|Year : 2018 | Volume
| Issue : 1 | Page : 28-35
Role of vascular adhesion protein-1 in patients with alopecia areata
Abd El-Aziz I El-Taweel1, Amany I Mustafa1, Ola S El-Shimi2, Fatma Y Abd El-Aziz Anwer3
1 Department of Venereology and Andrology, Faculty of Medicine, Benha University, Benha, Egypt
2 Department of Clinical Pathology, Faculty of Medicine, Benha University, Benha, Egypt
3 Department of Venereology and Andrology, Kafr Shokr Central Hospital, Benha, Egypt
|Date of Submission||17-Jan-2017|
|Date of Acceptance||16-Feb-2017|
|Date of Web Publication||28-Feb-2018|
Fatma Y Abd El-Aziz Anwer
Abd-Elrahman Gab Allah Street, New Benha, Benha City, Qalyubia Governorate, 13511
Source of Support: None, Conflict of Interest: None
Objective The aim of this study was to detect the role of vascular adhesion protein-1 (VAP-1) in patients with alopecia areata (AA).
Background AA affects about 2–3% of new patients attending dermatology clinics. The etiology of AA is still largely unknown. Clinically, AA presents as an area of well-circumscribed patch of sudden hair loss. The severity of disease is graded using Severity of Alopecia Tool score. Treatment depends on disease involvement. In 1996, Arvilommi and colleagues reported that about 75–100% of dermal vessels in skin lesions demonstrated positive VAP-1 expression.
Patients and methods This study included 50 patients with AA (group A), and 30 apparently healthy individuals of matched age and sex were chosen as a control group (group B). All patients were selected from the Outpatient Clinic of Dermatology and Andrology Department of Benha University Hospitals.
Results The mean±SD age of the patients was 27.3±14.05, whereas the mean±SD age of the control group was 32.8±11.27 years. Serum level of VAP-1 was higher in AA patients with progressive course compared with patients with constant course. Serum VAP-1 level was variable according to the pattern of hair loss in the patient group. The VAP-1 level according to the presence or absence of exclamation mark was not statistically significant. There was a highly significant variation in serum VAP-1 level according to psychological stress. VAP-1 level was not statistically significant according to the presence or absence of nail pitting. There was a highly statistically significant correlation between VAP-1 level and the Severity of Alopecia Tool score. There was a statistically significant correlation between VAP-1 level and the number of lesions. There was no statistically significant correlation between VAP-1 level and the size of the lesion. There was no statistically significant correlation between VAP-1 level and the age of the patients and the duration of the disease.
Conclusion Serum VAP-1 level is an important marker for the early diagnosis (prediction) of suspicious cases of AA, prognosis of the disease, the course of the disease, and whether the case will progress to alopecia totalis or universalis or it will remain stationary. It can be a target for new therapeutic lines for the disease.
Keywords: alopecia areata, severity of alopecia tool score, vascular adhesion protein-1
|How to cite this article:|
El-Taweel AAI, Mustafa AI, El-Shimi OS, Abd El-Aziz Anwer FY. Role of vascular adhesion protein-1 in patients with alopecia areata. Benha Med J 2018;35:28-35
|How to cite this URL:|
El-Taweel AAI, Mustafa AI, El-Shimi OS, Abd El-Aziz Anwer FY. Role of vascular adhesion protein-1 in patients with alopecia areata. Benha Med J [serial online] 2018 [cited 2019 May 26];35:28-35. Available from: http://www.bmfj.eg.net/text.asp?2018/35/1/28/226420
| Introduction|| |
Hair loss is one of the most common complaints among all patients consulting a dermatologist and is usually associated with severe psychological disturbances, distress, and even symptoms of depression .
Human hair can be classified as follows: (i) androgen-independent hair (i.e. eyebrows and lashes), and (ii) hair on hormone-dependent body regions (i.e. scalp, beard, chest, axilla, and pubic region) .
Hair follicles can be divided into lanugo, vellus, and terminal follicles, which differ in hair diameter, length, and pigmentation. The same follicle is able to produce lanugo hair during the fetal period, vellus hair during childhood, and terminal hair during adulthood .
The hair follicle is arranged in a stocking-like configuration, containing several layers of structures. The hair follicle contains an upper segment (infundibulum), a middle segment (isthmus), and a lower segment (bulb and suprabulb) .
Hair follicle, which is a stem-cell-rich mini-organ, undergoes life-long, cyclic transformations between stages of active regeneration (anagen), apoptotic involution (catagen), and relative proliferative quiescence (telogen) .
Alopecia areata (AA) is considered one of the most frequent diseases and a major unsolved clinical problem . AA is a common, chronically relapsing localized hair loss that occurs mainly on the scalp but can occur at any hair-bearing area. It is characterized by a sudden loss of hair . AA affects about 2–3% of new patients attending dermatology clinics, presenting a wide range of clinical heterogeneity .
In AA the follicles become smaller during the course, forming miniaturized hair, and are substituted by fibrous tracts . During the development of alopecias, exogen often occurs before anagen is renewed or when anagen is dystrophic. This results in a state called kenogen, in which no visible hair fiber is left in the hair follicle. Essentially, AA-affected skin can be said to be in a state of kenogen . In the chronic stages of AA, the most affected hair follicles are forced into prolonged telogen and no attempt at re-entering anagen or new growth of hair is observed. At this point, any inflammation present will typically be localized in the papillary dermis around miniaturized hair follicles .
The etiology of AA is still largely unknown, and currently available treatments are mostly unsatisfactory and inefficient for the more chronic and severe types of the disease; a useful biological marker implicating the disease’s activity has not yet been identified ([Figure 1]) .
The genetic basis of AA is strongly supported by its observed heritability in first-degree relatives, twin studies, and genetic linkage analysis of AA families . Patients with atopic and autoimmune diseases have an increased risk of developing a severe subtype of AA, suggesting a genetic susceptibility to develop autoimmune phenomena . An association between AA and human leukocyte antigen has been demonstrated .
Clinical and experimental studies showed that environmental insults such as emotional/physical stressors, hormones, and infections contribute to autoimmunity. Stress hormones, in their efforts to maintain dermatological homeostasis in reaction to their environment, are known to affect AA onset .
A viral etiology was proposed in the late 1970s, but subsequent studies have demonstrated no connection . Cytomegalovirus has been suggested as a potential promoter of AA, but several subsequent studies were not able to confirm the potential association. Genetic research suggests a possible role for a cytomegalovirus binding protein in natural killer cell activation . Hepatitis B virus, hepatitis C virus, and Epstein–Barr virus infections have also been implicated as triggers for AA .
Clinically, AA presents as an area of well-circumscribed patch of sudden hair loss. Patches up to 2 cm in diameter may appear overnight and then extend circumferentially at a rate of around 1 cm/week. The initial loss is by hair breakage, close to or just below the surface of the skin. Exclamation mark hairs, which are dystrophic hairs with fractured tips, are commonly seen around the margins of the hair loss patch. Although any hair-bearing area may be affected, 90% of lesions occur on the scalp. The skin over the affected areas may have no overt epidermal changes except for some mild erythema and atrophy .
Autoimmunity in AA is strongly supported by clinical observations that patients with AA are often diagnosed with one or more other autoimmune disorders, including vitiligo, lupus erythematosus, myasthenia gravis, scleroderma, ulcerative colitis, thyroiditis, celiac disease, and rheumatoid arthritis .
Diagnosis of AA had been made by a trained dermatologist based on the clinical evaluation of a diffuse or patchy alopecia and the presence of exclamation mark hairs .
Gauging severity of disease
The National Alopecia Areata Foundation working committee had devised ‘Severity of Alopecia Tool’ (SALT) score (Price and Gummer, 1989) . Scalp is divided into four areas as follows: the vertex, 40% (0.4) of the scalp surface area; the right profile of the scalp, 18% (0.18) of scalp surface area; the left profile of scalp, 18% (0.18) of scalp surface area; the posterior aspect of the scalp, 24% (0.24) of scalp surface area. Percentage of hair loss in any of these areas is percentage of hair loss multiplied by percent surface area of the scalp in that area. SALT score is the sum of percentage of hair loss in all above-mentioned areas .
Dermoscopic features that are seen in AA include yellow dots, black dots, broken hairs, short vellus hairs, and tapered hairs. These findings are not specific for AA but can be very helpful when the diagnosis is uncertain ([Figure 2]) .
|Figure 2 (a) Close-up view of exclamation mark hairs. (b) Dermoscopic examination of alopecia areata showing exclamation mark hairs and black dots .|
Click here to view
Treatment depends on disease involvement and some options are topical, intralesional, or systemic steroids, and topical immunotherapy or systemic immunomodulators . Other options are topical minoxidil, topical dithranol, and anthralin, contact immunotherapy, photochemotherapy, and oral cyclosporin. Intralesional triamcinolone is the most commonly used method of treatment .
In the late 1980s vascular adhesion protein-1 (VAP-1) was found to be an adhesion molecule that takes part in the inflammatory process of arthritis . VAP-1 was discovered by Salmi and Jalkanen (1992) during the production of monoclonal antibodies against human synovial vessels .
VAP-1 is a heavily sialylated homodimeric transmembrane glycoprotein with a molecular weight of 170–180 kDa, and the human AOC3 gene is located on chromosome 17 (17q21.31) .
VAP-1 has been suggested to play a role in immune cell trafficking and has been implicated in the pathogenesis of inflammatory diseases by influencing immune cell trafficking (through cell adhesion) and cell activation .
Its mechanism of action as an immune modulator and its structure–function relationship is not clearly understood. It is not clear whether VAP-1 function is dependent on the oxidase activity or on other structural features of the protein .
Its synthesis is induced/upregulated by chronic inflammation in the vessels of different tissues, such as tonsil, gut, skin, and synovium . VAP-1 is stored in intracellular vesicles, and during inflammation rapid translocation to the luminal surface of endothelial cells is observed. These cytoplasmic vesicles are responsible for recycling of VAP-1 .
A role for VAP-1 in mediating leukocyte migration and propagation of inflammation has been proposed to be facilitated by both its enzymatic activity and direct adhesion .
The importance of VAP-1 in chronic cutaneous conditions was first suggested by Arvilommi et al. (1996) when they noticed that about 75–100% of dermal vessels in skin lesions of psoriasis, lichen planus, and pemphigoid demonstrated positive VAP-1 expression .
VAP-1 is a target of interest for autoimmune, inflammatory, and fibrotic diseases. BIOTIE Therapeutics is conducting clinical trials with their anti-VAP-1 antibody . Several approaches to inhibit VAP-1 (e.g. small molecule inhibitors, small interfering RNAs, and function-blocking antibodies) have been proposed .
| Aim|| |
Our study aimed to evaluate the serum level of VAP-1 in patients with AA and to assess its clinical significance.
| Patients and methods|| |
This study included 50 patients suffering from AA (group A), and 30 apparently healthy individuals had been chosen as a control group (group B). All patients had been selected from the Outpatient Clinic of Dermatology and Andrology Department of Benha University Hospitals. Written informed consent had been obtained from all participants. The study had been approved by the ethics committee of Benha Faculty of Medicine.
Inclusion criteria were as follows:
- Clinically typical AA lesions with different clinical varieties such as patchy type, ophiasis, alopecia totalis, and alopecia universalis.
- Different degrees of severity of AA according to SALT score.
- Patients with AA on topical or systemic therapy.
- Pregnant and lactating women.
- Patients with other inflammatory disorders.
- Patients with peripheral vascular disorders.
- Patients with other diseases causing increased levels of VAP-1, such as atherosclerosis, autoimmune diseases, rheumatoid arthritis, systemic sclerosis, and psoriasis .
All studied participants were tested for serum level of VAP-1. Three milliliter of venous blood was collected from each participant aseptically by means of venipuncture using disposable plastic syringe and placed on plain tube (without anticoagulant) for serum separation. The tube was left at room temperature for 30 min until coagulation, and then was centrifuged at 1500 rpm for 15 min. The resultant serum was stored at −20°C for further testing. A double-antibody sandwich enzyme linked immune sorbent assay (ELISA) was used to detect serum level of VAP-1 using a commercial Human VAP-1 ELISA Kit (Thermo Fisher Scientific Inc., Carlsbad, CA, USA) for research use only.
The collected data were tabulated and analyzed using SPSS version 16 software (SPSS Inc., Chicago, Illinois, USA). Categorical data were presented as number and percentages, whereas quantitative data were expressed as mean±SD, median, range, and interquartile range. The χ2-test was used to analyze categorical variables. Continuous data were tested for normality using the Shapiro–Wilk test assuming normality at P more than 0.05 and using Student’s t-test for normally distributed variables, whereas the Mann–Whitney U-test, Kruskal–Wallis test, and Spearman’s correlation coefficient (ρ) were used for nonparametric variables. Receiver–operating characteristic curve was used to determine cutoff values VAP-1 with optimum sensitivity and specificity. The accepted level of significance in this work was stated at 0.05 (P≤0.05 was considered significant).
| Results|| |
The present study included 50 patients with AA (31 male and 19 female) and 30 healthy controls (12 male and 18 female) ([Table 1]).
The mean±SD age of the patients was 27.3±14.05 years, whereas the mean±SD age of the control group was 32.8±11.27 years. There was no statistically significant difference between the patient and control groups as regards age (P=0.07) ([Table 2]).
There was no statistically significant difference between the patient and the control group as regards sex (P=0.056) ([Table 3]).
The course of the disease was constant in 52% of patients, whereas it was progressive in 48% of patients. The mean±SD duration of the disease was 8.4±12.3 months. The pattern of loss of hair was patchy in 84% of patients, ophiasis in 8% of patients, totalis in 4% of patients, and universalis in 4% of patients. The mean±SD number of patches was 2.0±1.8. The mean±SD average size of patches in D1 was 2.4±1.3 and in D2 it was 3.1±1.9. There were exclamation marks in 13% of patients, whereas they were absent in 87% of patients. Psychological stress was not present in 30% of patients; it was mild in 6% of patients, moderate in 42% of patients, and severe in 22% of patients. The mean±SDSALT score was 22.8±25.1. Nail pitting was negative in 82% of patients and positive in 18% of patients ([Table 4]).
Family history was negative in 84% of patients, whereas it was positive in 16% of patients ([Table 5]).
The serum level of VAP-1 was higher among AA patients when compared with the control group and the difference was statistically significant (P=0.003) ([Table 6]).
|Table 6 Comparison of serum level of vascular adhesion protein-1 between the patient and control groups|
Click here to view
Using receiver–operating characteristic curve, the cutoff value of serum level of VAP-1 (the value more than which we could predict AA) was more than or equal to 7.26; the sensitivity was 68% and specificity was 60%. Its accuracy is 65%. There was statistical significance (P=0.003) ([Table 7]).
|Table 7 Vascular adhesion protein-1 in the early diagnosis (prediction) of alopecia|
Click here to view
Serum level of VAP-1 was higher among AA patients with progressive course when compared with patients with constant course, and the difference was statistically significant (P=0.03) ([Table 8]).
|Table 8 Levels of vascular adhesion protein-1 according to course of disease in the patient group|
Click here to view
VAP-1 is a significant test for prediction of the progressive course of AA (P=0.03). Its accuracy is 60%, sensitivity is 62.5%, and specificity is 57.7%. It has a cutoff value for prediction of the progressive course of AA of at least 11.61 ([Table 9]).
|Table 9 Performance of vascular adhesion protein-1 in the prediction of progressive course of alopecia|
Click here to view
Serum VAP-1 level was variable according to the pattern of hair loss in the patient group. In patchy type it ranged from 3.04 to 64.4 ng/ml and its median was 8.6 ng/ml. In ophiasis the median was 6.01 ng/ml. In totalis and universalis its median was 68.7 ng/ml. It was statistically significant (P=0.006) ([Table 10]).
|Table 10 Levels of vascular adhesion protein-1 according to the pattern of lesion in the patient group|
Click here to view
VAP-1 level is a highly significant test for prediction of totalis or universalis type of AA (P=0.001). It has a sensitivity of 100%, accuracy of 94%, and specificity of 93%. The cutoff value was more than or equal to 51.37 ([Table 11]).
|Table 11 Performance of vascular adhesion protein-1 in the detection of totalis/universalis types of alopecia|
Click here to view
The VAP-1 level according to the presence or absence of exclamation mark is not statistically significant (P=0.49) ([Table 12]).
|Table 12 Levels of vascular adhesion protein-1 according to the presence of exclamation mark hair in alopecia areata lesions|
Click here to view
There was a highly significant variation in serum VAP-1 level according to psychological stress in the patient group (P<0.001). If there was a severe psychological stress the median VAP-1 level was 55.25 ng/ml. If there was no psychological stress the median VAP-1 level was 5.21 ng/ml ([Table 13]).
|Table 13 Levels of vascular adhesion protein-1 according to psychological stress in the patient group|
Click here to view
VAP-1 level was not statistically significant according to the presence or absence of nail pitting in the patient group (P=0.29) ([Table 14]).
|Table 14 Comparison of levels of vascular adhesion protein-1 as regards nail pitting in the patient group|
Click here to view
Serum VAP-1 level median was 51.37 if the patient had a positive family history, but the median was 7.59 if the family history is negative. It is a statistically significant test as the P value is equal to 0.003 ([Table 15]).
|Table 15 Levels of vascular adhesion protein-1 according to family history in the patient group|
Click here to view
There was a highly statistically significant correlation between VAP-1 level and the SALT score (P<0.001). There was a statistically significant correlation between VAP-1 level and the number of lesions (P=0.034). There was no statistically significant correlation between VAP-1 level and the size of the lesion. There was no statistically significant correlation between VAP-1 level and the age of the patients and the duration of the disease (P=0.14 and 0.93, respectively) ([Table 16]).
|Table 16 Correlation between the vascular adhesion protein-1 level and the studied variables in the patient group|
Click here to view
| Discussion|| |
AA is a type of nonscarring hair loss affecting anagen stage hair follicles with a multifactorial autoimmune pathogenesis and an unknown etiology .
Hypotheses as regards the pathogenesis of AA include a lymphocyte-mediated inflammation that suggests an underlying autoimmune etiology, an association with human leukocyte antigen class II antigen alleles, and contribution from environmental factors such as hormonal fluctuation, infectious agents, vaccinations, and stress .
As many of the known endothelial adhesion molecules (ICAM-1, VCAM-1, E-selectin, and P-selectin) are inducible in inflammation, they were investigated for the possible association between VAP-1 synthesis and inflammatory status of tissues. In skin, chronic inflammation was accompanied with an increased synthesis of VAP-1 .
VAP-1 has been suggested to play a role in immune cell trafficking and has been implicated in the pathogenesis of inflammatory diseases .
Results of our study showed that there was a statistically significant difference between patients and controls as regards the level of serum VAP-1 (P=0.003). Moreover, there was a statistically significant serum level of VAP-1 in the early diagnosis (prediction) of AA as the serum level of VAP-1 had a cutoff value of at least 7.26. Our results showed that this test had a sensitivity of 68% and a specificity of 60%. The accuracy of this test was 65%. To the best of our knowledge it is the first study to assess the serum level of VAP-1 in AA patients and in the early diagnosis of AA.
Results of our study showed that there was a highly statistically significant importance of serum level of VAP-1 in the prediction of alopecia totalis or alopecia universalis (P=0.001) with a sensitivity of 100% and specificity of 93%. To the best of our knowledge it is the first study to reveal these results.
Results of this study showed that there was a highly statistically significant difference in the serum level of VAP-1 among AA patients according to the psychological stress in those patients (P<0.001). Moreover, there was a highly statistically significant difference in the serum level of VAP-1 according to SALT score (P<0.001). To the best of our knowledge it is the first study to assess the serum level in AA patients and compare it with these variants. SALT score was used to determine the severity of the disease as reported by Amin and Sachdeva  and Yenin et al. .
| Conclusion|| |
AA is a common disease affecting all age groups and both sexes. Mostly it is a clinically diagnosed disease, but we found that serum VAP-1 level is an important marker for the early diagnosis (prediction) of suspicious cases. Moreover, it is an important marker in the prediction of the prognosis of the disease as it can predict the course of the disease and whether the case will progress to alopecia totalis or universalis or it will remain stationary. It can be a target for new therapeutic lines for the disease.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
França K, Rodrigues TS, Ledon J, Savas J, Chacon A. Comprehensive overview and treatment update on hair loss. J Cosmetics Dermatolog Sci Appl 2013; 3:1–8.
Rosen J, Landriscina A, Friedman A. Nanotechnology-based cosmetics for hair care. Cosmetics 2015; 2:211–224.
Geyfman M, Plikus MV, Treffeisen E, Andersen B, Paus R. Resting no more: re-defining telogen, the maintenance stage of the hair growth cycle. Biol Rev Camb Philos Soc 2015; 90:1179–1196.
Liu C, Guan Y, Wang J, Zhu D. Reflectance spectroscopy for evaluating hair follicle cycle. Dyn Fluctuations Biomed Photonics 2014; 20:051011–0510116.
Blume-Peytavi U, Vogt A. Human hair follicle: reservoir function and selective targeting. Br J Dermatol 2011; 165:13–17.
Nissimov JN, Das Chaudhuri AB. Hair curvature: a natural dialectic and review. Biol Rev 2014; 89:723–766.
McElwee KJ, Gilhar A, Tobin DJ, Ramot Y, Sundberg JP, Nakamura M et al.
What causes alopecia areata? Exp Dermatol 2013; 22:609–626.
Devi M, Rashid A, Ghafoor R. Intralesional triamcinolone acetonide versus topical betamethasone valearate in the management of localized alopecia areata. J Coll Physicians Surg Pak 2015; 25:860–862.
Rajan MM, Ramani RP, Kartik VV, Kandasamy D. Alopecia areata and dental disease association − a review. Int J Pharm Bio Sci 2013; 4:978–982.
Alkhalifah A, Alsantali A, Wang E, McElwee KJ, Shapiro J. Alopecia areata update. J Am Acad Dermatol 2010; 62:177–188.
Wang E, Mcelwee KJ. Etiopathogenesis of alopecia areata: why do our patients get it? Dermatol Ther 2011; 24:337–347.
Amin SS, Sachdeva S. Alopecia areata: a review. J Saudi Soc Dermatol Dermatolog Surg 2013; 17:37–45.
Biran R, Zlotogorski A, Ramot Y. The genetics of alopecia areata: new approaches, new findings, new treatments. J Dermatol Sci 2015; 78:11–20.
Islam N, Leung PSC, Huntley AC, Eric Gershwin M. The autoimmune basis of alopecia areata: a comprehensive review. Autoimmun Rev 2015; 14:81–89.
Alkhalifah A. Alopecia areata update. Dermatol Clin 2013; 31:93–108.
Madej A, Reich A, Orda A, Szepietowski J. Vascular adhesion protein-1 (VAP-1) is overexpressed in psoriatic patients. J Eur Acad Dermatol Venereol 2007; 21:72–78.
Pannecoeck R, Serruys D, Benmeridja L, Delanghe JR, Van Geel N, Speeckaert R, Speeckaert MM. Vascular adhesion protein-1: role in human pathology and application as a biomarker. Crit Rev Clin Lab Sci 2015; 52:284–300.
Price VH, Gummer CL. Loose anagen syndrome. J Am Acad Dermatol 1989; 20:249–256.
Salmi M, Jalkanen S. A 90-kilodalton endothelial cell molecule mediating lymphocyte binding in humans. Science 1992; 257:1407–1409.
Noonan T, Lukas S, Peet G, Pelletier J, Panzenbeck M, Hanidu A et al.
The oxidase activity of vascular adhesion protein-1 (VAP-1) is essential for function. Am J Clin Exp Immunol 2013; 2:172–185.
Kurkijärvi R, Jalkanen S, Isoniemi H, Salmi M. Vascular adhesion protein-1 (VAP-1) mediates lymphocyte-endothelial interactions in chronic kidney rejection. Eur J Immunol 2001; 31:2876–2884.
Arvilommi AM, Salmi MK, Jalkanen KS. Lymphocyte binding to vascular endothelium in inflamed skin revisited: a central role for vascular adhesion protein-1 (VAP-1). Eur J Immunol 1996; 26:825–833.
Huang KP, Mullangi S, Guo Y, Qureshi AA. Autoimmune, atopic, and mental health comorbid conditions associated with alopecia areata in the United States. JAMA Dermatol 2013; 149:789.
Salmi M, Jalkanen S. Different forms of human vascular adhesion protein-1 (VAP-1) in blood vessels in vivo and in cultured endothelial cells: implications for lymphocyte-endothelial cell adhesion models. Eur J Immunol 1995; 25:2803–2812.
Yenin JZ, Serarslan G, Yönden Z, Ulutaş KT. Investigation of oxidative stress in patients with alopecia areata and its relationship with disease severity, duration, recurrence and pattern. Clin Exp Dermatol 2014; 40:617–621.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8], [Table 9], [Table 10], [Table 11], [Table 12], [Table 13], [Table 14], [Table 15], [Table 16]