the Italian Society of Alcohology (SIA),” Minerva Medica, vol. 110, no. 5, pp. 425–438, 2019.
View at: Publisher Site | Google Scholar
C. Sun, Z. Li, Z. Shi, and G. Li, “Current diagnosis and treatment of acute pancreatitis in China: a real-world, multicenter study,” BMC Gastroenterology, vol. 21, no. 1, 2021.
View at: Publisher Site | Google Scholar
P. Silva-Vaz, A. M. Abrantes, M. Castelo-Branco, A. Gouveia, M. F. Botelho, and J. G. Tralhão, “Multifactorial scores and biomarkers of prognosis of acute pancreatitis: applications to research and practice,” International Journal of Molecular Sciences, vol. 21, no. 1, 2020.
View at: Publisher Site | Google Scholar
G. Xu, Y. Shi, X. Ling et al., “HHLA2 predicts better survival and exhibits inhibited proliferation in epithelial ovarian cancer,” Cancer Cell International, vol. 21, no. 1, 2021.
View at: Publisher Site | Google Scholar
L. Ni and C. Dong, “New B7 family checkpoints in human cancers,” Molecular Cancer Therapeutics, vol. 16, pp. 1203–1211, 2017.
View at: Publisher Site | Google Scholar
J. L. Lines, E. Pantazi, J. Mak et al., “VISTA is an immune checkpoint molecule for human T cells,” Cancer Research, vol. 74, pp. 1924–1932, 2014.
View at: Publisher Site | Google Scholar
Q. Zhou, X. Tao, S. Xia et al., “T lymphocytes: a promising immunotherapeutic target for pancreatitis and pancreatic cancer?” Frontiers in Oncology, vol. 10, p. 382, 2020.
View at: Publisher Site | Google Scholar
A. Demols, O. Le Moine, F. Desalle, E. Quertinmont, J.-L. van Laethem, and J. Devière, “CD4⁺ T cells play an important role in acute experimental pancreatitis in mice,” Gastroenterology, vol. 118, no. 3, pp. 582–590, 2000.
View at: Publisher Site | Google Scholar
J. Huang, Z. Wu, S. Lu, J. Shen, X. Kong, and Y. Shen, “Soluble B7-H2 as a novel marker in early evaluation of the severity of acute pancreatitis,” Laboratory Medicine, vol. 46, no. 2, pp. 109–117, 2015.
View at: Publisher Site | Google Scholar
J. Gong, J. Huang, R.-X. Xu et al., “Clinical significance of B7-H2 expression in peripheral blood CD14+CD16+ monocytes in patients with acute pancreatitis,” world chinese journal of digestology, vol. 28, no. 4, pp. 149–154, 2020.
View at: Publisher Site | Google Scholar
X. Zhuang, J. Shen, Z. Jia et al., “Anti-B7-H3 monoclonal antibody ameliorates the damage of acute experimental pancreatitis by attenuating the inflammatory response,” International Immunopharmacology, vol. 35, pp. 1–6, 2016.
View at: Publisher Site | Google Scholar
J. Granger and D. Remick, “Acute pancreatitis: models, markers, and mediators,” Shock, vol. 24, Supplement 1, pp. 45–51, 2005.
View at: Publisher Site | Google Scholar
J. Glaubitz, A. Wilden, C. van den Brandt et al., “Experimental pancreatitis is characterized by rapid T cell activation, Th2 differentiation that parallels disease severity, and improvement after CD4⁺ T cell depletion,” Pancreatology, vol. 20, no. 8, pp. 1637–1647, 2020.
View at: Publisher Site | Google Scholar
R. J. Greenwald, G. J. Freeman, and A. H. Sharpe, “The B7 family revisited,” annual review of immunology, vol. 23, pp. 515–548, 2005.
View at: Google Scholar
A. Ghahremanloo, A. Soltani, S. M. S. Modaresi, and S. I. Hashemy, “Recent advances in the clinical development of immune checkpoint blockade therapy,” Cellular Oncology (Dordrecht), vol. 42, no. 5, pp. 609–626, 2019.
View at: Publisher Site | Google Scholar
Z. Cheng, Z. Liu, Y. Yuan, Z. Liu, Y. He, and P. Jiang, “Predictive value of serum soluble B7-H4 in acute pancreatitis,” European Journal of Clinical Investigation, vol. 50, no. 12, article e13346, 2020.
View at: Publisher Site | Google Scholar
Y. Chen, M. Li, J. Liu et al., “sPD-L1 expression is associated with immunosuppression and infectious complications in patients with acute pancreatitis,” Scandinavian Journal of Immunology, vol. 86, pp. 100–106, 2017.
View at: Publisher Site | Google Scholar
T. Pan, T. Zhou, L. Li et al., “Monocyte programmed death ligand-1 expression is an early marker for predicting infectious complications in acute pancreatitis,” Critical care, vol. 21, no. 1, 2017.
View at: Publisher Site | Google Scholar
C. Bedel, M. Korkut, and F. Selvi, “New markers in predicting the severity of acute pancreatitis in the emergency department: immature granulocyte count and percentage,” Journal of Postgraduate Medicine, vol. 67, 2021.
View at: Publisher Site | Google Scholar
J. Wu, Y. Wang, H. Li, W. Tan, X. Chen, and S. Ye, “Serum apolipoprotein B-to-apolipoprotein A1 ratio is independently associated with disease severity in patients with acute pancreatitis,” Scientific Reports, vol. 9, no. 1, 2019.
View at: Publisher Site | Google Scholar
S. M. Staubli, D. Oertli, and C. A. Nebiker, “Laboratory markers predicting severity of acute pancreatitis,” Critical Reviews in Clinical Laboratory Sciences, vol. 52, pp. 273–283, 2015.
View at: Publisher Site | Google Scholar
F. Başak, M. Hasbahçeci, A. Şişik et al., “Can C-reactive protein levels increase the accuracy of the Ranson score in predicting the severity and prognosis of acute pancreatitis? A prospective cohort study,” The Turkish journal of gastroenterology, vol. 28, pp. 207–213, 2017.
View at: Publisher Site | Google Scholar
A. Karpavicius, Z. Dambrauskas, A. Gradauskas et al., “The clinical value of adipokines in predicting the severity and outcome of acute pancreatitis,” BMC Gastroenterology, vol. 16, 2016.
View at: Publisher Site | Google Scholar
C. Zhong, Q. Lang, J. Yu, S. Wu, F. Xu, and Y. Tian, “Phenotypical and potential functional characteristics of different immune cells expressing CD28H/B7-H5 and their relationship with cancer prognosis,” Clinical & Experimental Immunology, vol. 200, pp. 12–21, 2020.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2021 Ruoxin Xu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

327

Downloads

921

Citations